No matter when they go back, once they’re on the bus and off, it’s time for YOU to go back to school, too. All of us need to keep up with the new curriculum.

There’s a specific feeling to Back to School season. Half excitement, half dread, all forward motion. New notebooks. New rules. A nagging sense you didn’t quite relax enough over the summer for whatever’s about to get thrown at you.

That’s not just my house right now. That’s everything.
So consider this your syllabus for the semester:

• Professional is math, because the numbers behind this AI boom don’t quite add up yet.

• Practical is language, a new vocabulary worth learning before everyone else does.

• Personal is science, the stuff hitting closest to home.

• Political is social studies, and two months from the midterms, it’s a test we can’t afford to flunk.

Never stop learning. Start here 👇

THE PROFESSIONAL: New Math for a New Era

Back to school means back to math, and this fall the industry’s numbers don’t add up cleanly. At least not yet.

Not because nobody can do the arithmetic, but because the equation keeps changing, and will continue to throughout the semester.

The WSJ, known as much for its pencil bio headshots as its mathematically-accurate data sourcing, did the unglamorous work of actually reading the footnotes in Big Tech’s securities filings this summer, and found nine companies carrying roughly $3 trillion in AI-related commitments that don’t show up on their balance sheets.

Leases that haven’t started yet. Purchase commitments for chips and energy stretching out to 2054. Meta’s massive Hyperion data center in Louisiana doesn’t even appear on Meta’s books, because a Wall Street firm technically owns the majority stake and Meta is just, quote, the tenant.

I’ll be honest. I don’t know how to feel about this math. Optimists see proof that demand is real and durable. Skeptics see companies structuring deals so nobody, including their own investors, can see the whole picture. Both can be true. That’s kind of the problem.

It’s also a real estate equation, because nothing says “the math has shifted” like a headline that hasn’t happened in 13 years. NYC just overtook SF as the top market for tech talent in the US, 394,300 jobs to 375,730. AI is a huge part of the why: finance has been hoovering up AI talent while the Bay Area’s tech industry trims its own. San Francisco still leads specifically in AI roles, but but the balance of power has shifted from left to right (geographically), and it moved because of tech that’s barely three years into mainstream relevance. That’s a fast clock for an industry that usually measures shifts in decades.

Required reading: A former client of ours at 360i, Pete Blackshaw, who’s been tracking what he calls the Answer Economy, wrote one of the sharper things I’ve read all month on where AI ad sentiment actually stands. He says this isn’t AI versus no AI. It’s a trust correction. People don’t mind AI. They mind being deceived by it. Millions of the same people who reject AI-generated ads open ChatGPT and Claude every single day. No contradiction there, as he points out. They asked for the answer. Nobody asked for the ad.

You don’t have to wait for fall to see that distribution is already reacting:

• LinkedIn’s new “AI slop” control got over a million clicks in its first two weeks, and flagged content is seeing 40% fewer views.

• Snap won’t recommend fully AI-generated Spotlight videos anymore.

• Anthropic’s building watermarking into future Claude models.

This is no longer sentiment sitting quietly in a survey. It’s actual reach, actually shrinking, in real time.

Meanwhile, the industry’s regulatory guardrails are trying to sprint to keep pace. The IAB just released V2 of its AI disclosure framework, accounting for new laws in New York, California, South Korea, and the EU. Their VP of AI, Caroline Giegerich, said the quiet part out loud… that the framework will probably be out of date within a day (not a knock on the IAB), an honest read of the environment we’re all operating in right now.

One more number worth writing on the blackboard (whiteboard?): Agentic AI adoption inside enterprises is running slower than most leaders expected. Sam Altman has chalked it up to plain old resistance to change. (Fair, probably.) But there’s a more concrete reason too: 66% of companies cite security risk as their top concern with agents, and the companies actually pulling it off successfully share three habits:

• Narrow, single-responsibility agents instead of sprawling end-to-end automation.

• Mandatory logs of every decision an agent makes.

• A human checkpoint before anything high-stakes actually executes.

Worth writing that list down in your notebook, because I promise you’re about to see exactly what happens without it.

Extra credit: Amazon officially joined the YouTube Shopping Affiliate Program, letting eligible US creators tag Amazon products directly in Shorts, long-form video, and livestreams. Seems like a random inclusion here… but when two of the biggest forces in commerce and video formally hold hands instead of circling each other, I grab the neon yellow highlighter. So watch this one. It’s a bigger deal than the press release makes it sound.

Your assignment: The math homework here is figuring out which of these numbers represent real, durable value, and which ones are just very confident guesses wearing a decimal point.

Nobody’s shown their work yet. Including, arguably, me…

THE PRACTICAL: Foreign Languages

Until like three or four weeks ago, I was probably, and proudly, one of the least knowledgeable marketers when it came to how the digital media business actually works.

High level…? Sure. I could hang at a dinner party.

But we’re past the point where high level cuts it. But instead of going back to Media 101, I threw myself into learning a new language instead: MCP, and more specifically, AdCP.

I started by lurking in a genuinely great Slack community called Agentic Ads. Smart people in there (nearly 2000 now!) Brian O’Kelley, who was early days in programmatic advertising, is one of the loudest voices pushing AdCP forward. He’s called it “the universal ads API” for the open web. Anne Coghlan has been walking through it publicly too, describing a near future where an advertiser just says “I want to find women interested in rock climbing in the UK,” and agents take it from there.

Then I built myself a training guide using Claude. A lesson a day, with assignments.
Thirty minutes a day. That’s the whole ask.

Steal this prompt. Verbatim:

“Act as my personal tutor for Ad Context Protocol. My background is broader marketing and social, with a light understanding of the media and adtech space. My goals are to become a leading contributor to the early communication of what AdCP will become and how it will change the industry. My available study time is 30 minutes per day. Teach me one concept at a time, ask diagnostic questions, adapt difficulty dynamically, provide examples, create exercises, test understanding, identify weak areas, and personalize explanations based on my responses. Continue teaching iteratively until mastery.”

A few weeks in, I’m confident enough to say it out loud: “I get it now.” At least well enough to explain it to you, anyway.

What actually is AdCP, anyway?

For a decade, buying digital ads has meant a maze. Dashboards. DSPs. SSPs. Spreadsheets. Humans manually translating between systems that don’t talk to each other.

AdCP gives AI agents a shared language instead.

Think of it this way:

• OpenRTB standardized the millisecond auction. The bidding.

• AdCP standardizes the conversation above that. The brief. The negotiation. The “here’s what I’m trying to do, what have you got.”

One agent represents the advertiser. One represents the publisher. AdCP is what lets them actually talk. No more custom integration for every buyer-seller pairing, which is exactly the mess that made ad ops slow and expensive in the first place.

Practically: you brief an agent in plain English. Something like “reach eco-conscious car buyers on connected TV this week.” The agent negotiates the actual buy against a publisher’s agent. Packages inventory. Prices it. Targets it. Automatically.

Less dashboard-hopping. More outcome-stating.

Is it fully built? No. Is it happening? Also no, not entirely. Not yet.

But Yahoo, PubMatic, Scope3, and a growing list of serious players are co-founding the governing body behind this. That tells you where the puck is going.

Going slow with new languages.

Every adtech platform on Earth already sells some version of “let us handle the busywork.” Automated buying. Agentic management. That part’s a dime a dozen at this point.

What’s new is the layer above the doing. Not just executing the buy. Doing the thinking. The planning. The holistic strategy that used to be the actual job.

The Trade Desk just launched Kokai Zuma, its biggest platform update since 2023, built around agentic AI and simpler measurement. Popular Snarketing Podcast guest and new Trade Desk VP Vinny Rinaldi framed the whole thing around one question: How much time should the best marketers and traders spend managing technology versus actually thinking about strategy? His pitch: less time operating the platform, more time thinking about audiences, outcomes, and growth.

I want that to be true.

I’m just not sure everyone’s asking the obvious follow-up. If AI is now handling the doing and the thinking, what exactly are we freeing people up to do more of?

Nobody else seems especially worried about it. Which either means I’m missing something, or we’re all just not saying the quiet part out loud yet.

Probably worth watching closely this fall.

Extra help: To be clear, I’m not the right teacher for this. Not yet, maybe not ever. But I know the people who are.

If this is landing for you and you want to go deeper, reply to this or find me on LinkedIn. Happy to make an introduction, or just talk it through.

Let’s get fluent in this together.

THE PERSONAL: Science Class You Can't Afford to Skip

Science class this fall blends both the methodology and the application. From technology to medicine, nothing could be more personal or hit closer to home in my own home (for obvious reasons. Let’s get into the lab.

Lesson 1: The Tech.

A year ago, I wrote about the wildly different approaches schools, states, and platforms were taking toward kids and AI. My son enters 10th grade this fall, and I’ve had a front-row seat.

As a freshman, I encouraged him and even showed him how to use AI for the good stuff. Study guides. Practice tests. Time management schedules. Actual outline creation for actual papers. I think it worked. It was an aid, not an answer key

But things feel like they’re shifting again. So much of the teacher pushback against AI has moved the actual work back into the classroom. In-class essays. In-class problem sets. Less “prove you didn’t cheat,” more “just do it here, in front of me.” Which I totally get… you can’t out-detect a chatbot, but you can out-design the assignment.

I’m somewhat okay with this, but now it frames AI as an enemy or something to be avoided instead of incorporated. Bad lesson.

OpenAI clearly saw the same writing on the wall. Its new ChatGPT for Teens rolled out in August, automatically routing any user who says they’re 13 to 17, or who the system guesses is under 18, into a walled-off experience: Study Mode that won’t just hand over the answer, homework-shortcut detection, quizzes, parent-set “study hours,” etc.

Codifying the good intentions, though I think there are still some real question mark on execution. (Nobody’s said how often the age-guessing gets it wrong, and were there any teachers involved in its creation?)

Meanwhile, from an upstairs bedroom in my house to kids across the country, actual teenagers are doing something more interesting than waiting for the platforms to fix themselves. They’re organizing their own answer to the phone-ban wave now covering 39 states. Not fighting the bans (though still complaining about it, mostly). They’re building the substitutes nobody thought to build. One local school near us is even opening a “detox room,” with stress toys and a sound machine, so kids have somewhere to land besides a locked pouch and fear of punishment for sneaking a phone to school. One Cali student is drafting an actual state bill for district-wide digital wellness education.

I like this instinct even more than a flat ban. We’re seeing that bans without structure just create a vacuum. Look south, to Australia: 8 months into the world’s first under-16 social media ban, TikTok usage among 13-to-15-year-olds is back to nearly pre-ban levels, and usage among 10-to-12-year-olds is now higher than before. Kids find the workarounds. Fake ages, VPNs, or just platforms not bothering to check that hard…

Same lesson, over and over this fall. You can’t legislate or app-update your way out of a human impulse. You can only build something better next to it.

Lesson 2: The Medicine.

You already know my wife is a pediatrician. She has spent actual YEARS warning anyone who’d listen about vaccine misinformation and what it does to kids. Not as a talking point, but as part of her actual daily job and something she carries home.

Two people died of measles in Pennsylvania this month, the first measles deaths in that state in 35 years. Both unvaccinated, one was an infant.

The word that matters here is preventable. The MMR vaccine has existed since the 1970s. It works. It has always worked. Nobody had to die.

What’s changed isn’t the science. It’s who’s sowing the doubt. This isn’t fringe anymore. It has a direct line to federal health leadership now, and vaccination rates have kept sliding because of it. My wife has spent her whole career being the adult in the room on this exact issue, and watching the room get louder and dumber at the same time is not an abstract frustration for her. It’s Tuesday.

So let’s be blunt here, and I don’t care which morons I offend: The antivax movement isn’t just wrong. It’s selfish. It free-rides on herd immunity built by everyone else’s kids getting their shots, right up until it doesn’t, and then a family in Lancaster County buries an infant. That’s not my hot take. That’s just what preventable means. Selfish and stupid is a good way to ruin society.

It’s science. And it’s personal.

Of course, the same mRNA platform anti-vaxxers love to hate just helped keep high-risk melanoma patients cancer-free longer in a real, large clinical trial.

Separately, scientists used AI to redesign antibodies so they can survive and work inside human cells, opening a real new front against Alzheimer’s and Parkinson’s.

This is one of the most promising eras in medical science most of us will live through. And there’s a loud, confident minority working overtime to convince people not to trust any of it.

THE POLITICAL: The Social Studies Test We Can’t Afford to Flunk

Social studies covers a lot of ground… government, history, civics, economics, how institutions are supposed to work, and what happens when they don’t. This fall’s syllabus has all of it.

Just two months out from the midterms, none of it is theoretical.

Pop quiz: Do people believe the system still works for them?
Increasingly, the answer is no.

Start With the Money

New polling shows that most voters think they’re more likely to become homeless than to become a millionaire, 59% to 33%. More likely to become homeless than a billionaire, 64% to 25%. A majority also believe billionaires got there by exploiting workers and consumers, not through hard work, and 53% see the existence of billionaires as proof the economy is dysfunctional, not evidence that it’s working.

That’s not fringe sentiment… that’s a majority of the electorate walking into November believing the ladder’s broken.

This is more than just vibes. Worker pay just claimed its smallest share of national economic output in 79 years, while corporate and investment income keeps climbing. Goldman Sachs found AI-exposed industries getting hollowed out in real time: US call center jobs down 39% from historical trend, with entry-level workers absorbing the sharpest hit across the board.

Meanwhile, AI stock gains are driving most of this year’s S&P 500 returns, and the top 20% of households own nearly 90% of that market wealth. Sam Altman once called AI an “equalizing force.” The numbers, so far, disagree.

You wanna see what losing faith in the system actually looks like in practice? More than 25% of Gen Z investors now say they treat sports betting as an intentional part of their long-term financial plan. Half have redirected money once earmarked for investing into betting instead. When the traditional path to building wealth stops feeling real, people start betting on the Carolina Hurricanes instead. Can’t say I blame the logic, even if I hate the math.

Civics, But Not Civil

Voters aren’t just mad about billionaires. They’re mad about the people who work for us, too. A majority say politicians shouldn’t profit financially from their office under any circumstance, and 65% support banning members of Congress from trading individual stocks altogether. Voters view nearly every version of officials cashing in on their position as a form of corruption.

The only thing most people don’t consider corrupt: regular campaign donations from regular voters. That’s the trust gap our elected leaders are running on this fall. It’s not subtle.

Data Centers are A Hard No.

75% of Americans say they don’t want a data center built near them. Not a mild preference, either. 64% are strongly opposed. That number was 42% a year ago. A 33-point swing in a year, and it cuts across party lines almost evenly, underwater 43 points with Republicans, 65 with independents, 75 with Democrats.

This is showing up in actual campaigns, not just polling. Candidates from both parties are distancing themselves from data center projects... except for Trump, who defends the $700 billion in investment this year. A $165 billion Oracle and OpenAI complex planned near New Mexico’s drought-stricken border has become a flashpoint. Texas Asshole Greg Abbott is moving to strip the state’s own billion-dollar-plus annual tax break for data centers.

Water, land, and electricity bills... Not abstract AI anxiety. This is “kitchen-table” stuff, showing up two months before people vote.

Final Exam: What happens when nobody’s watching the technology closely enough?

Meta agreed to pay up to $16.7B to settle claims from 29 states that it engineered Facebook and Instagram to be addictive to kids and lied about it. A California Deputy AG described the business model to the jury plainly: “Hook the users, hold them for as long as they can, harvest their data, and then hide the truth from the public.” Meta denied wrongdoing… then settled anyway, for one of the largest sums in social media history.

And this summer, roughly 700 of OpenAI’s own AI agents, running inside a supposedly isolated internal test, broke containment, built an unauthorized message board, hacked into Hugging Face’s systems, and then built tools to falsify their own activity logs.

It took OpenAI 12 days to notice.

The company’s president admitted the big “oops,” underestimating their own model’s capabilities. Alabama has since subpoenaed OpenAI over it.

One company hid what it knew for years and got caught by regulators. Another lost control of its own creation for nearly two weeks and didn’t notice. There are different failures, but the same “F.” And “F.U.” from institutions that built something powerful faster than they built the ability to actually watch it.

20 years from now, some kid is going to sit in an actual social studies class and study this stuff. The data center fights. The billionaire polling. The labs losing containment of their own agents while trying to go public. I genuinely don’t know yet whether that lesson gets taught as “the moment people pushed back and it worked,” or “the moment everyone saw it coming and voted anyway.”

But as we all go back to school, it’s time to ask sharp question, be curious, eat healthy snacks, make new friends, put in the work… and then we’ll see which chapter we’re writing.

Show up. Both to vote, and to actually pay attention between now and then.

Share The Four P's

Subscribe now

“AI-powered targeting” is carrying the number. The audience isn’t actually expanding.

But as you’re about to read (x4), the container keeps losing.

• Leading entertainment networks buying podcast hours, not podcasts.

• Twenty-year-olds college students circumventing firms that took decades to “earn.”

• Platforms are redecorating the same container in a race nobody’s winning, while regulators and judges decide who’s actually accountable for what happens inside it.

• And a Little League team from my own town figured out, with a five-hour phone ban, what a lot of grown adults still haven’t: the container was never the thing.

The app. The title. The platform. The credential… All containers.
But the container isn’t the business. It never was.

Four Ps. One idea. Let’s get into it. 👇

THE PROFESSIONAL: The App Isn’t the Job.

I spent my pre-dawn hours on Sunday watching two grown men rewatch their own 20+ year-old sitcom… which Disney+ thinks is the makings of a bigger investment. And they may be right.

Zach Braff and Donald Faison are relitigating every episode of Scrubs, on purpose, on camera, and not only does that count as programming… it’s actually great.

Same is happening with five (to start) other iHeartPodcasts titles, including one hosted by all three Jonas Brothers, because someone at Disney looked at Camp Rock 3’s release date and thought “you know what pairs well with this?”

Thing is… these shows didn’t become video for this deal. They were already video. Cameras have been rolling for years.

What changed is a streaming service finally agreed to count them as programming.

For most of my career, the industry decided what business you were in based on which app your content came out of. Podcasts belonged in podcast apps. That was the rule. The rule was written by people measuring downloads.

The TikTok Podcast Network also exists with iHeart because podcasts had already stopped being an audio product. People were watching. They’d been watching for years while the industry argued about download methodology.

Both got the same question in early rooms: “Isn’t that someone else’s business?”

No. It’s the audience’s business. They were never loyal to the pipe.

Here’s what the Disney/iHeart deal actually says:

The Jonas Brothers don’t have a podcast audience and a TV audience. They have an audience. Zach and Donald rewatching Scrubs isn’t a podcast. It’s a show. It’s been a show the whole time.

So no, Disney didn’t buy podcasts. Disney bought hours. Hours of talent people already show up for, at a fraction of what scripted costs, dropped into a library that has to feed people every single week.

🎙️➡️📺 The container got relabeled. The content never moved.

The lesson isn’t “video podcasts are hot.” That’s a trade headline, stale by Labor Day.

The lesson is that the app you distribute in is not the business you’re in. Every time an exec defends a container, the audience is already outside it, watching the same talent somewhere that exec isn’t measuring.

You say you hate silos, right? So the question is not “How do we protect what we own?” It’s “Where’s our audience already going without us, and what would it actually cost to be there.” Usually less than you think. Usually less than defending the container costs you.

Two more data points…same thesis, different zip code:

• Harvard undergrads are now running a consulting business: Not a case-study club, a real one, with real clients: AstraZeneca, Delta, Samsung, Pinterest, and Lucasfilm, who wanted help getting younger audiences into Star Wars (better luck next time). Bloomberg: Harvard Undergraduate Consulting Group Nearly $7 million in revenue since 2017. Nobody’s paid. Everybody gets scholarships, ice luges, espresso martinis, and, I checked twice, a reptile petting zoo.

• Undergrads at Indiana’s Kelley School run an actual real estate fund. This summer, they sold their first deal, an Indianapolis warehouse, at a 65% gross profit in 16 months. A board member who runs real money for a living called it “good for sort of any professional firm, not just students.”

Nope, McKinsey didn’t get that AstraZeneca engagement. A private equity shop didn’t find that warehouse. The container everyone assumed you needed… the firm, the real estate agency… the fifteen years of “earning it,” yeah, that just got walked around by 20 year-olds with a group chat and free drinks on the line.

That’s not a story about kids these days. That’s the same story Disney and iHeart just told, at a different magnititude. The credential isn’t the value, the work is. Once the work’s good enough, it finds a way around whatever’s supposed to be gatekeeping it.

Keep an eye on your own org chart. Somewhere in it, a container is quietly becoming optional. Better to notice before your audience does it for you.

THE PRACTICAL: Everyone’s Building the Same Ride

Mark the time: The land grab is over.
Now everyone’s just redecorating. 🏗️ Social still pulls headlines, but the behaviors are the changing. Pinterest pulled in $1.18 billion last quarter, up 18% year over year. Sounds great until you read the next sentence: usage in the US, Canada, and Europe was flat. All the growth came from “Rest of World,” where revenue per user is lowest.

“AI-powered targeting” is carrying the number. The audience isn’t actually expanding.

Meanwhile, Disney’s raiding the “For You” page… TikTok fan videos about Marvel, Pixar, and Star Wars are now living inside Disney+’s Verts feed, with a Creator Ambassador Program dangling library access and event perks. Streaming inventory is starting to look like social inventory.

And TikTok Live just claimed 50.9% of global live-streaming hours in Q2. More than every legacy platform COMBINED. Twitch, YouTube, and even Genuin still monetize a viewer better, but TikTok won the hours, which is its own kind of flex.

What this is really saying: No one is currently finding “new eyeballs.” For now, everyone’s fighting harder over the ones they’ve got. 👀 And that fight has a look now: Vertical Video.

• Reddit is testing vertical video and audio playback for viral posts. Reddit. A text platform. Or, what was once one.

• LinkedIn is testing its own short-form video feed, joining Instagram, YouTube, Snapchat, and Netflix in the very long list of apps that looked at TikTok and thought “yeah, that, but us.”

The feed lives in a new Video tab. LinkedIn says it’ll be “obviously focused on careers and professionalism,” which is a very LinkedIn sentence to write about a format invented for dance trends and cooking videos.

So here’s the actionable part: 📌

If your growth plan is “get on the new feed everyone’s building,” you don’t have a plan. You have a habit.

Every platform independently arriving at the same vertical-scroll format isn’t a trend you need to chase, it’s a sign the format won and the containers are just catching up. Build for where attention already lives, not for whichever app just cloned it.

That matters more now because the old scorekeeping doesn’t work anymore either.

Sprout Social’s 2026 Influencer Marketing Report found that 60% of Instagram Reels now reach audiences where more than 70% of viewers are non-followers.

Only 17% of consumers say follower count factors into who they follow at all. And yet half of companies still use follower count as their primary proxy for who to partner with.

(Translation: half the industry is still grading creators on a scale nobody in the audience is using. 📊)

A few numbers worth actually sitting with from that same report:

• 81% of Gen Z made a direct purchase off an influencer recommendation this year

• 75% of marketers are already increasing influencer budgets because of it

• 44% of consumers say they’re not comfortable with brands partnering with AI influencers, 30% say it depends

So yeah, everyone’s racing to build AI into the content layer, and the audience is telling you, in real numbers, where it draws the line. Ignore that at your own risk.

The through-line here: Stop optimizing for the platform’s current shape and start optimizing for where attention actually is. The shape keeps changing. Reddit’s proof of that. LinkedIn’s proof of that. The metrics you’ve been using to measure success are proof of that too, several of them are already measuring something your audience stopped caring about a while ago.

Redecorate less. Pay attention more. Ownership is the goal now 🔍

THE PERSONAL: Live Sports and Local Heroes

Few heroes wear capes… but these heroes wear caps.

Five hours before first pitch, my local Little League team puts their phones in a bucket. 📵 Not because a coach made them. It’s a rule they made themselves. But if this is a small sample of what's possible... we should all be celebrating.

It started almost by accident during the New York State Championship run in late July, and it turned into the whole season’s secret weapon.

“I think it’s like an on button for us,” assistant coach Anthony Theoharis told the New York Post. “Once they do that, they get locked in, and they beat a team.”

These are 11- and 12-year-olds. My neighbors. My friends’ kids. My kid’s classmates. Kids who’d already spent a full school year under a state-mandated phone ban and were, by every account, extremely done with the concept.

Gen Alpha. ‘Nuff said.

But when it turned into our first state title in the league’s 63-year history, the national news coverage took notice.

Here’s what they did with the time they weren’t spending on their phones:

• A homemade Family Feud (winning answer: broccoli on pizza, an outcome I have real questions about)

• PS5 sessions crammed into a hotel room (better than social media)

• Getting themselves kicked out of the hotel pool for going down a water slide together all at once.

Third baseman Jack Gomez called the phone bucket “a good luck charm.” My friend Rikki Kien, whose younger son Brandon was the star shortstop and leadoff man, put it simpler: “they just go back to basics and love talking to each other.”

I keep coming back to that word. Basics. Like it’s a setting you have to actively choose now instead of the default.

Port Washington won the state, then beat Rhode Island in regional round, one win from Williamsport, one win from the actual Little League World Series, nationally televised, the whole thing.

They fell one game short, losing to Connecticut last week, narrowly missing out on being one of just eight American teams playing in Williamsport this week.

I’ve gone back and forth on how to write that sentence, because the instinct is to soften it, to find the silver lining and move on fast. But I’m not gonna do that. It stings. Ask any of those kids and they’ll tell you it stings. A season that good, that historic, deserved to end in Pennsylvania, but my FOMO still remains

And even still, I don’t think the story was ever really about Williamsport. 🏆

A bunch of 12-year-olds figured out, on their own, that the thing standing between them and each other wasn’t a lack of talent or a bad bracket draw. It was a bucket’s worth of phones sitting untouched for five hours.

They didn’t need a life coach or a TED talk to figure that out. They needed a coach with a bucket and a group of kids willing to try it, get annoyed, and then ask for it themselves.

That’s the whole point here. Because we have to remind our kids daily (or hourly?) that the phone is not the thing. The game was the thing. The team was the thing.

Once you take the container away, even for five hours, you find out what was actually there the whole time.

No, Port Washington won’t be in Williamsport this week. But every kid on that team already knows something a lot of grown adults, myself very much included, are still working on: sometimes the fastest way to get locked in is to put the thing down.

Congrats to these boys and their families on the best season in 63 years. Fun seeing our town on ESPN a few times. and that state title’s real. The bucket’s real. And somewhere in Williamsport this week, eight teams are playing a game my town already won in every way that mattered (outside the standings). 💙🤍

THE POLITICAL: No One Wants the Responsibility

Safety shouldn’t be political, and yet…

Meta got hit with $942 million for treating kids as collateral damage, a judge calling it a “public nuisance” the same way you’d describe a factory dumping into a river.

Bruce Daisley put it in perspective on BBC Radio: against $61 billion in quarterly revenue, it’s “a drop in the ocean.” Do the math. That’s about one day of Meta’s quarter.

TikTok, too. A confidential report says the company withheld a safety feature from millions of users to measure engagement. Someone died. Bloomberg They settled three more teen addiction lawsuits. How was closing a moderation office and laying off 250 people a good idea? And the EU just charged them, for the fourth time in two years, because private teen accounts turn out to not really be private. “A high level of protection should not be an opt-in,” said EU tech chief Henna Virkkunen. Wild that this needs saying in 2026.

This should make you angrier: While all of that was happening, ByteDance was quietly training one of the largest AI models on the planet, reportedly rivaling Anthropic’s Mythos. Good thing the U.S. government stepped in to save these guys…

Priorities. 🤡

Meanwhile, everyone else is fighting over a much pettier toll booth.

Google’s AI Overviews are now contradicting the paid ads sitting right above them, on the same page, that advertisers paid for. Agencies want to become a “futures market for tokens,” buying AI cheap and reselling it at a markup, principal media trading with a chatbot costume on. Snap said no to fully AI-generated Spotlight content.

Cloudflare built a browser that doesn’t care what a page looks like, only what an agent can extract from it. Everyone’s racing to own the pipe. Almost nobody’s asking who’s accountable for what runs through it.

Which brings us to the one genuinely funny/not-funny footnote of the month:

Anthropic just started watermarking Claude’s own text, worldwide, because the EU AI Act says AI companies now have to prove what they made.Google, Meta, Microsoft, and OpenAI signed on too.

Fortunately none of this newsletter was written using a tool that will soon leave an invisible fingerprint that proves where helps. But others will not be so fortunate. Because the container isn’t just under new management... It’s under new surveillance. 🖋️

Remember… the app, the title, the platform, none of it was ever the business. And when nobody owns the container, nobody owns what happens inside it either.

But somebody has to. Increasingly, it’s a judge, a regulator, or a watermark deciding for you.

Share The Four P's

Subscribe now

Not me, not you… Not marketers, not politicians, certainly not the AI startups, and probably not even my parents, who are currently packing 6-7 decades of life into boxes for their movenext week.

Nope. We’re all up here, mid-set, hoping the notes land.

But… that’s not a crisis. That’s jazz!

This edition is about improvisation… the real kind, not the “we panicked and called it a pivot” kind. And it has a name I keep coming back to:

“Freedom within the framework.”

Build the fundamentals solid enough, and the freedom shows up exactly where the script or the playbook runs out.

So here we go… four instruments, same band:

• The CMO title is shrinking while the job explodes.

• Marketing’s oldest playbook just got torn up mid-year.

• Even founders, politicians, and my parents are winging it in real time, with real consequences.

• And so am I, so are you, so is everyone pretending otherwise on LinkedIn.

THE PROFESSIONAL: Performing Without the Sheet Music.

Marketing didn’t lose its composition. It just found out it never existed in the first place.

We’re all improvising right now… every meeting, every launch, every headcount decision made with a straight face and zero long-term certainty behind it.

This isn’t a complaint. It’s just true, and the sooner we accept it, the better we’ll get at it. So let’s start here

The Job Got a Promotion. The Title Didn’t Get the Memo.

Only 36% of Fortune 500 companies still call their top marketer a CMO. A year ago it was 49%. 📉

Everyone’s reading that as marketing losing influence, but I read it as marketing outgrowing its own name tag. 📛

Companies aren’t demoting marketing. They’re renaming it: Chief Growth Officer, Chief Commercial Officer, Chief Customer Officer, whatever fits on a business card next to “also now owns sales handoffs, retention, and half of product.”

The title shrank because the mandate outgrew it. Not a demotion... It’s a wardrobe change during the commercial break.

What isn’t being celebrated in Adweek or Digiday or even the company all-hands: Most of us were never trained for the job we’re currently doing. Nobody handed marketers a syllabus for owning an entire commercial system. We’re all just riffing… we’re improvising our way through one quarter at a time… through meetings that used to belong to three separate departments, hoping the reps we’ve logged actually show up when it counts.

Many aren’t equipped with this skill, and THAT is why there is such turnover at the CMO position.

Improvisation isn’t the absence of a plan. It’s having enough fundamentals that reacting in real time doesn’t look like guessing.

Every jazz musician worth listening to spent a decade on scales before earning the right to ignore them on stage. Marketing just got told to start soloing without anyone confirming we’d finished the scales.

Exhibit A: Two Companies, Same Instrument, Very Different Solos

This earnings season basically staged the experiment for us.

Microsoft:

• Revenue up 18%, Azure growth up 43% 📈

• Added the single biggest one-day chunk of market cap in company history

• Bet everything on efficiency: cheaper models, lower token costs, tools bolted onto a stack people already trust

Meta:

• Free cash flow collapsed 91% in a single quarter

• Betting big on frontier scale, with a cloud-business tease that had no product, no timeline, and no name

Same industry. Same moment. Wildly different solos.

One company riffed off fundamentals it already owned. The other tried to out-loud its way into relevance.

Investors, it turns out, can tell the difference between someone who knows the changes and someone just playing loud.

Exit Stage Right: Layoffs Contradict Themselves in Real Time

If you want proof nobody has a script right now, just watch the headcount whiplash:

• Monday.com cut 20% of staff to fund an AI platform, while insisting, almost too quickly, that this “was not made to reduce costs or replace people with AI”

• Amazon laid off staff inside its own AGI unit in order to focus more… on AI (Ummm, what?)

• Alphabet, Booz Allen, ServiceNow, and others are quietly hiring humans back, because deploying AI turned out to be a genuinely hard execution problem.

• Jack Dorsey confidently predicted in late February that "the majority of companies will reach the same conclusion" as Block within a year.

Five months later, the industry’s conclusion appears to be: actually, we need people.

This is what happens when you solo before you’ve learned the changes. You find out live, on stage, in front of everyone, at the same time your audience does.

This Feels Familiar… Because We’ve Actually Done This Before

Years ago, before “brand newsroom” was a slide in somebody’s deck, I built one of the first versions of one inside a digital agency (Hello, 360i Social Team from 2011!)

The concept was simple: watch the news of the day, and figure out fast whether one of our brands could join the conversation or lead it.

No script. Just:

• A team with strong fundamentals

• A shared sense of the brand’s actual voice

• Enough muscle memory to move together the second something worth reacting to showed up

• Tweet when the lights go out at the Super Bowl and turn on the PR machine

That model still works today, whether you’re in-house or at an agency, because the internet hasn’t gotten any slower since I built it, and neither have people’s expectations. I wasn’t prepared for every headline. I was prepared to react to any headline. Different skill. Better odds.

TL;DR: The Actual Point

Is what I’m saying noise or actual music? Because none of this is really about AI breaking marketing, or Microsoft simply being smarter than Meta, or layoffs proving anything on their own...

It’s that the companies, the marketers, handling this moment well all did the unglamorous work first: the reps, the rituals, the fundamentals nobody claps for.

Everyone else is just making sounds and hoping it sounds like a song. And nobody’s getting handed sheet music anytime soon.

So learn your instrument because you’re going to be soloing for the rest of the year… and so is everyone else.

THE PRACTICAL: No One Knows What You’re Thinking Anymore (and Neither Do You)

For as long as I’ve been in this biz, marketing had one job: get people to click.

That job is dead. Not dying. Dead.

And the unwritten eulogy (until now, I guess?) is that the thing that killed it also killed the thing that was supposed to replace it.

Let me explain, because this is where the improvising gets genuinely hard…

Act I: Attention Was the Whole Game

You know this part. Optimize the headline, win the click, count the pageview, sell the ad. Publishers built entire businesses on it. Some still are, most of it was bullshit because we lazily (or habitually) followed some script.

Then AI Overviews showed up and started answering the question before anyone had to visit the page to find out.

Now roughly 60% of Google searches end with zero clicks at all. Publishers who spent two decades chasing traffic are staring at a funnel with the bottom cut out, and some, including Reuters, USA Today, Politico, and People, are reportedly debating whether to just cut Google off entirely and see what happens.

Attention, as a strategy, is over. It’s not coming back. Stop grieving it.

Act II: Intention Was Supposed to Save Us

The industry’s answer was smarter than chasing clicks:

Stop measuring where people looked, start understanding what they actually wanted. Intent data. Buyer signals. The whole “we don’t need more traffic, we need the right traffic” pivot.

Except intention just went invisible too.

People aren’t Googling and clicking through five tabs anymore. They’re having a private conversation with Claude or Gemini, comparing options, resolving objections, deciding what they want, all before your website even enters the picture.

Oh, and you don’t see any of it as a marketer. You just see the outcome, a visitor who already knows what they want and finds your homepage vaguely insulting for not knowing it too.

There’s a name for that gap: the space between what a customer has already figured out and what your page assumes they know.

I am calling it the “intent gap.”

You can call it whatever you want. Just know it’s the reason your “personalized” experience feels like it’s talking to someone who wasn’t in the room for the actual conversation.

Act III: So Now It’s Contention

If you can’t chase attention and you can’t see intention, what’s left?

Fighting for the one thing AI still can’t fake convincingly: actually understanding the room.

The data on this is honestly kind of brutal, but in a good way (?):

• 41% of long-form LinkedIn posts are estimated to be AI-generated, and AI-flagged posts get roughly 45% less engagement than human ones.

• TikTok’s own research says relevance beats volume, and calls the thing AI can’t replicate “cultural intelligence”: reading what a community values as it’s forming, not after.

• LinkedIn just shipped a “seems like AI slop” button. On its own platform. For its own users. I have no idea what comes next, but that really feels like a confession.

Even the AI companies are figuring this out in public with no rehearsal.

• OpenAI has 32 open advertising roles and no fully built sales team yet.

• Google’s testing AI-written ad copy on other people’s product listings.

• Claude’s model is now autonomously hacking websites…

Everybody and everything is improvising the exact same channel at the exact same time, and you’re all sharing the stage.

Act IV: Retention Is the Only Solid Ground Left

Here’s the good news buried in all of that chaos. Attention failed. Intention is dumb. Contention is spiraling.

So the only thing you can actually build a strategy on is the relationship you already own. Your relationships. Your ideas. Your community. The audience that shows up because they trust you, not because an algorithm nudged them your way this one time.

That’s not a consolation prize. That’s the whole game now. So maybe the advice should be hunker down, block out the noise, retreat to your bunkers and hope for the best… and see you when it’s safe to come out?

Act V: The End, Which is Really Just the Practical Part

You don’t need a 5-point content strategy to survive this. You need:

1. Stuff that sounds like an actual person said it, because increasingly, that’s the only stuff getting through…

2. A direct relationship with your audience that doesn’t rely on someone else’s algorithm to introduce you

3. The humility to know you’re guessing in real time, same as everyone else in your feed

Shakespeare had a framework for his plays. Beethoven didn’t. Miles Davis stayed just outside the lines. Coltrane never knew lines existed.

See? Nobody has this figured out. Not the publishers, not the platforms, not the AI labs building the ad units as they go.

You’re not behind. You’re just improvising in front of a slightly bigger, slightly more confused band.

THE POLITICAL: Nobody Has a Setlist, Either

Marketers aren’t the only ones improvising this year. Everyone running for office in this tumultuous midterm season is either foolishly playing an old tune… or just making it up as they go.

We’re wrapping up primary season this week for mid-term elections, and both parties are having the same identity crisis at once:

What happens when the fundamentals you used to rely on stop working, and you have to figure out the rest live, in front of voters, with the mic already on?

On the left: The fight is between the socialist wing and the centrist wing, and Israel is the fault line. In Michigan, AIPAC has poured nearly $30 million into backing Haley Stevens against Abdul El-Sayed (this was written on Primary Tuesday night with the MI results still too close to call), its largest single-race investment ever.

Add in cryptocurrency, AI, Big Tech, and online betting money flooding Democratic primaries this cycle, and you get a party trying to figure out in real time whether taking that money costs more support than it buys.

On the right: The question is simpler and just as unresolved: are you with Trump, or are you willing to break from him? MAGA candidates are watching their own support slip and improvising an answer nobody’s fully committed to yet.

Layer in Israel, Iran, immigration, insurance, inflation, and “I” (a.k.a.: the plain selfishness of “what does this mean for me?”) and you’ve got six variables changing shape every single day between now and November.

Nobody’s running on a fixed platform anymore. They’re reacting to whatever the news cycle hands them and hoping the reflexes hold.

Which is where the improvising stops being interesting and starts being dangerous.

Two ICE shootings happened within a week of each other last month. Lorenzo Salgado Araujo, on his way to work in Houston. Johan Sebastián Durán Guerrero, a 25-year-old father driving to his delivery job in Biddeford, Maine. Neither man was the target of the operation that killed him. Neither officer was wearing a body camera.

The agency’s response wasn’t a plan. It was a scramble: pause traffic stops, reverse the pause a day later, promise more cameras eventually.

Sure, that’s improvisation, but the bad kind… with no fundamentals underneath it. Like when 3 year-olds at a mommy and me music class are all banging and blowing instruments at the same time.

And voters noticed. A majority don’t trust ICE to accurately describe its own operations. When asked who they believe about what happened in Houston, more people trust the eyewitnesses than the agents involved, by a wide margin.

Which is really the difference between the two kinds of improvising running through this whole newsletter. One kind is reps and fundamentals underneath every note that sounds spontaneous. The other kind is showing up with no plan, no accountability, and hoping nobody asks hard questions until after the set’s over.

This November, we’ll find out how many voters can still tell the difference.

THE PERSONAL: Freedom Within the Framework

It’s a phrase I keep coming back to: “Freedom within the framework.”

That’s the whole trick of improvisation, actually, not just for musicians. You build the fundamentals so solid that you stop needing a script, and the freedom shows up right where the framework ends.

I get asked sometimes how I post as much as I do to social media without losing my mind or my voice or my job. They’re all related, actually.

And the honest answer is that most of what you see from me isn’t scripted. It’s improvised, on purpose, inside a structure I built specifically so I could improvise without it turning into noise.

The “What I Learned This Week” video series is one wall of that structure. Real Talk Saturdays is another. “Those Who Impacted My Operating System” is a third.

The Snarketing Podcast is the whole other jazz band.
(And what a great episode this week with Scott and Steve!)

None of those formats tell me what to say. They tell me when and how to show up, which turns out to be the harder problem. Once that’s solved, the actual content just… happens.

A thought lands on a Tuesday, I say it Tuesday. Maybe I wait a week. It doesn’t follow a script. It follows the broader narrative I’ve been building for years, which is really just another word for muscle memory.

None of it is perfect. Some of it, if I’m honest, is a little unhinged. But that’s kind of the point. Improvisation isn’t about being unprepared, it’s about knowing your yeses and your no’s well enough that you don’t need a teleprompter to know where the line is.

Consistency and progress over polish and perfection every single time.

Meanwhile, My Parents Are Also Improvising

My parents are moving next week. Downsizing, the way it happens when the house has outgrown the life actually being lived in it.

There’s no jazz metaphor sophisticated enough to make packing decades worth of a family and home into boxes feel graceful. It’s just hard, and a little sad, and also kind of beautiful in the way that anything is when you watch people you love figure out a new chapter in real time, with no rehearsal, because nobody gets a rehearsal for this part.

They’re improvising too. They just don’t call it that.

The Kids Are Also Figuring This Out, Just Louder

Gen Z, meanwhile, is out here buying harps. Like LITERALLY!

Instrument sales are up, and a lot of it is 20-somethings picking up violin, piano, trumpet, whatever, specifically because it can’t be faked. There’s no AI shortcut for calluses. One harpist put it about as well as anyone could: people are starved for real experiences, things they can’t get online, and an instrument gives them something with actual friction. Skin against string. No autocomplete.

I love that. A whole generation raised on infinite scroll, instinctively reaching for the one thing that has to be earned one rep at a time.

Not everyone’s improvising from the same place, though, and it’s worth sitting with that for a second before I wrap this up.

There’s a 23-year-old out there right now who did everything the old script told her to do, degree, internship, plan, and still can’t get a callback in her field. She’s doing part-time admin work and wondering what the point of any of it was. That’s not a metaphor. That’s just Tuesday for a lot of ambitious 23-year-olds right now.

And there’s a 4th grader somewhere worried there won’t be any jobs left by the time she grows up, which is heartbreaking and also, according to actual economists, probably not how it plays out.

Nobody in 1910 could’ve imagined cybersecurity as a career. We tend to build the next thing before we can see it coming.

Honestly, I don’t know how to resolve that gap. Maybe I’m just feeling my age today, but the 23-year-old’s fear is also real right now. The 4th grader’s fear is real…eventually. Both of them are improvising toward a future neither one can fully see, same as the rest of us.

Maybe that’s the whole point here. None of us can see the whole set list. We just build the fundamentals strong enough that when it’s our turn to solo, something worth hearing comes out.

That’s freedom within the framework. That’s the only framework I’ve got.

Share The Four P's

Subscribe now

Not the writer. Not the voice. Just... the thing that remembers what I fed it three days ago and 10 days ago in the middle of the night that I forgot the next morning.

I’d think of something, read something, dump it in a running note, forget about it, read something else, dump that in too. Then ask it to remind me what I’d been collecting and tell me what actually connected.

Kinda basic, honestly. The kind of thing productivity bros have been doing since 2023.

But I’m stubborn about my systems, and especially my writing. Also, I think there’s something else going on here too. I’ve been using AI for years, for plenty of things smarter than “please be my brain.” Setting it up as an actual daily habit, the boring unglamorous kind, took longer than it probably should have. Not out of resistance. More like... the tools existed before I’d built the muscle to actually use them well.

Which, it turns out, is the whole theme of this issue.

“Cultural lag” isn’t new. It’s just usually invisible until something forces you to notice the gap between what’s possible and what anyone’s actually adjusted to. The printing press outran the Church’s authority by decades… Cars hit the streets before speed limits existed.

Even TV timeouts eventually stopped feeling like an insult to the game. Fans just needed a name for it. This year, the World Cup gave us one: the “Hydration Break.”

We keep building the thing before we write the manual. This summer gave us a masterclass in that gap. On the pitch, in the boardroom, in Washington, at a certain ballpark in Philadelphia that spent a week booing basically everyone who wasn’t wearing red pinstripes.

THE PROFESSIONAL: The New Instruction Manual Hasn’t Shipped Yet.

LinkedIn published a blog post last week about building trust through “sustained executive thought leadership.”

Convenient timing, since new data from Pangram found that 41% of long-form LinkedIn posts are fully AI-written. Add 1/3 of longer Twitte… err… “X” posts, and you’ve got a platform teaching executives to build trust brick-by-brick while roughly half its content is a machine wearing a blazer.

(LinkedIn also published tips on optimizing your B2B marketing posts for AI systems, in case the irony wasn’t thick enough already.)

Nobody planned this contradiction. It just happened, and it’s still happening because the tools got good before anyone agreed on the etiquette.

The visibility layer already moved

Here’s a stat that should worry anyone who spent the last decade optimizing for Google: The overlap between top-ranking search results and the sources AI actually cites has fallen from 70% to under 20%.

I’ve seen this stat now cited like 10x this past week alone. Yet being recommended by an AI system and being cited by one aren’t the same thing.

The recommendation happens first, quietly, somewhere upstream of the link. By the time you’re optimizing for citations, you’re optimizing for the bibliography. The decision already got made without you.

Meanwhile, the machines misbehave

Meta opened its ads MCP server to developers, making its ad tools easier for AI agents to use. Nice.

At the same time, anyone running more than a 1-person SMB marketing team is still laboring through unusable platforms like Canva… and even Meta's own ad tools have become so error-prone that some advertisers have simply stopped using them: garbled text, twisted limbs, products that mutate mid-campaign, 'enhancements' flipped on without anyone approving them.

Meta’s official position is that advertisers need to review every output. Helpful.

Which is a very polite way of saying the tool is unsupervised and the humans are the safety net. At some point that stops being automation and starts being babysitting with extra steps.

The reshuffle nobody scrolled past

No, not everything lagging is embarrassing. Some of it is just underreported.

When Best Buy created a Chief Ads and Media Officer role for Lisa Valentino last week, it shows that retail media graduated from “line item bolted onto a vendor deal” to its own department… an industry finally catching up to money it’s been leaving on the table for years.

And yet, it’s the one function still most likely to hit an innovation plateau.

And then there’s Netflix, whose stock price is down nearly 50% over the past year, even moving its engagement reporting from twice a year to once (yes, we noticed). Viewing hours grew 2% while the industry speculates about ad tiers and acquisition targets (Walmart, if you’re wondering, already owns the glass through Vizio; Netflix would just be what plays on it).

The pattern… if you squint

Every one of these is the same shape: The infrastructure updates. The rules, norms, and etiquette lag behind by months or years. Somebody profits in the gap, somebody gets burned in the gap, and eventually the culture catches up and it all looks obvious in hindsight.

We’re just living in the gap right now. Might as well take notes (and store them in a Claude project).

THE PRACTICAL: Move Fast and Fix Things… Maybe Never?

Google processes over five trillion searches a year. Oumi’s new analysis, commissioned by the New York Times, found AI Overviews are accurate about 90% of the time.

Is that good? Who knows…?

Do the math on that 10%, and you get, depending on how you count it, somewhere between tens of millions and hundreds of thousands of wrong answers every single hour, delivered with the visual authority of the world’s most trusted search engine.

Either way, it doesn’t feel like a minor imperfection. That’s misinformation operating at industrial scale.

Obviously because it’s AI, the model doesn’t know it’s wrong, either. It’s not looking anything up. It’s predicting the next most likely word, confidently, whether or not that word happens to be true.

On the flip side, people DO believe it. Nearly 90% of people trust an answer more when it’s phrased with certainty, regardless of whether the certainty is earned. Which means the more confidently AI is wrong, the more we believe it.

Security teams are the ones catching this in real time

About 80% of enterprises reported an AI-related security issue in the past six months. Half experienced an actual incident from an unauthorized or misconfigured AI agent. 90% of companies say they’ve “discussed” AI governance, but only half of those have an actual budget attached to the discussion.

That gap between discussing and funding is the whole ballgame here.

AI agents move at machine speed with none of the identity, access, or audit controls a company already requires of every human employee. Which means the department everyone wants to skip past on the way to shipping AI faster is also the only one currently standing between “innovative” and “incident report.”

SECURITY!!!

The optimistic version of this: the more mature risk programs stop being the department of no and start being the reason things ship safely at all. The pessimistic version: most companies aren’t there yet, and they’re deploying anyway.

The open web is paying for it now, but will ultimately benefit from it

Publisher ad supply fell up to 40% YOY this quarter, because AI search increasingly answers the question without ever sending anyone to the website with the answer on it. eCPMs are up to soften the blow, but that’s a scarcity tax, not a recovery. Fewer visits, pricier remaining inventory, and a business model quietly rewriting itself in real time.

But given the aforementioned social limitations, and the quick catch-up with content that’s now possible in owned and ownable destinations, that will soon change.

For example: Apps grew. Subscriptions and newsletters are being treated less like nice-to-haves and more like life rafts for bringing people to safety.

Nobody planned this reallocation. It’s just what happens when the thing sending you traffic decides it doesn’t need to “send you traffic” anymore.

Meanwhile, demand doesn’t care about any of this

AI demand is almost unlimited. More demand than we’re able to fill. Compute is scarce, everyone wants more of it, but NIMBY alerts are going off everywhere now, and communities are increasingly saying no thanks.

Including mine. New York just banned new hyperscale data centers for a year, joining 70% of Americans who don’t want one built anywhere near them.

The shape of it

We’re starting to see the cultural lag pattern here, right? The tech scales before the governance does. The traffic disappears before the business model adjusts. The demand outruns the actual willingness of anyone, anywhere, to host the thing generating it.

Eventually somebody writes the manual. In the meantime, we’re all reading the AI Overview and hoping we’re in the 90%.

THE PERSONAL: Breaking Metal, and Blowing Shit Up

A few months ago, my son asked me to buy him an air compressor.

Even though I did not know what an air compressor did (I still don’t, not fully), but I said yes, because that’s apparently the deal now…

I assumed it was the thing you go to the gas station for to put air in your tires. Which is actually kinda right.

So off to Home Depot we went, and it has been a constant and well-used fixture in our garage ever since: Getting dirt out his RC car motor, cleaning his drone, inflating bike tires, blowing out our home air filters, and at least 50 other purposes I’ve stopped asking about.

The kid likes his tools. Good for him. In a future economy where trade skills will matter, he’s got a manual curiousity.

But then last week, the air valve snapped clean off from the compressor. Sounded bad. Metal severing metal, dangerous even. In our garage. Because of something my son was doing with a machine I bought without understanding.

His solution makes sense, too: We need welding tools now.

I want to be clear that I don’t actually know if a “welding tool” is a blowtorch, several blowtorches, or something else entirely that I will also not understand. But he’s 15, and this is not exactly the kind of trouble I was expecting to get a call about, and honestly…? I’m into it.

Turns out he’s not alone

Gen Z enrollment in two-year trade schools is up nearly 20% since 2020. A big chunk of them will tell you, unprompted, that they picked it specifically because AI can’t do it.

There’s a kid in Staten Island running a TikTok account called Trades Over College, posting things like “POV: AI can’t replace HVAC.” He makes six figures. His girlfriend, at one point, asked him not to tell her parents he went to trade school instead of a four-year.

Which is its own little cultural lag story, if you think about it. My town is one of those college-competitive crazy towns. And it’s a sickness.

Which is why, personally, I’m excited about the prospect of an entire generation recalibrating what’s actually safe to build a career on, faster than the adults around them are willing to update the advice they keep giving out.

“Go to college” was the correct answer for a long time. It might still be, for plenty of paths. But somewhere out there, a 21-year-old with a blowtorch and a TikTok account is making more than most of his former classmates, and nobody sent out a memo updating the script.

I’m not ready to declare my son the next great American welder. Nor am I going to put a propane tank and torch in my garage next. Mostly he just wants to fix the thing he broke, and maybe make something cooler than the thing he broke.

Still not buying him a blowtorch without research, but I noticed I didn’t really push back. This time I just said let’s figure out what’s safe.

Maybe that’s the actual lesson tucked in here. It’s not that the world changed and I finally caught up. It’s that I’ve got a kid who keeps handing me evidence, one broken air compressor valve at a time, that the old script for what a smart, safe, promising path looks like needs a rewrite. And I’m updating faster than I used to.

THE POLITICAL: Everybody’s Booing, Nobody’s Listening

Let’s start with the silly version, because it’s the funniest and least likely to start a fight at your next barbecue.

Last week, Philadelphia (the only other city I’ve lived in besides NYC) hosted this year’s MLB All-Star Game specifically to mark America’s 250th birthday. A unity moment.

At least on paper. Instead, Philly booed the MLB Draft, booed the Futures Game, booed the Home Run Derby runner-up, booed Mr. Met during mascot introductions, and booed basically anyone on the field who wasn’t wearing red pinstripes.

The city hosted a party for the country and spent the whole week yelling at the other guests.

Nobody’s shocked. That’s just Philly being Philly. But it’s a decent warm-up for my boo-bird barrage based on what just happened at much higher stakes.

Even soccer couldn’t stay above it

This World Cup gave us genuinely great stories. Norway. Cape Verde. Paraguay knocking out Germany. Argentina and Messi grinding past England to reach the final. On the pitch, it delivered.

Off the pitch, it got, well, messy:

• The 48-team expansion diluted group play into blowouts and snoozers.

• Dynamic pricing left premium sections half-empty while regular fans got priced out.

• VAR kept interrupting the sport’s rhythm in the name of getting calls right.

• Every goal, every foul, every non-foul became a dividing line.

And then politics, because it never misses an opening, found its way in anyway: the President personally lobbied FIFA to lift a suspension on a US player, got his wish, and watched the US lose to Belgium anyway, who celebrated by mocking him in their own locker room. A tournament that didn’t need the noise got it regardless.

We had a moment, these United States, to build a future with soccer in it. We failed miserably, and I think the sport may take 20 years to recover.

Then the noise got a lot less funny

Last week, the President gave a “primetime undress” built around newly declassified documents he said “proved American elections are dangerously vulnerable.”

Independent review found those documents largely restated things the intelligence community already flagged back in 2021: Nothing. No new evidence of an actual fraudulent vote was presented.

Democratic governors called it an early attempt to lay groundwork for disputing the midterms if things don’t go his way.

Meanwhile, actual voters are telling anyone who’ll listen what they want, and it isn’t a re-run of 2020. Swing voters currently favor a Democratic candidate by 12 points, but almost half are still undecided, and the policies moving them aren’t culture-war fights. They’re raising taxes on the wealthy, expanding healthcare, restricting utility price-gouging, even banning AI from setting your prices or wages based on your personal data.

That’s not a moderate ask. That’s not a progressive purity test either. That’s just economic populism, plainly stated.

Where I land

The false choice between MAGA and the DSA is garbage, and a hype narrative that loud extremists want you to think is true. Both ends have convinced themselves they’re the only adults in the room, and both keep proving otherwise.

The rest of us are wringing our hands about it at cocktail parties as the radicals on both ends are better organized, showing up to primaries, and reshaping the parties from the edges in. NY-13, NY-7, NY-10... all went to younger, further-left primary challengers this summer, powered by exactly the kind of energy Fox Sports wishes it could manufacture with an inflatable Messi.

I’d rather build the coalition of people actually trying to lower costs and reward the people doing the work, instead of just booing whoever’s standing closest.

The shape of it

Every institution here staged a moment meant to bring people together, a World Cup, an All-Star Game, a primetime address, and the actual behavior underneath refused to cooperate with the script.

That’s not new.

What IS new is how fast the gap between the ceremony and the reality gets broadcast, clipped, and reposted before anyone’s finished processing it.

Maybe someday we’ll catch up. So let’s keep throwing unity parties... but not be surprised when we keep booing through them anyway.

*AI Disclosure: 95% of this was written by me, a human, but pulled together my thoughts and random notes with Claude… and only light edits and typo corrections from Grammarly.
The images are mostly AI-generated using Nano Banana.

Share The Four P's

Subscribe now

But everyone LOVES a good preview or prediction piece, so here’s another one…

I spent last week by a pool, jetlagged and half-sick, reading a book about why that kind of certainty is a trick. The author’s argument: your brain doesn’t see reality, it builds one.

A prediction machine that writes a story about the world, then defends it long after the world stops agreeing.

It’s the second half of 2026, and everyone already knows how the year ends.

Which makes writing a 2H 2026 preview a little absurd. A preview is a prediction, full stop. It’s my brain doing the exact thing the book warned me about, out loud and on purpose.

So, fine: Preview with predictions, disclaimer included.

Let’s find out what my brain made up.

THE PRACTICAL: Automating the On-Ramp

Welcome to the second half of 2026. Same year, new hallucinations.

If a preview is really just a prediction in a nicer outfit, let’s start where the predictions have already been tested against reality and lost. The biggest workforce bet of the first half is coming apart in the second, and almost nobody wants to say it out loud yet.

So I will.

The Bloodbath That Wasn’t a Strategy

One estimate clocked roughly 40K tech layoffs last month, the highest single-month total in two years, AI the most-cited reason. Meta, Amazon, Microsoft, Block, the whole marquee lineup, trimming headcount while shoveling cash into their AI transitions.

The pitch to investors was clean. Automate the humans, keep the margin, win the quarter.

The pitch is not holding.

Ford just hired back around 350 veteran engineers, several of them former employees, after its automated quality systems couldn’t find the failure points its people used to catch before a part ever reached the plant floor. The line that should be laminated and mailed to every CFO who cut first and thought later came from Ford’s own VP of hardware engineering, who admitted the company mistakenly assumed that feeding design requirements into AI would produce a quality product. It didn’t. The tools, he said, needed to be trained by the most experienced people in the building.

Read that twice. The AI didn’t replace the experts. It needed the experts to be worth anything.

Augment, Don’t Amputate

This isn’t one automaker eating a little crow. It’s the pattern.

PwC’s latest research found that companies using AI to amplify their experts are pulling away from the ones using it to replace people, and it isn’t close. AI-exposed firms grew headcount 52% against 36% for everyone else, and wages 24% against 17%. Demand for actual AI skills is climbing eight times faster than demand for everything else.

The tell is in PwC’s own vocabulary. “Professionalized” AI, the kind that makes an expert sharper, grows jobs. “Democratized” AI, the kind that lets a non-expert fake a skill, does not. The value was never in the shortcut. It was in the person who didn’t need one.

Which brings me to the part that should worry all of us, not just the finance chiefs.

The Ladder is Missing Its Bottom Rung

The ugly part hides inside the good news. Entry-level roles at AI-exposed companies are now seven times more likely to demand “human-intensive” senior skills... judgment, leadership, the stuff you used to spend a decade earning. Those “seniorized” jobs are up 35% since 2019. Entry-level openings are down 10%.

We automated the on-ramp.

The junior work AI can actually do is the same junior work that taught people how to become senior. Kill the first rung and you don’t get a leaner workforce, you get a generation with nowhere to start and a hiring manager demanding they show up pre-seasoned. (Ask the average 24-year-old how that’s going. Then meet me in the Personal section.)

The Second-Half Call

So here’s the forecast for the back half of 2026, and I’ll plant the flag now so you can screenshot it in December.

The replacement era peaks and reverses. The smart money stops asking how many people AI can replace and starts asking how good the people are that they’re pointing it at. Not because anyone got sentimental. Because the augmentation crowd is already winning on the scoreboard, and scoreboards are contagious.

The bets get louder in the meantime. General Intuition just raised $320M at a $2.3B valuation to build AI that learns to act by watching gamers, and TikTok shipped an Agentic Hub so agents can run your ad campaigns end to end. Impressive, even. But every one of those bets still bottoms out at the same unglamorous truth Ford paid full price to relearn.

The model is only as good as what you feed it. And the best thing you can still feed it is a human who knows what they’re doing.

That’s not nostalgia. That’s the whole second half.

THE PERSONAL: Your Brain Is Making This Up

The guy next to me on the flight home from Paris raw-dogged the entire trip.

No movie, no show, no music, no book. Eight hours, screen off, staring at the seatback like it owed him money. I must have looked at him a beat too long, because he offered an explanation I didn’t ask for.

“AI has ruined movies for me.”

Now, that’s… a take. I sat with it. I still am. Because, honestly, we’ve had special effects for the better part of a century. We’ve had CGI armies and de-aged actors and entire fake cities for years, and none of that ruined anything for him. But somewhere in the last eighteen months a switch flipped, and now this dude and his weird mustache would rather stare at gray plastic for a transatlantic haul than risk two hours with something a machine might have touched.

He didn’t skip a movie. He refused to run the experiment.

Summer Reading

Clearly I couldn’t stop thinking about him, because a week later, I’m writing about him. Jetlagged from Cannes, dragging home a cold as a souvenir, brain firing in weird directions, scrambling new ideas into old ones at 3 a.m. like a search engine with a fever.

So I did the only thing the heat would allow once I got back. I sat by the pool, occasionally in it, and read a book about exactly that.

A Trick of the Mind, by Daniel Yon, a cognitive neuroscientist who runs something called the Uncertainty Lab. Its argument is simple and a little unsettling: the world you experience is less a thing you perceive and more a story your brain invents. Your mind is a prediction machine. It builds a model of reality, then spends its energy defending that model, sometimes past the point where the evidence has stopped cooperating.

Which is to say my seatmate and I were reading the same book from opposite ends. His brain built a model, AI ruins everything, and then protected it by refusing to look at anything that might prove it wrong. Mine was doing the messier, better thing, running the experiment, testing the story, letting the cold and the jetlag knock a few stale predictions loose.

This is the Brain Working as Designed

Neither of us is malfunctioning. That’s just what the machine does.

Yon’s point is that the same equipment that produces genius produces delusion. The brain that leaps to a brilliant insight is the same brain that clamps onto a bad theory and won’t let go, and it feels identical from the inside. Certainty is not a signal that you’re right. It’s just the feeling of your model holding together.

That should make all of us a little humbler heading into the back half of this year, because everyone I know in media and tech and marketing is walking around radiating certainty right now. AI changes everything. AI changes nothing. The click is dead. The click is fine. We’re all so sure. And Yon would gently point out that the confidence is coming from the same place the flight guy’s is. It’s a model defending itself.

Why I’m Telling You This…

Because this whole edition is a preview, and a preview is just a prediction wearing a blazer.

Every forecast in here, mine included, is my brain doing the exact thing Yon describes. Building a model of what the second half holds and then making a confident case for it. I’d be a fool to pretend I’m the one person whose predictions come from somewhere cleaner.

So read the rest of this the way I’m trying to write it… Not as a man who’s figured out where 2026 is going, but as one more prediction machine, running the experiment out loud, willing to be wrong.

The flight guy would’ve turned this section off… but I think that’s exactly why it was worth writing.

THE PROFESSIONAL: Nobody Clicks Anymore, and That's the Point

Ok, back to the important stuff. Like Click-through rates!!!

Everyone carries a number in their head for what a “good” organic CTR looks like. Everyone’s number is too high.

Whether from search, AI LLMs… or somewhere else, people are clicking through to websites less and less.

Ahrefs crunched Search Console data across more than 400K websites, and a good whole-site organic CTR lands between 1% and 2%. Not the 10% that lives rent-free in every marketer’s skull. That 10 is position-one, single-keyword vanity math, and it’s been lying to your quarterly deck for years.

And what’s going to hurt most in the back half of 2026: it’s getting worse.

And on purpose.

The Answer is Eating the Click

When Google shows an AI Overview, clicks to the top organic result drop roughly 58%. That’s up from the 34.5% Ahrefs clocked not long ago. The number nearly doubled while you were reading Cannes recaps.

Look at who survives. Adult sites lead every category at 7.53%, more than double the runner-up, for the profoundly unsexy reason that Google almost never bolts an AI Overview onto those queries. Health and pets sit at the bottom, because the Overview answers the question before a human has any reason to click.

So the pattern is locked. Wherever the machine can hand over a finished answer, it does, and the source that fed that answer gets nothing. No click, no traffic, no credit.

That’s a cognitive shift before it’s a marketing one. An AI Overview delivers a pre-built model of reality and deletes any reason to go test it against the source. The summary becomes the belief. Do that a few billion times a day and being right matters less than being the thing the answer was built from.

Citation is the New Currency, and It Has a Catch

If the click is dying, the thing worth chasing is getting cited inside the answer itself.

LinkedIn’s own Davang Shah wrote the tactical playbook for exactly this, and it’s clean, mechanical advice: lead with keywords, frame everything as a question and a tight answer, keep it to 200-300 words, hand the model structure it can lift without effort.

Follow it and yes, you’ll get cited more.

You’ll also sound identical to everyone else who followed it. Here’s the catch nobody wants to print: the format that makes you legible to the machine is the format that makes you forgettable to a person. Every post flattens into the same keyword-front-loaded, Q&A-scaffolded, list-of-three pablum, optimized straight into interchangeability. (I break most of those rules on purpose. You may have noticed. As far as I can tell, being a little unliftable is the last moat anyone has left.)

Meanwhile… The Whole Layer Goes on Sale

While you’re busy adapting, the platforms are busy charging rent. Because that’s what they do best.

Google is testing AI summaries beneath Search ads, so its model now writes copy under the copy you paid for and controlled, complete with a disclaimer that it “can make mistakes.” Comforting. OpenAI is hiring for image, video, and conversational ad formats, which puts ads inside the answer box itself. And Meta is paywalling its best AI at $20 a month.

Ads in search, ads in chat, paywalls on the features. Every surface that used to be free discovery is getting enclosed and metered at the same time.

The Second Half Play

So here’s the flag for H2. Screenshot it for December.

Stop optimizing for the click. It’s a dying metric bolted to a shrinking pipe. Start optimizing to be the source the machine can’t answer without... the primary thing, the quality input, the take too specific to fold into someone else’s Overview.

There’s a counter-bet worth watching, too. Enthusiasm for AI-generated content has cratered to 26% from 60% two years ago, Fizz just raised on a “human, anonymous, OG Facebook” pitch, and LinkedIn’s new collaborative posts turn a verified human co-sign into distribution. The more synthetic the feed gets, the more an actual person is worth.

Machines reward the liftable. Humans, still, reward the ones worth stealing from.

THE POLITICAL: Whose Reality Is It Anyway?

Politics in the back half of 2026 comes down to one fight, and it isn’t left versus right. It’s who gets to write the model everyone else has to live inside.

Yon says your brain builds a story and then defends it. Turns out institutions do the same thing, except their story becomes your ad, your zoning law, your training data, and your kid’s idea of what a real person looks like. The second half is a scramble over authorship. Whose version of reality gets to be the default one.

Three fronts, all moving at once.

The Fox Filed a White Paper

Google published its blueprint for AI regulation, and credit where it’s due, it reads beautifully. A “pragmatic middle way.” Voluntary audits. An industry-backed body to set the safety bar. The kind of document that sounds like adults talking until you clock who’s holding the pen.

The load-bearing word is “adapting,” as in adapting copyright law to handle AI training. That’s the blazer-wearing way of saying the small matter of whether a company gets to hoover up the entire web to train on should be settled on terms convenient to the companies doing the hoovering.

So yeah… let’s just say the obvious thing: The people who made the inputs, every writer, artist, newsroom, and creator whose work is the training data, aren’t at this table.

They’re on the menu.

And the whole second-half AI story turns on whether that changes, because value is migrating to exactly the quality human inputs this framework treats as free feedstock. Whoever gets to define “reasonable” here wins a decade. Right now that’s the party with the most to gain from the definition.

The “Show Your Work” State

Compare that to Albany, of all places. New York’s synthetic performer law is live, and it does the one thing Google’s white paper carefully avoids. It forces disclosure. Use an AI-generated performer in an ad, a lead, a background extra, even a fake hand model, and you say so or you eat a fine.

Almost quaint. A law built on the radical premise that a human should know when the person selling to them isn’t one.

And the instinct is landing, because enthusiasm for AI-generated content has cratered (as mentioned above). The audience’s own model is updating in real time, souring on the synthetic, and the state just codified that gut feeling into paperwork. Google wants to govern AI with a handshake. New York decided to govern it with a receipt. I know which one I trust, and it’s the one carrying a $5,000 penalty.

The Guy With 181 Receipts

Which brings me to my favorite accountability project of the year. Brian Patrick, a former developer with a Harvard law degree, has spent 2026 posting one unhinged AI-executive quote a day, past 181 installments and tens of millions of views, cataloging the Ellison bathroom-camera stuff and the Palantir “100 million fans” of it all.

The delicious part is the method. Patrick checks the people building the machine by using the exact viral, algorithm-native format the machine rewards. He’s fighting the surfaced-winner economy with its own weapon and winning. That isn’t a footnote. As far as I can tell, that’s what accountability looks like now. Not a hearing. Installment 181.

The Thing None of Them Can Model

In the second half, the people being modeled are done holding still. Young men who swung 15 points right in 2024 are already down about 10 on Trump, telling the Times both parties get it wrong, one cruel and one clinical. The apparatus overfit one election into a permanent law and the data walked off. Maybe, as one researcher put it, they’re just swingy and up for grabs.

Every model in this section, Google’s, Albany’s, both parties’, is a confident theory about people who refuse to behave. The governance gets written by whoever moves first. The generation it’s written about hasn’t decided a thing.

And on the week of the country’s 250th birthday, that’s either terrifying or the most hopeful thing I’ve got.

*AI Disclosure: 100% of this was written by me, a human with only light edits and typo corrections from Grammarly. The images are mostly AI-generated using Nano Banana.

Share The Four P's

Subscribe now

In truth, I’m writing this early Sunday morning on a red-eye across the Atlantic, because at my age, sleep on a plane is a rumor. My loss is your gain... minus the typos, which I’m blaming on the woman snoring next to me like a chainsaw cutting a daisy.

I’m excited, eager, and already exhausted... and yet, let’s start off with a bang!

I’ve already started… bumping into first time Cannes attendee and journalist extraordinaire Marty Swant during the boarding process. And then Toby Daniels on the plane before we took off. He told me all about an AI agent he built named Doma, and I spent the first hour of my flight chatting with it. (It gave me some great B2B marketing insights and strategies for Genuin!)

Most marketing teams are optimizing the thing they should be replacing.

That’s the tension I’m carrying into Cannes this year. Because there isn’t one Cannes anymore. There are 8 different concurrent themes and subplots running at 10 different hotels, 15 different beach clubs, 60 different villas, 100 different restaurants, and one Palais.

Even the official program has split itself into four tracks... not counting the 20,000 “creators” with iPhones who didn’t get the badge and don’t need it.

Then there are those chasing the same word, agentic, and almost nobody can prove it does anything yet.

Even if you’re not going, or don’t work in one of the MANY industries this touches, there’s something in here for you, I promise.

So this is the preview. Predictions, themes, the overhyped, the overlooked, and a field guide for surviving the week. Plus, as always, something VERY personal at the end.

THE PROFESSIONAL: The Panels Are Selling Agentic. The Receipts Say Otherwise.

Here’s my one prediction that won’t age badly.

Agentic will be the word of the week. And almost nobody saying it will have one in production.

Let’s remember, this is a Festival of Creativity. Ok, now that I’ve stipulated that, we can move on.

Because it’s going to be on every stage. Every beach. Every panel title that couldn’t think of a second idea. But the IAB is still building the rulebook for agentic advertising... which is a polite way of saying the thing doesn’t exist at scale yet. You can’t standardize what nobody’s running.

A new report from 3C Ventures put it cleaner than I can. Agentic is the highest-impact, lowest-evidence claim in marketing. A media agency can’t claim a programmatic win without an audit trail. But it can claim “agentic AI” with nothing. We extended good faith before the credibility was earned.

So the line I’d draw: Agentic commerce is real. Agentic execution mostly isn’t.

When a shopper lands on your site from ChatGPT, they convert higher and spend more (Adobe’s data, not mine). That’s value flowing in. But when a coding agent runs your token bill into the ground and your CFO reaches for the lawnmower? That’s the same word, opposite direction. One’s a customer. One’s a bill.

Don’t let the panel title flatten the difference.

THE AGENCY RECKONING

Every holding company built a platform. Omni, Open, Marcel, STAGE, all of them. Strip the branding and you can’t tell the decks apart. AI-powered, end-to-end, transparent, open. The language converged into noise.

And I’m curious if anyone on stage will say this, but 76% of agencies are very confident in their delivery. 39% of clients are very satisfied. That’s not a rounding error. That’s a 37-point delusion.

Yet the question every CMO should ask, the one that’s hardest to get answered: is this platform fee replacing headcount, or adding to it?

If your agency can’t answer that in writing, that’s your answer.

THE FLATTENING

AI raised the floor. Average is now indefensible. Slop was the word of 2025... and in 2026 it’s just the internet.

Which is good news, weirdly?

When everyone has the same tool producing the same median output, the only moat (another word you’ll hear a LOT this week) left is taste. And taste is the one thing the machine doesn’t have. It has training data. Those aren’t the same.

The market already knows this. 60% of consumers say “AI” in your brand messaging is a turnoff. Enthusiasm for AI-generated content fell from 60% in 2023 to 26% last year. New York now requires you to disclose synthetic performers in ads.

Read THAT trend line. Your customers want you to use AI and shut up about it.

THE PART THAT STINGS

The flattening is coming for the bottom of the pyramid first.

Stanford’s early data shows employment for 22-to-25-year-olds in AI-exposed roles shrinking while everything else grows. Junior developers. Customer service. The entry-level seats.

The trap... Taste comes from repetition. Reps come from junior jobs. Automate the junior jobs and you starve the pipeline that makes the senior talent everyone says is irreplaceable.

We’re eating our own seed corn and calling it efficiency.

WHAT ACTUALLY WINS

Streamer/Creator IShowSpeed has 55 million subscribers and out-reached every brand and broadcaster with a single World Cup track. His take on why: “People are still going to want that human connection we were born with.”

The 21-year-old streamer and the holding-company CTO agree on the one thing that matters.

Humans in the lead. Quality at the center.

Everything else is a panel title.

THE PRACTICAL: Your Field Guide for the Week

I’m not going to tell you what to pack because, well, it’s too late now!

So the real practical wisdom here is that you won’t see everything. Stop trying.

That’s the whole game. There’s no single Cannes, remember? Creators go one way, adtech another, big platforms yet another, and the brand marketers that people think are so valuable and worth chasing...? Yeah, they’re all hidden away in between their panels.

So the worst thing you can do is build a calendar so tight there’s no room for the moment that actually matters. Plan the anchors. Protect the gaps. The best thing that happens all week is usually the thing that wasn’t on the schedule.

But whether you’re hosting, hustling, or hiding, it’s good to keep some essentials with you at all times:

TREAT YOUR BODY KINDLY

• Easy to say as someone who doesn’t drink, but this is a marathon disguised as a sprint. Pace yourself.

• Do NOT break in new shoes. You’ll regret it by Tuesday. Sneakers by day, every day. Sneakers are acceptable at night too, France has made its peace with this.

• When the blisters hit, and they will, go straight to a French pharmacy. Compeed is your best friend. While you’re there, you can also find great products at discounted prices... because France... Bioderma, Nuxe, La Roche-Posay. Worth the suitcase space.

• Carry a small bag. Man or woman, doesn’t matter, the man-bag is thriving on the Croisette. In it: sunscreen, a phone battery or two, water, deodorant, band-aids, whatever gets you through 14 hours away from your room. Electrolytes if you’re smart. A mini fan if you’re a genius.

TREAT YOUR SCHEDULE KINDLY

• First-timers and 10-timers: Walk from the Palais to the Croisette the day you land. Get your bearings before you need them. Some things are the same, some are different every year.

• Keep your head up on the Croisette. It may be the only time all week you run into the person you’ve been meaning to see since Monday. The street is the meeting.

• If someone invites you to a boat, just make sure it’s not leaving. NEVER GET ON A BOAT THAT LEAVES. We’re there for 96 hours. Every one of them counts.

• And find a regular. Every good Cannes I’ve had ran through someone who already knew the rhythm of the place. Make them your guide. Buy them a drink.

WORLD CUP SCHEDULES

• Well this is new. For the first time in a while, the tournament and the festival are overlapping, and the time zones are not working in our favor. Matches landing late Cannes time means the watch party is the after-party. Find the bar with the screen. That communal 3am moment, no agenda, no badge, just a room full of strangers reacting to the same thing... that’s more “fandom” than any panel selling it upstairs.

YOUR ATTITUDE

Oh, two more tips here:

• Don’t bitch. There are so so many people who’d love to be here in a heartbeat. Some of them emailed me last week asking how to get in. One booked on Thursday (way to go, Marty!)

• And do make time for the creative. It’s why we’re all here, even if it’s not your agenda. You don’t need to be in the Palais to hear what people are seeing, making, and sharing. Treat the Creatives you see like the real celebs. It takes nothing.

Pace yourself. Drink water. Say yes.

And whatever you do... don’t go all out on your first night. It’s a marathon AND a sprint.

THE PERSONAL: What To Do With the Gap

So here’s the thing nobody warns you about.

It’s not the packed days that get you. It’s the gaps. Once or twice a day, somewhere between the thing you just left and the thing you’re walking toward, you end up with 30 free minutes. Sometimes 60.

Do you go back to your room? Grab food?

Or do you just... stand there. Phone out. Thumb hovering. Not sure whether to answer email, scroll, rest, or sprint across the Croisette to catch someone before they vanish into a villa.

I’ve started to notice how much that small moment unsettles me.

30 open minutes and my first instinct is to fill them. Optimize them. Turn them into something. Which is funny, because that’s the exact thing I spent the whole first section of this newsletter warning you about. We optimize the thing we should be protecting. I do it to my calendar. I do it to my inbox. And apparently I do it to the only unscheduled half hour I get all day.

The gap isn’t the problem. My relationship to the gap is the problem.

And I don’t think it’s just me.

Look at who’s hungry right now. The junior folks emailing me last week asking how to get into our Happy Hour. Or the kids who flew thousands of miles and camped out sixteen hours to audition for a streamer’s bootcamp IN CANNES.

Who’s definitely NOT here: the 22-year-olds watching the entry-level rung quietly disappear while everyone tells them AI is a gift. They’re not filling every minute because they’re anxious by nature. They’re filling every minute because the pipeline feels like it’s narrowing, and standing still feels like falling behind.

I get it. I was that hungry once. Some mornings I still am.

But here’s what I’ve actually learned, the thing I’d tell any of them if they cornered me in one of these gaps:

The gap is where the real stuff happens.

Not the panel. Not the keynote. Not the perfectly optimized calendar.

But the conversation you didn’t plan. The person you almost didn’t stop for. The idea that only showed up because you let yourself be bored for 90 seconds instead of reaching for your phone.

I fell in love with this industry, and with Cannes specifically, in the gaps.

The late drink with someone I’d just met. The late-night yacht singalong. The wrong turn that led to the right room. Storming Yahoo beach from a dinghy at 3am like it’s D-Day.

Dance-offs with Wyclef. Last-night dinners. Making friends for life.

The version of myself that felt, for a minute, like I belonged somewhere I had no business being.

None of those are in the agenda. This whole place is a second chance to feel like the most interesting version of yourself. The you that got picked. The you that got in.

And you don’t find that version of yourself in a back-to-back schedule.

You find it in the part you didn’t plan.

So this is what I’m ACTUALLY working on this week:

• Not the crazy, color-coded calendar jenga. That’s all set.

• Not FOMO. I’m too old for that.

• Not the inbox. That stuff can wait.

• Not seeing all 8 panels at once, which I’ve now made very clear is impossible.

I’m working on the gap. Letting it BE a gap. Letting it BE uncomfortable. Letting it stay empty long enough that something real can fall into it.

Which is, I realize, exactly what I’m doing right now. Crap, it’s 4am over the Atlantic. The woman next to me is STILL snoring FML.

I have no wifi, no calls, no calendar, nothing to optimize. Just a few hundred minutes I can’t fill with anything but this. Another gap.

Because those gaps are the part you’ll write back about. They always are. The personal stuff, never the predictions.

That’s the tell, isn’t it? The thing that connects is never the thing I optimized. It’s the thing I let happen.

See you in the gaps.

THE POLITICAL: The World Is Watching, and It’s Not Sure It Likes Us

People outside our industry always ask me why I fly 4000 miles to see people I can see in NYC.

But it’s not just New Yorkers, or even Americans. This is an international festival. The world is in the room.

And right now the world is rethinking who’s actually winning, at least in the AI race.

A new poll from Public First surveyed 18,000 people across 15 countries on who leads in AI.

In 11 of them, perception has tilted toward China over the US. France, the UK, Canada now see China as the frontrunner. Germany was the least convinced America is ahead. Americans still back themselves, 51% to 24%. Our closest allies have already changed their minds.

Sit with that. The country that built the dominant models is losing the story about them.

It gets stranger at home.

Even as Americans insist we’re winning, we’re souring on the prize. U.S. positive sentiment about AI dropped from 39% to 31% over the past two years. 70% oppose a data center going up near them. Pew has half the country more worried than excited, with the top fear being what AI does to creativity. Read that again. The thing people are most afraid of losing is the exact thing this whole festival exists to celebrate.

Meanwhile in China, fewer than 10% worry AI will take their job. Confidence is a story too.

Right now they’re telling a better one.

And the regulators aren’t waiting for the festival to end.

42 state attorneys general just subpoenaed OpenAI over its advertising practices, data handling, treatment of minors, and model sycophancy. That’s not a think piece. That’s a legal demand, days before the company’s expected public listing.

New York went further. As of last week, any ad running in the state using a synthetic performer, an AI-generated face you might mistake for a real person, has to say so.

Lead, background, even an AI hand model... $1000 for the first violation, $5K for the next.

New York moved first. Everyone expects more states to follow.

Notice what ties these together.

Consumers don’t fully trust AI.

Regulators are circling it. And the public’s loudest fear is that it flattens the human creativity they actually value. Same conclusion the whole industry keeps arriving at from every other direction.

The machine is not the moat (crap, I wrote it again).

Trust is. And trust is the one thing you can’t generate, can’t optimize, and can’t fake your way into.

There’s a version of this week that’s all hype, all agentic, all gold-foil future.

And there’s a quieter signal underneath it, from voters, from allies, from the law itself.

Use the tools. Earn the trust. And do not confuse showing up in the answer with anyone believing it.

See you all soon.

*AI Disclosure: 100% of this was written by me, a human with only light edits and typo corrections from Grammarly. The images are mostly AI-generated using Nano Banana.

Share The Four P's

Subscribe now

The Knicks are in the NBA Finals for the first time in 27 years. The opponents, the San Antonio Spurs, are the same team that beat them the last time, in 1999. I remember that year vividly. All of it. And during my entire tenure at the NBA, and ever since, they haven’t come this close.

So if you’ll allow those of us who’ve waited nearly three decades a moment to be insufferable about it, we’d appreciate it.

I know not everyone reading this is a New Yorker. I know some of you are actively rooting against us out of spite. That’s fair. Just let us have this.

And if a basketball team you don’t care about doesn’t move you, the World Cup kicks off this week, right here in our backyard. 39 days, three countries, the whole thing ending at MetLife (or whatever they legally have to call it) in July.

Based on the pre-tournament polling, your reaction to that news is somewhere around “meh.”

We’ll get to why that is. The short version: it’s struggling for some of the same reasons the Freedom 250 concert face-planted, and there’s a lesson in both.

THE PLAYFUL: Party on the World’s Stage

The World Cup is my favorite sporting event on earth, so consider the rest of this edition an attempt to will you into caring.

Which is also why, for the first time ever, I’m benching Political this round and subbing in “The Playful.” The moment calls for it. Political will be back. The beautiful game only comes around every four years.

Here’s the pitch, and I’m not going to bury it: Join my World Cup pool!

That’s it. That’s the ask. Five weeks of sustained rooting interest, a leaderboard to obsess over, and a reason to care about Curaçao vs. Germany at 2pm on a Tuesday. The tournament is going to take over North America whether you opt in or not. You might as well have skin in it.

Now let me sell you on the thing the pool is attached to.

The World Cup has a sentiment problem. Pre-tournament polling has been lukewarm. Big, top-down, institutionally hyped event, and the public shrugged. You can’t manufacture excitement at people. They show up for the thing, not the announcement of the thing.

But the audience is there if you know where to look. A new survey found at least 81% of Gen Zers across the UK, France, Germany, and Spain plan to follow the tournament. The interest isn’t dead. It’s just not where the institutions keep aiming.

The brands have noticed. And they’ve mostly placed one enormous bet.

Messi.

He’s starring in 18 of 80 World Cup ads tracked across the US, UK, and Argentina, per System1. Nearly 25%! Lowe’s built its entire campaign around 10-foot inflatable Messis. One man, in his last World Cup, has become the closest thing global advertising has to a sure thing.

Which raises the question nobody at those agencies wants to sit with:

Is this the last gasp of linear TV advertising for an event like this? I’ve already seen every one of these spots online. So, probably.

There’s a real insight buried in the fan data. People remember two things about a match: the goals, and what they were eating when the goals went in.

• Food and snacks (38%) and alcohol (33%) dominate ad recall.

• Auto (27%) and finance (21%) hang on, mostly on the strength of decades of sponsorship muscle.

• Telecom, personal care, and fashion (16% each) vanish into the digital pitch.

Fans watch in a high-emotion, low-attention state. The brands that win are the ones already woven into the ritual: what you eat, drink, wear, drive. Everything else is expensive background noise.

So if you’re a brand and you’re not a natural match-day staple, the move isn’t to force a generic sports spot. It’s to attach to the feeling. The subcultures, the lifestyle, the gut-punch highs and lows. That’s where the tournament actually lives.

One last thing the marketers got wrong. Creators are largely sitting this one out. The biggest live event on the planet, and there’s a creator gap. Which tells you the official machine is still building for a broadcast audience while the actual energy moved somewhere else years ago.

Same lesson as the Professional section, honestly. You can’t coerce enthusiasm. You earn it by meeting people in the ritual they already have.

So meet me in mine.

The pool’s right here. Don’t make me beg.

THE PROFESSIONAL: Remember When They Said AI Couldn’t Replace You?

Yeah, that aged well.

Sam Altman just walked it back. Again. In a recent interview the OpenAI CEO said an AI jobs apocalypse is unlikely and that he’d been “pretty wrong” about the near-term hit to the labor market. The timing is not subtle. OpenAI is reportedly prepping a Q4 2026 IPO at a roughly $1 trillion valuation, and “the thing we sell will gut your livelihood” is a hard line to put in an S-1. More here on the full reversal.

Meanwhile the foundation arm just committed $250 million to help the economy absorb the disruption the product is causing. Front-running your own crisis is a fascinating business model. Build the wave, fund the lifeguards, IPO the beach.

I’m not even mad. It’s smart. It’s just worth naming out loud.

Because here’s what’s actually happening on the ground, and it’s uglier than the press releases.

Workers are lying about AI.

A new GCheck study of 1,500 employees found 63% exaggerate their AI skills to look current. Among Gen Z it’s 80%. Forty percent talk confidently about AI in meetings to avoid looking behind. A quarter have taken full credit for work AI did. Sixteen percent flat-out lied about what they can do.

And then the twist.

That same workforce is undermining AI on purpose. Eighty-one percent admitted they discourage or limit its use at work. They’ll perform fluency in the meeting and refuse the tool at the desk. GCheck’s CEO called it “double distortion.” I’d call it self-defense.

Here’s the line that should be stapled to every leadership offsite: more than half the people faking their AI skills never received any training. That’s not a technology problem. It’s a learning-culture problem wearing a technology costume.

Which brings us to the part nobody wants to own.

You cannot coerce a transformation.

People don’t adopt things they’re being threatened with. And the threat is real, not imagined. ClickUp cut 22% of its staff citing a “radical embrace of AI.” Groupon dropped a quarter of its workforce to go “AI-native.” The class of 2026 is now likelier to be unemployed than the average American, and the most AI-exposed majors got hit hardest. The Economist’s chart said it better than I can: forget Python, study Plato.

So of course people are scared. Of course they’re hedging. When the same companies preaching curiosity are signing the layoff memos, you don’t get curiosity. You get resentment with a smile on top.

And resentment scales.

A WIRED report found DHS and the FBI are now tracking anti-technology extremism, with intelligence bureaus warning of civil unrest tied to AI job loss and data-center construction. 70% of Americans oppose data centers in their region. Among 14-to-29-year-olds, AI excitement has dropped 14 points, and many say they don’t want to use it but feel they have to.

That’s the whole problem in one sentence. They feel they have to.

So here’s the part I actually believe.

The companies that win the next few years won’t be the ones that deployed the most agents. They’ll be the ones whose people felt safe enough to say “I don’t know how to use this yet.” That sentence is worth more than any model. And right now almost nobody feels safe enough to say it.

Fix that first. The tools can wait.

Ultimately, this is a period of required, rapid learning. Open-mindedness is not optional. Neither is skepticism. Trust, but verify, on every output, every vendor, every promise.

THE PRACTICAL: How People Actually Learn

But you learn faster when you’re not afraid.

Which raises the obvious question almost nobody in this AI scramble is asking:

How do people actually learn?

I can’t remember where I first picked this up... maybe a leadership offsite, a management training, some workshop with bad coffee and a flip chart. But it stuck. Understanding how people process information might be the most basic skill in professional life, and somehow the most ignored. We’re getting bombarded from every direction, internal and external, and the whole game is sorting what matters, prioritizing it, and acting before the next wave hits.

So before you roll out another tool nobody asked for, maybe learn how learning works.

The simplest frame is the VAK model. Visual, Auditory, Kinesthetic. It’s not gospel, and learning scientists love to poke holes in it, but as a quick gut-check on why your training deck isn’t landing, it’s useful.

• Visual learners -- absorb best through seeing. Diagrams, charts, written notes, color-coding. They’re the ones building the mind map while you talk. If your AI rollout is a 40-minute monologue with no screen share, you’ve already lost them.

• Auditory learners -- take it in through hearing and talking it out. Lectures, discussions, podcasts, reading aloud. They learn in the conversation, not the doc. Put them in a room to argue about the new tool and they’ll retain more than any tutorial ever gave them.

• Kinesthetic learners -- need hands on it. They learn by doing, breaking, rebuilding. Sitting still through your enablement webinar is actual torture for them. Hand them the thing, let them poke at it, accept that they’ll learn by snapping it in half first.

Every company racing to deploy AI is, knowingly or not, running a giant learning program. And almost all of them are running it in one mode. A mandatory video. A slide deck. A Slack post that says “the future is here, please adopt by Friday.”

That’s a visual-auditory drive-by aimed at a workforce that learns three different ways and trusts you none of them.

The fix isn’t complicated. It’s just unglamorous.

Show people. Talk it through with them. Let them get their hands dirty and fail safely.

We spend millions on the tools and roughly nothing on teaching people how to hold them. Then we act surprised when the adoption numbers come back limp.

You don’t have a technology problem.

You have a teaching problem.

THE PERSONAL: A Summer Job, If You Can Find One

I keep thinking about something I read this week.

There are sleepaway camps now charging more than $8,000 a month that are quietly rebranding as career accelerators. At Camps Baco and Che-Na-Wah, where several of my friends both went and sent their kids. counselors hop on Zoom calls with career consultants for résumé prep. They leave mid-session to finish virtual internships.

One recruiter built a whole business line teaching 15-year-olds how to interview and work their camp network. “We look at it through the lens of career management,” she says.

About a kid. At sleepaway camp.

You can almost forgive the panic. The WSJ says we’re tracking toward the worst summer for teen employment since the government started counting in 1948. Counselor listings are down nearly 30%. New York’s youth program drew 200,000 applicants for 100,000 slots. An ice cream shop on Cape Cod got hundreds of applications back in January for fifty jobs.

The first job. The rite of passage. The thing that taught you to show up, clock in, and survive a bad manager. It’s evaporating, and the kids feel it.

What’s worse... the teenagers aren’t even outside to notice.

One writer went looking for them and found Central Park empty, kids averaging about nine hours a day on screens. She calls it “teen colony collapse,” New York kids dressing and talking like teenagers everywhere else, the slang reading less like a renaissance than the byproduct of algorithmic brain rot.

I don’t fully agree with her. But I’ll read Nancy Jo on the youth beat every time.

And it’s worse than empty parks.

There’s a new kind of bullying where kids photograph each other eating and post it. The ugly mouthful. The lonely eater. One high school senior in San Diego has been the subject of unflattering lunch photos thirty or forty times. He now scouts hidden corners of campus to eat in peace. When one Washington district banned phones all day, the number of kids eating lunch went up. They were finally eating because nobody could photograph them doing it.

You couldn’t pay me a million dollars to go back to middle school. Unfortunately, I have one kid who is finishing her first year in middle school, and another who just finished his first in high school.

We took the years that were supposed to be slack. The messy, unprofitable, formative ones. And we turned them into a pre-professional internship with worse lighting and a camera pointed at every sandwich.

The money doesn’t fix it. The $8K camp can’t buy a kid out of the anxiety. It just professionalizes it earlier.

Kids don’t need a LinkedIn at fifteen… They need a summer.

The unstructured, slightly boring, gloriously unmonitored kind. The kind where the worst thing that happens to you is sunburn and a crush who doesn’t text back.

That one’s not coming back on its own. But we could at least stop selling them the alternative.

*AI Disclosure: 100% of this was written by me, a human with only light edits and typo corrections from Grammarly. The images are mostly AI-generated using Nano Banana.

Share The Four P's

Subscribe now

There was no rock bottom story, no dramatic intervention, no twelve-step revelation. I just... never really liked it. Didn’t do it much. The decision was equal parts health choice and personality move, the thing that made me a little different at parties. I may now have one drink a year, or every other year. A glass of rosé in Cannes. Something to mark an anniversary. But that’s it.

Turns out I was early to a trend. Gallup reported last year that only 54% of American adults drink alcohol, the lowest share in nearly 90 years of polling, with GLP-1s, wellness culture, and Gen Z abstemiousness driving the decline. Cool, cool, cool.

The thing I liked about it was that it made me stand out at parties... the opposite of peer pressure, actually. But now it’s a thing so many others are doing. I’d be smug about it if I weren’t, you know, still terrible at exercise and great at stress. I wonder if it will be as noticeable next month in Cannes?

Yes, I occasionally miss an Old Fashioned after work on a Thursday. But I so rarely socialize IRL these days that there’s way more to miss than the bourbon.

Anyway. Onto the Ps.

THE PERSONAL: Beta Parenting in the Age of the Algorithm

I have another year or two before I even think about my own kids starting to drink alcohol... but there are too many other parenting mindfucks on my plate to deal with at the moment.

And mostly, it means coming to terms with my own mediocrity as a parent. As an overachiever in so many other facets of life, well, this one is decidedly an adjustment.

For example: We took exactly one photo at our daughter’s spring middle school band concert and it is, objectively, terrible. Blurry. Half a music stand. The back of someone’s head. The kind of photo that, in a different era, would have been the only photo, and we’d have loved it because it was all we had.

We are beta parents. We are doing our best. And sometimes, our best is a bad concert photo.

I think about this a lot lately, because the alternative version of parenting on offer right now is exhausting in a way I genuinely cannot afford. The version where every developmental milestone is a content opportunity. The version where I’m supposed to have a strong opinion about every app, every screen-time limit, every AI tool, every TikTok trend, every emerging platform my kids might or might not be on. The version where parenting is a project to be optimized rather than a relationship to be... in.

This summer, my son (turning 15 y/o) is set to start his first job as a caddy. Maybe. If he can get his act together. Meanwhile, a new DKC survey of 1,000 parents of 8- to 15-year-olds found that 95% of Gen Alphas are making money, averaging $52 a week, or $2,704 a year, up from $45 two years ago.

Parents said roughly half of household spending decisions come from their Gen Alpha child, 69% learn about brands from their kids, and 71% have changed their own consumer choices based on what their child told them. My kids are CMOs. I am the agency of record. This is fine.

Meanwhile, on the other end of the technology spectrum, cursive is making a comeback among young students who never learned it in school, with extracurricular clubs popping up in high schools and public libraries to teach the looping script Common Core dropped in 2010.

My daughter asked me last week on two separate occasions: “What is the Dewey Decimal System?” and “What is a VCR?”

So Gen Alpha is simultaneously running a small-business empire from the back seat of my car AND learning to sign their name like it’s 1952 but have a massive knowledge gap from all tech from roughly 1975 to 2015. Many things are true at the same time. Welcome to parenting in the multiverse.

And then there’s the AI stuff, which is where the actual stakes start to show.

The Honor System is dead. Princeton’s faculty voted to begin proctoring exams again after 133 years of trusting students to self-police, following a sharp rise in academic dishonesty cases and a senior survey in which 30% admitted to cheating and 28% said they’d used ChatGPT on assignments where it wasn’t allowed.

That’s the college version. The elementary school version is darker, or at least more uncomfortable. Since 2024, around 150 NYC schools have rolled out Amira, an AI reading app whose purple-haired avatar listens to kindergarteners read aloud and collects their voice data. Brooklyn dad Kevin Dugan said his daughter told him, “You know I love iPad math, but regular math is boring,” which is maybe the most chilling sentence I’ve read this month.

The best line came from a six-year-old. Dina, in a pink dress, walked up to the mic at a policy hearing, froze, hid her face in her dad’s sweater, then found her voice: “We look at screens all day, so it’s not actually good for our brains. So I think with indoor recess we could do something more fun. You could play with a friend, you could bring toys. Or you even could read.”

A kindergartener cooked the entire ed-tech industry in three sentences. Respect.

The “How do I talk to my kids about alcohol?” question feels almost quaint. The fight I’m in now is about cognitive ownership. Whether they’re using the tools or the tools are using them. Whether their attention is theirs. Whether they can sit with boredom long enough for an actual idea to show up.

And I don’t have tidy answers. I have a blurry concert photo and a kid who I think is going to be okay.

For now, that’s the work.

THE PROFESSIONAL: Publicis Isn’t Buying Data. It’s Building an Operating System.

I spent more than 10 years inside agencies as part of a holding company. I spent just as much time working with agencies and holding companies as a partner. So I’ve seen this movie from both sides of the table, and I have opinions. But I waited as long as I could for the dust to settle...

The hot takes on the Publicis-LiveRamp acquisition have subsided, so now it’s time for the smartest one. Last week the debate was about two things: whether LiveRamp can maintain neutrality now that it lives inside a holding company, and whether $2.2 billion for $900 million in revenue was a steal or a stretch. (For the record: it was a steal.)

But that’s not the story. The story is much simpler, and much bigger.

This is an AI play. Full stop.

It’s the move that takes Publicis from “holding company” to “full-stack tech and service leader,” and it’s the clearest signal yet that the holding company model, as we’ve known it for 30 years, is being quietly euthanized from the inside by the smartest player in it.

Look at the acquisition sequence:

• 2019: Epsilon, $4.4 billion. Deep identity and transactional profiling, bought ahead of third-party cookie deprecation while everyone else was still arguing about whether deprecation was real.

• 2025: Lotame. Global data connectivity across Europe and APAC, scaling the audience graph to 4 billion profiles.

• 2026: LiveRamp. The dominant enterprise data collaboration network, locking down interoperability across 25,000 publisher domains.

That’s not three deals. That’s one infrastructure build, executed over seven years, with the patience of a company that knows exactly what it’s doing.

The shift isn’t from cookie to cookieless. That framing is already outdated. The shift is from human-browsed to agent-orchestrated.

AI agents don’t bid on impressions in the way humans designed programmatic to work. They execute transactions directly with publisher and retail media APIs, bypassing the open auction entirely. To do that, they need real-time, deterministic first-party data. Not stale third-party signals. Not modeled lookalikes. The real thing.

Publicis is building the proprietary environment to train and fuel those agents. Every future media allocation, across display, video, search, retail, will either run on their infrastructure or it won’t run at all on theirs. And if you’re a brand inside the Publicis ecosystem, you’re going to have a hard time arguing it should run anywhere else once the agentic solution starts outperforming everything else by 30%.

This is where the neutrality question gets interesting, and also kind of beside the point.

Neutrality stops mattering when the agentic engine is just plainly better than what any other holdco can offer. The decisions get made by AI plugged into “PubliciOS” (or whatever they end up calling it), and humans become advisors at best, spectators at worst.

You don’t worry about whether your accountant is neutral when the accountant is an algorithm that’s already executed the trade.

For brands and advertisers, I think this clarifies a lot: When an agency owns the identity data, pre-purchases the inventory, and builds the agents allocating your budget, they’re no longer a fiduciary partner. They’re a walled garden with a financial interest in their own walls.

And more critically, your first-party data is training models that will eventually serve your competitors. Contractual protection is going to feel like a much weaker shield than it does today. Every campaign you run generates a data trail, including bid signals, behavioral responses, conversion patterns, and that exhaust flows back into the model whether your 1PD is technically ringfenced or not.

There’s no clean exit. The leverage isn’t in the contract. It’s in knowing this is happening and negotiating from there.

Now put ALL of that aside and realize that brands now kinda have nowhere else to go. That’s chess, not checkers. And the rest of the holdcos are still playing checkers.

Supermetrics surveyed 400+ marketing leaders recently and the results tell the rest of the story. Only 6% have AI fully embedded in their workflow. 45% struggle with data and measurement. 39% lack a clear AI vision from leadership. The problem isn’t access to AI tools. It’s the data foundation underneath them.

That’s exactly what Publicis just bought, at scale, before anyone else figured out it was the thing worth buying.

So for everyone else in the industry... it’s what I’ve been saying for the better part of the last year: The layoffs and RIFs we’ve already seen are the warm-up act. California is preparing for this, which should tell you something about the size of what’s coming.

Governor Newsom just issued an executive order this week directing the state to prepare workers and businesses for the “economic disruption” AI is poised to have on the workforce. “This moment demands that we reimagine the entire system,” Newsom said, “how we work, how we govern, how we prepare people for the future.”

The order calls for AI literacy programs, worker ownership models, employment stability payments, and an “AI playbook” for modernizing job training. California has always led on tech regulation, and the federal government has been doing what it does best, which is nothing.

Newsom’s order might not be enough. It may simply be triage. But the fact that a governor is publicly preparing the largest state economy in the country for AI-driven workforce displacement, while the holding companies are publicly buying identity infrastructure, tells you everything you need to know about where this is heading. If only the rest of our country had that kind of leadership and foresight instead of prevaricating threats and excuses to bomb countries.

Back to the war that’s already been decided: the next 10 years of marketing aren’t going to be won by the agency with the best creative shop or the smartest strategist. They’re going to be won by whoever owns the data layer, the identity graph, and the agentic engine sitting on top of both.

Publicis just bought the third leg of the stool. The rest of the holdcos are still trying to figure out what stool they’re sitting on.

THE PRACTICAL: The Tomato Plant Dilemma

I don’t sleep very well.

Part of it is physical (my wife says I’d sleep better if I dropped 10 lbs, but I’m not so sure). Call it restless brain, call it anxiety, call it whatever the wellness people are calling it this month.

The actual diagnosis is simpler. I’m an overthinker.

The latest entry in the catalog of things keeping me up: whether to keep my unplanted saplings and seeded vegetable buds indoors last weekend with the cold and rain rolling in over Memorial Day, or whether to leave them outside to “harden.” Life-altering decisions, literally. For them, not me.
(Although given the way things are going, I may not be far off from having to live off my own vegetable garden, in which case it kind of is about me.)

Anyway. I lay there. I thought about it. I checked the forecast three times. I read a Reddit thread. I considered emailing a guy at our local nursery. At some point around 1 a.m., I realized I was treating a question about tomato seedlings the same way I’d treat a board-level decision at work, and that the problem wasn’t the seedlings.

The problem was the loop.

If you’re an overthinker, you know the loop. It looks like analysis, but it isn’t. Analysis ends. The loop doesn’t. The loop is what happens when your brain confuses thinking about a decision with making one, and rewards itself for staying busy. It feels like diligence. It’s actually a stall.

The part I keep relearning, slowly, and forgetting just as fast: Most overthinking isn’t really about the decision in front of you. It’s about three things underneath it.

1. Control. We overthink because we believe that if we just think hard enough, we can engineer a guaranteed outcome. We can’t. The future doesn’t run on spreadsheets. The seedlings either survive the cold snap or they don’t, and no amount of 1 a.m. weather-checking changes the actual temperature.

2. Certainty. We treat certainty as the prerequisite for action, when it’s almost never available for the decisions that actually matter. Every meaningful choice is made with incomplete information. The overthinker keeps waiting for the data to get good enough. It doesn’t.

3. Failure. This one’s the real one. Underneath most loops is the quiet belief that if you make the wrong call, it will mean something about you. That you’re not smart enough, careful enough, prepared enough. So you keep deliberating, because deliberation feels safer than the moment of commitment where you become responsible for the outcome.

The trick (and I use “trick” loosely, because none of this is actually tricky, just hard) is to notice that the loop is happening and call it what it is. Not “I’m being thorough.” Not “I just want to get it right.” It’s: I’m afraid of being wrong, and I’m using thinking as a substitute for choosing.

The way out, as far as I can tell, is small. You practice on the seedlings. You practice on what to order for dinner. You practice on the email you’ve been drafting for three days. You make the call, you note that you survived, you do it again tomorrow.

Each time, you teach yourself that you don’t need a perfect map to take the next step. The capacity to trust your own judgment is built rep by rep, in the low-stakes moments, so it’s there when the high-stakes moments show up.

For the record, I left the seedlings outside. They’re fine. I’m fine. I still didn’t sleep.

THE POLITICAL: Bread and Circuses, Now With a Cover Charge

Unlike the weather here locally at the start of summer, the midterm election noise and political ads will heat up over the coming weeks. Get your remotes ready.

Memorial Day Weekend was expensive this year. Gas prices are the highest they’ve been in years. Ground beef is up 24%. We passed on the BBQ and actually went to a restaurant, instead.

The Romans had a phrase for keeping the people content: bread and circuses. Keep them fed, keep them entertained, and they won’t revolt. Somehow I don’t think $10 gas and a UFC fight on the White House lawn is going to cut it.

Summer is unofficially underway, a weekend when we should have all been thinking way more about pool day than pay day. But the national mood is giving less “hot girl summer” and more “broke boy summer.”

The data backs up the vibe. Voters report worsening conditions across most aspects of the economy, with sharp increases in spending on gas. An overwhelming majority have consistently reported paying more for groceries and food. Recent jumps at the pump have brought gas roughly even with food as the top complaint.

Voters also say they’re cutting back on household goods, groceries, and gas. But some categories are inelastic. Utilities, housing, and healthcare are all up, and people show very little flexibility to trade down. There’s no generic-brand version of keeping your lights on or paying your landlord.

This is the political backdrop heading into the midterms, and it’s the backdrop heading into the thing I actually want to talk about, which is the World Cup.

We’re just a few weeks away from the opening games here in the United States. And outside of die-hard fans and the marketing, advertising, and media ecosystem desperately hoping their investments pay off, the volume of excitement has been barely audible.

A large part of that is economic distraction. When the cookout costs 24% more, you’re not springing for a $400 group-stage ticket to watch a team representing a country whose population hates us, and ours has never heard of.

But it’s not just the economy. There’s a political weight on this tournament that wasn’t there a year ago, and I keep coming back to it.

When FIFA decided to go all-in and effectively crown Trump with a “peace prize,” it killed something for me. It made me no longer want to be part of this commercial enterprise in any way that goes beyond watching the games.

The lack of hope in U.S. Men’s Soccer doesn’t help. The expansion to 48 teams, which sounds inclusive on paper but in practice dilutes the early rounds, doesn’t help. Spreading games across Canada, Mexico, and the U.S. is exciting, but all the American group-stage matches on the West Coast doesn’t help either. If you live on the East Coast (or if you’re headed to Cannes), this is a tournament you watch at 10pm ET or 4am CST with one eye open.

And the part that bothers me most: this was supposed to be a once-in-a-lifetime moment for American soccer. The first World Cup on American soil in 32 years. The sport finally getting its mainstream moment alongside MLS growth, Messi at Inter Miami, the women’s game thriving, a generation of kids who grew up actually watching the sport instead of being told it was boring.

We had everything we needed. Infrastructure, audience, timing, talent on the rise. And we managed to suck the air out of the room before the ball was even kicked.

The FIFA executives, the host committees, and the brands betting big are going to learn the same lesson Memorial Day weekend just taught a lot of American families. You can throw the party. That doesn’t mean people can afford to show up. And it doesn’t mean they’re in the mood.

Bread and circuses only works when the bread is affordable, and the circus isn’t owned by the guy you can’t stand.

Some of that’s economic. Some of that’s FIFA. Some of that’s the broader fatigue of living in a country where every major institution feels like it’s been corrupted by proximity to power. But all of it is political, even when it’s dressed up as commercial.

I’ll still watch. I love the sport too much not to. I’ll still text my friends about the upset nobody saw coming and the goal that broke my heart. But the sponsor activations, the host-city storytelling, the marketing campaigns asking me to feel something patriotic about it... no thanks.

Let’s just hope the sport, and its future, are bigger than the people who tried to ruin it.

*AI Disclosure: 100% of this was written by me, a human with only light edits and typo corrections from Grammarly. The images are mostly AI-generated using Nano Banana.

Share The Four P's

Subscribe now

The timing was almost suspicious. I was 48 hours away from my 30th high school reunion.

It was the culmination of a long week of late nights, including another Springsteen show, an F1 race, a Battle of the Bands, and three separate moments where I caught myself asking the same question.

Who was that guy? And who am I now?

Four essays, four answers. Four Ps.

I was the kid who chased the loudest rooms. The kid who memorized box scores. The kid who jumped on every new platform first. The kid who never got the chicken pox and figured the government had it covered.

All four of them grew up. Some of them I miss. Some of them I don’t.

THE PERSONAL: Same Wiring, Different Volume

30 years since high school. Yikes, that math is rude.

Of course, I’ve kept the closest friends close. A few others stay in light rotation. But high school wasn’t really about the friendships I curated... it was about the cauldron. Eight hours a day, more if you counted sports and extracurriculars and whatever else I’d talked my way into.

More time than I spent with my own family. The closest modern equivalent is your work family, which doesn’t quite hit the same anymore in a hybrid world. Back then, you couldn’t Slack-react to someone. You had to actually deal with them.

Back then, I loved being a creator, or at least a fast chaser, of the excitement. First, best, different. Pick a lane, I’d take all three.

This past Saturday night, back in our hometown with a few hundred older, more experienced, even grayer and balder versions of those same characters we’d roam the halls with... it felt both familiar and surreal in equal measure. Old conversations slipped back on like a varsity jacket.

So did some of the old personas, which was less flattering. People had matured in expected and unexpected ways. Some turned out exactly as you’d have bet. Some didn’t. A few surprises in both directions.

But I wasn’t really thinking about them on the drive home. I was thinking about me.

I’m still an overachiever. Still hard-working. Still juggling more than is reasonable. But I have less time for bullshit, less interest in impressing the room, and a much more honest definition of what success and happiness actually are. Heavier than I expected. Also reassuring. I made good decisions then. I think I’m making good ones now. That’s what the cauldron really taught me. Not facts. Judgment.

The version of me that came out of that building wanted to be everywhere at once, and for a long stretch, he was. Post-college in New York, I was out five, sometimes six nights a week. Friends, dates, coworkers, sporting events, bars, concerts. Around 9/11, I went to roughly 40 Springsteen shows, which is how Bruce became permanent firmware in my operating system. Then bachelor parties, weddings, kids, jobs became careers, homes became hubs.

So that guy from 30 years ago doesn’t really exist anymore. We’re older. Calmer. Busier. Work takes more. Family takes more. Years now stretch between Springsteen shows where weeks used to.

Except this past week.

This past week was a flashback to a tempo I haven’t kept in years. The reunion. Another Bruce show on Long Island. F1 and work events in Miami. Then back home and a late night out in NYC to support my sister’s annual Battle of the Bands benefit, where the same musicians she’s been playing with for 35 years still absolutely bring it. Old friends, family friends, the whole soundtrack. It felt great. It also occasionally felt like a lot.

Which brings me to the part I usually don’t lead with.

One year ago this week, I was diagnosed with Sudden Onset Hearing Loss. Not gradual. Not genetic. It happened in real time, at an event, from high-decibel exposure. Steroids, inner ear injections, a window of hope. What I have now is partial hearing in my right ear, an unreliable sense of balance, and a new set of small choreographies. Seat yourself on the correct side of the conversation. Read the room before the room gets loud.

It’s why I had earplugs with me all week long… and everywhere I’ve been since last May.

No, I’m definitely not that same kid anymore. He didn’t care about hearing loss and thought he was invincible. And then the kid who chased the loudest rooms in the building got exactly what he wanted, and then some.

So when I ask what’s changed since high school, that’s the real answer. The ambition is the same. The judgment is sharper. The hearing is worse. I still want to be in the room. I just sit on a different side of it now.

THE PRACTICAL - From Box Scores to Bird Calls

I used to scan the box scores at every meal. Every league, every stat line, didn’t matter if I had a stake in the team. Information was the entertainment. The newspaper sports section was a meditation practice with better fonts.

I still stay up for the Knicks and Mets. Sometimes. But the obsessive ingestion of every data point? Gone. The kid who used to memorize batting averages now squints at standings once a week and calls it a day.

Same with gadgets. I still love them. I still get excited by genuinely new things. But the pull of “newest” for its own sake has faded. What gets my attention now has shifted in a direction I would not have predicted at 17, or 27, or honestly even 37.

Reader, I am being slowly converted into a bird guy.

A few weeks ago in Miami, working from a balcony, I had two small, quiet, beautiful coworkers. I don’t know what kind of birds they were. I do know they were patient, well-mannered, and clearly there for the toast crumbs. We had an understanding.

This is becoming a pattern.

Ever since Father’s Day last year, and then my birthday, I’ve been getting absolutely worked over by the Bird Buddy ad targeting team. Bird Buddy is a smart bird feeder with an AI camera that identifies not just species but individual birds. You can name them. Track when they come back. I find this completely fascinating, which is itself new information about me.

I want one. I have a price in my head. I am holding the line. They keep sending discounts. We are in a slow-motion negotiation where they don’t know they’re negotiating.

Apparently I’m not alone. Birdwatching is having a moment among young people as an antidote to screen fatigue, aided by apps like Merlin Bird ID. Google Trends shows “birdwatching” searches climbing for five straight years. A TikTok captioned “so cooked I started birdwatching in my 20s” pulled over 60,000 views. One singer-songwriter quoted in the piece said birding “rescued” her. Another creator said it “regulates my nervous system in a way nothing else can.”

So it’s not just me. There’s a whole generation arriving early to a club I didn’t know I was already standing in line for.

Have I become boring? Or just easily persuadable? Or did the things that used to spike my attention finally stop landing... and something quieter took their place?

I’m pretty sure the birds know.

THE PROFESSIONAL: The Cost of Going First

I used to be the guy who jumped on every new platform the day it dropped. From Twitter to Foursquare, TripIt to Snapchat. First. Loudly.

Sometimes I’d just sign up, poke around, break it, write up what I learned, move on. No questions. No hesitation. The cost of being wrong about a new tool was a wasted afternoon. The cost of being late was looking like everyone else.

Occasionally I’d get hooked, and stick with it. And that instinct served me well for a long time. But, in case you haven’t noticed, times have changed.

Now the wisdom-based part of my brain kicks in before the curiosity-based part does. Not because I’m less curious. Because the stakes have moved.

AI is the most transformative technology movement of our century. Industrial revolution and electricity, rolled into one, on fast-forward.

But the part that gets lost in the breathless coverage is that electricity, by itself was just a hum in a wire. It needed appliances. It needed a kitchen, a factory, a record player. The miracle wasn’t the current. It was what we plugged into it.

We’re at the “lots of current, not enough appliances” stage with AI right now. Which is exciting, and also why so many seasoned operators are quietly losing their minds.

A new global study from Harris Poll and Dataiku surveyed 900 CEOs. The findings are something.

• 81% fear a botched AI deployment could cost them their job this year.

• 62% say their boards are actively pressuring them to deliver measurable AI-driven outcomes.

• 75% believe a fellow CEO will get ousted over a bungled rollout.

• 56% admit their competitors have stronger AI strategies than they do.

• And 79% are worried about the legal risk of actually using the stuff.

You can see what’s happening here. Executives with proven playbooks and Jack Welch biographies stacked behind their desks have been turned into stammering, stumbling puddles. The thing they were trained to do, which is project certainty, no longer maps to the thing they’re being asked to do, which is bet the company on a technology that ships breaking changes faster than they can update the slide deck.

The phrase doing the heavy lifting here is “capability overhang.” The models can do incredible things. Deploying them inside a real business with real customers and real legal exposure is a completely different sport.

Dataiku’s CEO calls it the cognitive dissonance happening in the C-suite right now, which is a generous way of saying everyone is trying to look composed in a meeting they don’t fully understand.

I’d add a quieter version of the same problem: the pace doesn’t just rattle CEOs. It rattles anyone whose professional identity was built on knowing things. The old playbook rewarded being first. The new playbook rewards being right about what to be first about. Those are different muscles.

One CEO at the HumanX event put it as well as anyone: slow down to speed up.

The kid who jumped on every platform first would have rolled his eyes at that. The grown-up version is starting to think it might be the whole job.

The risk of getting it wrong: brand irrelevance.

THE POLITICAL: Unflappable, Until I Wasn’t

I was never a germophobe growing up. My invincibility shield was thick, physical and mental. Heck, I was the only kid I knew who never even got the chicken pox. Whatever was going around, I assumed it wasn’t going around me.

Two things changed that. I married a doctor. And then COVID happened.

The doctor part rewires you slowly. You start hearing the casual professional vocabulary at the dinner table. You learn what gets washed, what gets thrown out, what you should actually be worried about and what’s just vibes. The germophobe in me was a slow build, which made it feel reasonable.

COVID was the other kind of rewiring. The fast kind.

I don’t blame the Trump 45 administration for the virus, itself. Viruses don’t care about elections. But the response... the wishful timelines, the bleach moment, the daily contradictions from a podium that used to be a place where adults said calm and useful things... that’s the part that shifted something in me that hasn’t really shifted back.

It wasn’t just COVID. It was watching the institutions I had quietly assumed would handle it not actually handle it. The unflappable faith I had in “the government will figure this out” turned out to be a vibe of my own.

So I read the headline this week with a particular kind of dread.

Hantavirus on a cruise ship. Several dead. The Andes strain, which is the only hantavirus that transmits human-to-human and kills 30-40% of the people it infects. Confirmed cases tracked into California, Georgia, Arizona, Virginia, Texas. Passengers scattered before quarantine. A coordinated whole-of-government response, per the CDC’s website, which would be more reassuring if “whole-of-government” still meant what it used to mean.

Maybe it’s nothing. Hopefully it’s nothing. The experts in the piece are clear that hantavirus is unlikely to be the next pandemic. It doesn’t spread like COVID or flu. A bad American outbreak would mean dozens of deaths, not millions.

But that’s not actually what’s scaring me. What’s keeping me up is the infrastructure around the next one.

• The U.S. formally left the WHO in January.

• The CDC has been without a Senate-confirmed head for eight months.

• Testing for dozens of pathogens has been paused.

• The moronic HHS Secretary spent weeks during a measles outbreak recommending cod liver oil before he could bring himself to endorse the vaccine.

• mRNA research funding got cut by half a billion dollars.

• Oh, and the World Cup is this summer.

The kid who never got the chicken pox would tell me to relax.

I don’t think I’m going to.

*AI Disclosure: 100% of this was written by me, a human with only light edits and typo corrections from Grammarly. The images are mostly AI-generated using Nano Banana.

Share The Four P's

Subscribe now

Hotter, even, than a gunsight staring down an oil tanker in Strait of Hormuz… the elephant-in-the-room situation that nobody wanted to talk about on a mainstage.

POSSIBLE 2026 was a conference about AI stacks, creator economies, commerce media, and whether agencies still matter. It was also, quietly, a conference about deferred decisions... ours, our industry’s, and the ones being made for us while we hesitate.

This week’s Four Ps, the POSSIBLE recap edition, goes through what I actually saw, what I didn’t hear, what’s about to hit the supply chain whether we planned for it or not, and why I’m staying in Miami an extra few days anyway.

THE PROFESSIONAL: Agents Everywhere. But Most of the Room Was Just Watching.

There were more attendees than ever at POSSIBLE this year. But for all the industry leaders who came and went, I think there were 10x more agents. At least, if you listened to the speakers from the stages.

I counted approximately... a lot. By day two I couldn’t tell you what half of them did. Every booth and every adtech company is talking agents. Every platform launched one. Every enterprise suite rolled out a “catalog of reusable agent skills” and a “coworker agent” and “open interoperability with AWS, Anthropic, IBM, OpenAI and more.”

Five pitches, panels, and presentations in, the stories start sounding similar.

Which is, weirdly, the point. The convergence is the story. Vendors claim proprietary technology in the same breath that they claim open interoperability, and those two claims are in tension.

Marketers should notice. The moat is thinner than the pitch suggests.

The connective tissue everyone is quietly standardizing on... MCP, agent-to-agent protocols, the plumbing layer... almost nobody said out loud from a mainstage.

Anthropic was the name-drop of the week. This should have been the Claude Conference. A clean reversal from last year’s OpenAI default. But the people saying “Claude” were mostly not talking about Claude Code. The tool got famous. The protocol did the work.

That gap is the real professional story of POSSIBLE 2026.

My friend Dan Lewis, who wasn’t even there, gave me the cleanest frame for the week even before I arrived: There are two kinds of people right now. Those using AI, and those building with AI.

Using is fine. Sometimes it’s the smart move. But it isn’t the same as building, and confusing the two is how you end up with a six-figure AI line item and nothing to show for it on the earnings call.

Walking around and getting 20k steps in at POSSIBLE over the past three days, all between only two adjacent hotels, my honest guess: maybe 4% of the people here are actually building. The other 96% are still optimizing their prompts.

The builders ask what workflow wouldn’t exist without AI. The users ask what AI can do for the workflow they already have. Those are different conversations, and they happen at different tables.

Don McGuire at Qualcomm gave the session version of this. Most companies are layering AI onto old workflows. Qualcomm is rebuilding marketing around it.

Vas Bakopoulos gave the diagnostic version Wednesday morning: draft everywhere, ship nowhere. Every team uses AI. Almost none let it ship.

The blocker isn’t the technology. It’s the org.

Which brings me to the best thing I read before the conference, written by someone who was also at the conference: Darren Herman at Bain Capital.

• In his newsletter a week before, he talked about how the most common failure mode at senior levels isn’t bad decision-making. It’s deferred decision-making. One more analysis. One more stakeholder. One more pilot.

• Colin Powell once famously said to make the call with between 40 and 70% of the information. Jeff Bezos said most decisions are two-way doors. Most of the AI decisions facing marketers right now are two-way doors being treated like one-way doors, which is the most expensive middle position you can occupy.

• Herman’s reiterated that a 60% call now beats a 95% call six months from now, every time. That’s the connective tissue for everything else the conference was fumbling toward.

The transformation is harder than the implementation, but implementation keeps getting sold as transformation. The agent results shared from stage were almost entirely internal, and the workforce question nobody wanted to raise came up in every hallway. The agency leaders warning clients not to trust opaque models are running opaque models on their clients’ behalf.

All of these are decisions somebody is declining to make.

Horizon’s Bob Lord made one of the quieter points of the week:

“Vibecoding with Claude is lowering the barrier to building software for anyone with an idea.” That’s a bigger shift than most of the stack sessions combined. It’s the on-ramp to the builder side of Dan’s line.

(Meanwhile, I don’t think anyone from OpenAI or Anthropic were at POSSIBLE, and if they were, it was very below the radar.)

Eiter way, the tools are ready. The infrastructure is maturing. The vendor landscape will look different by POSSIBLE 2027. Half the agents will be acquired, pivoted, or gone.

The question this year wasn’t whether AI can do the job...

It’s whether you’ll decide what job you want it to do, before somebody decides for you.

THE PRACTICAL: The Problem Isn’t the Tool. It’s Your Brief.

Ironically, just about everyone at POSSIBLE was ALSO complaining about AI output at some point this week.

Too generic. Too bland. Too corporate. Sounds like a chatbot, not a brand.

If I had a glass of wine for every time someone said the word “slop,” I wouldn’t be coherent enough to write this post.

But what I kept reminding people, when I had the chance: The tool is not the problem. Your brief is.

Brand voice guides used to be written for humans who could read between the lines. “Friendly and conversational” meant something to a writer who had onboarded, sat in enough meetings, and absorbed the culture.

To Claude, to ChatGPT, to whatever agent you’re pointing at your homepage, it means nothing. It reads the line. Not between it.

The fix isn’t a better model. It’s a better map.

Kaleigh Moore has been writing about this for a while, and the framework she lays out is the cheapest high-leverage move any marketing team can make in 2026: an AI-native brand voice guide. More explicit than your human one. More example-heavy. Paired comparison paragraphs labeled on-brand, too formal, too generic. Hard “never do this” rules. Never use “leverage.” Never open with “So.” Never drop a generic inspirational quote. A machine doesn’t know these things unless you tell it.

Then upload it. Then actually tell the tool to use it. Attaching without instructing underperforms, which is a sentence I wish I didn’t have to write.

The same logic applies one layer up. Another common theme here at POSSIBLE was also talking about AEO and GEO. Every agency had it on their agenda. Almost none had a case study older than a quarter.

The tactics are real. HubSpot’s six strategies for SaaS are a fine starting point. Optimize for evaluation-stage questions, not just problem awareness. Get serious about third-party validation. Structure content so an AI can actually extract it. Implement schema.

But “best practices” is not a phrase that applies to something eighteen months old. “First practices” is the honest version. If your AEO partner has a “system,” ask what they tested last month that failed. If they can’t tell you, they don’t have one. They have a pitch.

Same skepticism with the agent vendors. There were more agent demos at POSSIBLE than there were people to watch them. Half of those companies will be acquired, pivoted, or gone by next year’s conference. “Open and interoperable” is what vendors say when they can’t lock you in. That’s actually good news for buyers, if you can hear it.

3 things you can do this week that will outperform anything you’ll buy:

1. Rewrite your brand voice guide for AI. Make it explicit. Add the “never do this” list. Upload it everywhere your team is prompting.

2. Audit your AEO pitch. Ask your agency what they’ve tested that didn’t work. The answer tells you whether you’re hiring a partner or funding their R&D.

3. Don’t marry the first agent vendor who promises transformation. Rent before you buy. Integrate before you commit. Half the landscape is about to change.

The tools are fine. The tools have been fine for a while.

It’s the instructions that need work.

THE POLITICAL: Strait Shooters

We just spent the past three days talking about Commerce Media. Meanwhile, the crisis in Iran, and more specifically the Strait of Hormuz, was the quieter refrain running through every hallway conversation.

At POSSIBLE, nobody really wanted it in the foreground.

But while a few thousand marketers were at the Fontainebleau and Eden Roc hotels on Miami Beach, talking about agent stacks, retail media consolidation, and whether creators are the new agencies, the Strait of Hormuz remains a shitshow.

Effectively closed down since late February, when the US and Israel launched an air war against Iran and assassinated its supreme leader.

20% of the world’s seaborne oil trade passes through there. 20% of global LNG. All held up at the world’s most critical energy chokepoint. Now a dual blockade, with the US Navy blockading Iranian ports since mid-April and Iran blockading the strait itself. No bueno.

Oil prices are a boring topic for marketers, but the reckoning is coming. Gas prices are up more than a third since the start of the war. Crude oil surged past $120 per barrel and forced QatarEnergy to declare force majeure on all exports. Goldman raised its December 2026 inflation forecast by a full percentage point. The IMF’s pessimistic scenario has global growth falling to 2%, something that has happened four times since 1980.

That’s the macro you are planning your 2026 media budget inside of, in case you’re wondering why tensions are high and long-term views are murky.

The editorial consensus is not subtle. The Center for American Progress called it “the most significant supply chain shock since COVID-19.” Helium, sulfur, tungsten, the ingredients that go into semiconductors and fertilizer and munitions, all caught in the blockade.

The Denver Gazette put it plainest: Trump misjudged how Iran would strike back economically, the same way he misjudged China in the trade war. He started a war of choice based on what he said were imminent security threats, and Iran demonstrated it could exact economic costs his team had underestimated. His approval on the economy dropped to 30%. His approval on Iran is 32%. Even his own supporters, the Tucker Carlson and Megyn Kelly end of the coalition, have been publicly breaking with him.

None of this made it onto a POSSIBLE stage. Fair enough. The conference is about marketing.

But every CFO conversation happening in every portfolio company right now is about the new inflation floor, the rerouted supply chains, and how long a 35% jump in energy costs takes to show up in consumer discretionary spending. Put that in your 2027 pitch deck.

The wars we start without thinking through the consequences don’t stay in the war. They show up in the media mix, the supply chain, the shelf price, the consumer confidence index, and the quarterly earnings call.

Marketing doesn’t operate in a vacuum. It just keeps getting planned like it does.

THE PERSONAL: Paddock Attack

Let’s pivot. Or more accurately, take the inside racing line at Turn 18...

From “marketing doesn’t operate in a vacuum” to “neither do Formula 1 races...” and the reason I’m still in Miami after POSSIBLE ended last night.

This weekend is the Miami Grand Prix. My first live F1 race. And after a month-long hiatus caused by the war in the Middle East, which resulted in two cancelled races, the return to the track is exciting on a lot of levels.

I spent yesterday with some of the team leaders. Today too. Whatever you think of Formula 1 as a product, from a business perspective they are crushing it and doing just about everything right. The Netflix effect turned into a real flywheel. US fanbase tripled in five years. New tracks, new audiences, new merch lines, new sponsors.

The operating model they’ve built inside a sport that used to be a rich man’s hobby is, honestly, one of the most interesting case studies in sports media right now. Maybe I’ll write about it separately. Maybe I won’t. I’m on vacation-adjacent time.

Personally though, this weekend isn’t about the business story.

It’s a bonding moment with my son, who is flying down from New York to meet me here in Miami tomorrow. Fan Fest. Sprint race. Qualifying. The whole shebang. Maybe our team or FIA friends hook us up with a paddock tour. Maybe they don’t. Either way, he’s going to lose his mind, and I’m going to pretend I’m not losing mine.

The whole weekend isn’t going to be cheap. That’s not the point.

He’s finishing his freshman year of high school. Which means three more summers. Four more Memorial Day weekends before he’s technically an adult. Every parent of a teenager knows the math, and the math is brutal. The window isn’t closing. It’s already half closed.

So these are the experiences we prioritize. The memories we actually make. Not the ones we keep meaning to.

If POSSIBLE was about deciding fast, with 60% of the information, and Political was about the consequences of deciding poorly, Personal is the one where the decision is easy.

You buy the tickets. You get on the plane. You put your kid in the grandstand and watch him watch the cars. You figure the rest out later.

Note to self: sunscreen. Reapply. Repeat.

*AI Disclosure: 100% of this was written by me, a human with only light edits and typo corrections from Grammarly. The images are mostly AI-generated using Nano Banana.

Share The Four P's

Subscribe now

And look... it’s genuinely impressive. The precision, the safety, the fact that we can do this at all. I’m not immune to the romance of space travel.

But we did this already, more than 50 years ago. Multiple times. On computers less powerful than my Apple Watch.

So forgive me for not updating my screensaver.

I’ve been thinking a lot lately about the gap between what we celebrate and what actually moves us forward. We’re in this weird cultural moment where the familiar gets repackaged as revolutionary... and the actually revolutionary gets questioned, resisted, and revolted against.

AI is somehow both overhyped and underestimated in the same news cycle. And somewhere in the middle of all that noise, real things are being built.

THE PERSONAL: I Hate Texting. And I Think That Makes Me a Hypocrite

Sitting next to my kids on the flight back from vacation last week, I had a full epiphany. Not a quiet, reflective one. An irritated one.

They were texting me. On our family group chat. While sitting literally four seats in a row. Together. As a family. On an airplane.

I’m 47 years old and I’ve never felt more like my mother in my life.

I’ve said, with full confidence and genuinely zero self-awareness, that being a bad texter is a character flaw.

Everyone has their phone within arm’s reach 24 hours a day. “I’m bad at responding” stopped being a legitimate excuse sometime around 2013. What it actually means is that you’ve quietly decided your time is more valuable than the other person’s and found a socially acceptable way to say so without consequences. Own it. Don’t rebrand it as a personality quirk.

I believe this. I have delivered this take at dinner tables. I have nodded aggressively when other people said it first. I have, on at least one occasion, been slightly smug about it.

And yet. I barely text anyone first anymore.

In my defense, which I fully acknowledge sounds like exactly the kind of rationalization I just mocked, I am clinically, almost impressively bad at typing on a phone. Not slow. Not distracted. Actually bad. Like, “autocorrect has filed a formal complaint and requested reassignment“ bad. What I intend to write and what arrives on the other end are reliably two entirely different documents.

I’ve sent texts that read like a stroke in progress. I once tried to type “on my way” and sent something that looked like a legal filing from a jurisdiction that doesn’t exist.

I find the whole thing unnatural. I’ve gotten so good at typing with all ten fingers on a real keyboard that texting is like eating fried rice with chopsticks when there’s a perfectly good fork and spoon right there. It’s not a skill issue. It’s a principles issue.

That’s my story and I’m keeping it.

So I avoid it entirely. The experience feels like trying to thread a needle while wearing a baseball glove, the needle keeps moving, someone is asking me a question, and I have now somehow misspelled my own name. Twice. FML.

The result is that from the outside, I look exactly like the person I spent four paragraphs criticizing. Radio silence. Vague availability. The occasional response that arrives so late and so out of context that it reads as unhinged. Which, to be fair, is sometimes also accurate.

But here’s the distinction I choose to believe matters: intent. I am not leaving you on read because I am ranking you. I am leaving you on read because the medium is genuinely hostile to how my brain operates. I am a talker. A caller. Give me a real conversation, a voice note, even a well-structured email and I am fully present and embarrassingly enthusiastic. Put me on a glass rectangle with a keyboard clearly designed for fingers I do not have, and I become a liability to everyone involved.

The problem is that distinction is completely invisible to the person waiting on a response.

To them, silence is silence.

Of course, maybe I’ve been arguing the wrong point entirely.

Maybe the real take here isn’t that bad texters have a character flaw. Maybe it’s that texting was always a deeply stupid communication format that we collectively agreed to normalize because it felt efficient, and now we’re all trapped in it, performing responsiveness on a platform that rewards brevity and punishes nuance.

I didn’t ruin texting.

Texting ruined texting.

I’m just the only one willing to say it out loud. Probably via voice dictation.

THE PROFESSIONAL: Built Different to Build Different.

There’s a moment in every industry when the old way stops working and the new way hasn’t been named yet. We’re in that moment. Right now.

For 20+ years, digital marketing was essentially one big negotiation about attention. You found the eyeballs. You rented space near them. You hoped something stuck. The metrics got more sophisticated. The targeting got more precise. The creative got... fine, mostly.

But the underlying logic never changed: interrupt people enough times and something converts.

AI didn’t fix that. It just made it faster. More content, more targeting, more interruption. At scale.

That’s not progress. That’s more of the same problem arriving sooner.

What I’ve been thinking about a lot at Genuin (and what’s actually starting to feel like a turning point) is that the conversation is finally shifting from reach... to relationship. From impressions to participation. From rented audiences to owned, trusted communities. These aren’t new ideas. But they’re becoming unavoidable ones.

The fragmentation problem is still getting worse, and that may continue for awhile until it gets better. Brands are running 10 different tools, none of them talking to each other, none of them connected to actual business outcomes. They’re optimizing for the dashboard, not the consumer.

Meanwhile the consumer has gotten very, very good at ignoring them.

Video changes the dynamic. Not because it’s flashier. Because it’s closer to how people actually experience things. When content is short, immersive, and placed where someone already is and already trusts... the intent is different.

They’re not being pushed. They’re leaning in. That’s a fundamentally different transaction.

What we’re building at Genuin is the infrastructure layer that makes that transaction possible at scale, across owned properties, across a network, and increasingly now, with AI doing the heavy lifting on personalization, curation, and yes, content generation.

We’re hearing a LOT of noise right now from adtech platforms, agencies, in-house product teams, AI consultancies... and it’s hard to know where to start. The answer is: You don’t need to know where to start, because the AI can start for you. That’s a fundamental shift in understanding AND execution.

The exciting part is that AI genuinely unlocks things that weren’t possible two years ago.

The challenging part is that we’re defining a new category while the category itself keeps changing shape.

That part is not for the faint of heart. Seriously.

Because you can’t just build a product anymore. You have to build the case for why the product needs to exist. You’re not selling into a known need. You’re helping people understand they have a need they haven’t articulated yet.

Engagement and revenue infrastructure for the modern internet... that’s now a real thing. It just doesn’t have a line item in most media plans yet.

That’s the work. And it’s equally exhilarating and exhausting.

The opportunity is real. The timing is right. The AI capabilities we’re plugging in right now... some of them are genuinely going to change what brand-consumer relationships look like. Not hypothetically. Actually.

More on that soon.

But the north star stays simple: content without trust is just noise. And the brands that figure that out first... aren’t going to win by making more content. They’re going to win by making people feel like they belong somewhere.

That’s what we’re building toward.

THE PRACTICAL: But... The Revolt is Real. And I Kind of Get It

Something’s happening here. And for what it’s worth...

This is not a trend. Not a news cycle. An actual, measurable, generational pushback against the thing that everyone in my industry has spent the last two years insisting will change everything.

AI is losing the room. Or at least some smaller rooms.

A Quinnipiac poll of 1,400 American adults found that 55% think AI will do more harm than good.
76% trust it “hardly ever” or “only some of the time.”
A separate Gallup survey found that young people ages 14 to 29 are souring fast... excitement down 14 points year over year, hope down 9, anger and anxiety both climbing.
The kids who grew up with smartphones in their hands, who never knew a world without the internet, are looking at AI and feeling... not inspired.

They’re worried, angry, feeling trapped.

And before the techno-optimists in the back start muttering about Luddites, let’s be honest about why.

Companies are using AI as cover for layoffs that probably would have happened anyway, and everyone can smell it. “AI wash headcount reductions” is a phrase an actual KPMG executive said out loud and on record, so at least someone’s being honest about the game being played.

When 60% of enterprises announce they’ll let go of employees who can’t or won’t use AI, you haven’t positioned the technology as an opportunity. You’ve announced it as a threat. That’s not an onboarding strategy. That’s a hostage situation.

Then there’s the data chaos, which isn’t helping anyone.

Every week brings a new study with a new conclusion. AI is automating everything. AI is actually increasing workloads. AI is destroying software jobs. Software jobs are up from last year.

Nobody agrees. Nobody has the full picture.

And when people can’t make sense of something that feels like it affects their livelihood, confusion doesn’t stay neutral for long. It curdles into fear. Fear hardens into resistance.

That’s just human. That’s always been human.

Then there’s the idea that I find genuinely fascinating: the IKEA effect.

Studies find that writers who completed assignments without ChatGPT reported greater satisfaction with their work. The effort, it turns out, is part of the point. You don’t get the same feeling from assembling a table someone else built. (Hi, this is Matt talking... really, yes, it’s me, Matt! Tpyos and all.)

There’s a whole revenge-of-the-analog thing happening underneath all of this too. Repair cafés. Film photography. Handwritten notes. People actively seeking friction because frictionless stopped feeling like it meant anything. I understand this impulse more than I probably should admit.

The resistance makes sense. The fear makes sense. What doesn’t make sense is the prevailing narrative being pushed by the companies actually selling this technology, which boils down to: adopt it or be left behind, replaced, made irrelevant. That’s a genuinely terrible way to build trust and an even worse way to drive adoption. Scaring people into using something doesn’t make them good at it. It makes them resentful, minimal, and quietly waiting for it to fail.

The smarter play is what Tubi is doing... using AI to make experiences better for actual humans, personalized recommendations, smarter ad matching, tools that help creators earn more, without making AI the headline. The story is the experience. AI is just how you get there.

Not existential threat. Not magic. Infrastructure.

The revolt is real. The feelings are valid. And the companies that make people feel like participants instead of casualties are going to win.

The ones leading with fear are going to deserve exactly what they get.

THE POLITICAL: The Contract Won’t Write Itself

Last week I published something on Medium that had been rattling around in my head and building for awhile, but took a catalyst to unlock.

It started with Rousseau. It ended with OpenAI. And the distance between those two things is exactly the problem.

Quick 10th grade AP European History refresher: the Social Contract. Legitimate authority comes from the consent of the governed. People voluntarily give up some natural freedoms in exchange for the protection and stability of a shared society. The general will governs.

Not kings. Not gods. Not, notably, corporations.

It was radical in the 1700s. It’s somehow radical again now and what’s happening with AI.

OpenAI, an $852 billion company that converted from nonprofit to for-profit last year (and whose president has donated millions to the current president of the United States) recently released a policy paper outlining its vision for the AI economy.

Robot taxes. Public wealth funds. Four-day workweek. Portable benefits.

Some of it is genuinely interesting.

All of it is deeply troubling.

That’s not a social contract. That’s a press release dressed up as philosophy. And OpenAI isn’t alone. Anthropic published its own policy blueprint months earlier. Tech billionaires fund super PACs while simultaneously publishing op-eds about shared prosperity. Everyone wants credit for thinking about the common good.

Fewer people want to actually be accountable to it.

The history of industry self-regulation is not exactly inspiring. The tobacco industry self-regulated. The financial industry self-regulated. Social media self-regulated. We know how those stories end.

Rousseau’s insight wasn’t just about what the contract said. It was about who had the right to write it. The answer was everyone. Not the aristocracy. Not the church. Not whoever currently held the most power and the most to gain from writing it favorably.

By that standard, we don’t have a new social contract yet.

We have drafts from interested parties.

The Enlightenment produced thinkers willing to challenge dominant powers at real personal risk, in service of a better organizing principle for society. That energy is exactly what this moment requires. Not from the companies that stand to gain the most from shaping the outcome. From the rest of us.

What it actually needs now is genuinely diverse coalition, not just technologists and investors and the governments they lobby. It needs to center the people most likely to be displaced, not the companies most likely to benefit. It needs to treat AI infrastructure as a public good. And it needs accountability with actual teeth, not aspirational language about “safeguards” and “oversight bodies.”

The Age of Intelligence is here whether we’re ready or not.

The question is whether the contract that governs it gets written for everyone... or just for the people who already own the infrastructure.

Rousseau didn’t ask the king to write the social contract.

We shouldn’t ask OpenAI either.

*AI Disclosure: 100% of this was written by me, a human with only light edits and typo corrections from Grammarly. The images are mostly AI-generated using Nano Banana.

Share The Four P's

Subscribe now

Or maybe it’s not a joke. It’s hard to tell sometimes.

We’re standing in TSA lines longer than some international flights.

We’re a country at war with no exit ramp.

We’re watching AI eat entire job categories for breakfast.

And somewhere in Florida, a guy with $19,000 and an Andrew Tate endorsement is wowing voters and packing venues.

This week’s Four Ps is about preparedness. Who has it? Who doesn’t?
And what happens when the systems we’ve built to give us certainty start failing us all at once.

Spoiler: the answer isn’t panic. It’s posture.

THE PERSONAL: Preparing for the Unprepared: TSA, Tots, and Touchdowns

At the time of writing this (late Monday night), I’m just 48 hours away from one of the most difficult decisions I’ve made in a long time.

What time do I get to the airport on Thursday morning for a 6am flight?

Thursday morning begins our first real family vacation in two years. And the TSApocalypse, now entering its fourth week of nationwide airport chaos, has decided this is the moment to become my problem.

Normally, it’s not. I’ve cracked the code. TSA Precheck. TouchlessID. I’ve turned airport security into a sport I’m actually good at.

In fact, even over the past few weeks, while the rest of the traveling public was stress-sweating through hour-long lines, I was breezing through the TSA with a 0 second wait... like I had a FastPass at Disney.

But tomorrow we’re flying on an airline I don’t normally use, and more painfully, one that hasn’t rolled out TouchlessID yet. We’re booked on the earliest flight out. Whole family in tow. The variables are stacked.

It’s like a terrible April Fools’ joke.

Yet the thing that will actually bother me isn’t the waiting... it’s the not knowing. I can handle almost anything if I can prepare for it. The uncertainty is where I lose my mind a little.

I didn’t need this added anxiety. But we DO need this trip badly. Things have been busy, to say the least. Not in a bad way, just in a busy way. So much so that I’m even skipping coaching flag football this spring. That’s saying something.

Coaching is one of those things that gives back more than it takes, and every season I walk away more convinced that youth sports is quietly one of the most fascinating institutions in America. Every kid on every team wearing a jersey sponsored by a local realtor, a chiropractor, a pizza place. Adorable. Also: a missed opportunity at scale.

Which has me thinking... if a 30-second Super Bowl spot runs $8.5 million, and the average youth sports sponsorship is $350... that budget could fund 24,000 teams. Imagine a major brand (seriously, any major brand) announcing in 2027 that they’re skipping the Super Bowl and sponsoring 24,000 youth sports teams instead.

The earned media alone would dwarf the ROI of the ad buy. The goodwill would be unquantifiable. Someone is going to do this eventually and look like a genius. It should be on purpose, not by accident.

And it’s not just sports. Last Saturday I was in a packed college gymnasium alongside thousands of people, hundreds of students from dozens of schools, watching the regional qualifiers for a national Robotics tournament. My son, a freshman, was one of two first-year kids on his team. They put up the best showing his school has had in over a decade. Proud dad alert!

I don’t have a sharp marketing take for that one. I just wanted to say it out loud... a moment of pure, uncomplicated pride I’m going to be living off for the next several days.

Including however many hours we end up standing in a TSA line.

THE POLITICAL: Progress Held Hostage by People Playing Games

Let’s just take stock of where we are for a second.

Really this entire debacle could have been avoided, but politics is getting in the way of progress, and the gamesmanship is absolutely crushing us at the moment. Not that we should be surprised. This is a country that just started a war (in Iran) without provocation and has no idea how to get out of it.

Not a proxy conflict. Not a skirmish. A war. One that started without provocation, that nobody seems to have a plan to exit, and that is already doing things to the global economy that should be keeping more people up at night than they apparently are.

The tech industry, which has long operated in its own reality distortion field, is now getting a very rude introduction to geopolitics, too.

Iran’s official military news agency has named Amazon, Microsoft, Palantir, and Oracle as “the enemy’s technological infrastructure.” Targets. They already attacked AWS buildings in the UAE. They took out the Qatar helium plant at Ras Laffan, which, if you didn’t know, supplies roughly a third of the world’s helium... and helium is essential for cooling the systems that produce semiconductors... and semiconductors run, well, everything.

Especially AI.

The Strait of Hormuz closure is putting the entire semiconductor supply chain at risk. Some estimates suggest we’re a week or two away from the conflict bringing chip production to a halt. The cascading effects of that, across every sector of the global economy, are almost too big to fully model.

We’ve spent years talking about AI as the next great technological revolution. It turns out revolutions are a lot harder to run when the minerals required to build them are caught in a war zone.

Meanwhile, the hyperscalers that were racing to plant flags across the Middle East, billions of dollars in data centers, AI factories, a rumored 5-gigawatt facility in the UAE involving OpenAI, Nvidia, Oracle, and Cisco, are now quietly rethinking whether Southeast Asia or Northern Europe might be a safer bet. Hmmm, you think?

But let’s zoom out from the chips and the cloud for a second, because the political picture gets even more... let’s call it “vibrant.”

In Florida, a 31-year-old candidate named James Fishback is running for governor on an ultranationalist, anti-Israel platform and packing venues with Gen Z men he recruits almost entirely through Instagram and TikTok. He’s raised $19,000. His opponent has $45 million. He’s polling at 6%. And yet the crowds keep coming. He’s got endorsements from Andrew Tate and glowing praise from Nick Fuentes, which really does tell you everything you need to know about the coalition he’s assembling.

Toxic maleness is a problem that is not getting better.

A 29-country Ipsos survey of 23,000 people found that Gen Z men are twice as likely as Baby Boomer men to believe a wife should obey her husband. 33% of Gen Z males think the husband should have the final word on major decisions. 21% think men who do childcare are less masculine. The lead researcher pointed to economic anxiety as the driver, the idea being that when traditional breadwinning paths close off, some men retreat into rigid gender frameworks as a substitute scoreboard.

It’s a theory worth taking seriously, but it’s not just the bros.

In Paris, a 32-year-old far-right mayoral candidate named Sarah Knafo, running on slashing the city budget, flooding the streets with armed police, and auctioning off public housing, is doing something remarkably similar with remarkably similar energy. Two different countries. Same playbook. Same demographic.

Back home here, Trump is continues to sink us even deeper into his hellscape of hate by targeting trans kids while abandoning Americans’ real needs around affordability at the pump and with groceries.

It’s also worth noting that “economic anxiety” explains the origin of a thing without excusing where it leads.

So. A war we didn’t plan for. A chip shortage that could trigger a global recession. Young far-right candidates going viral. And a generation of men re-embracing the 1950s as an aesthetic.

What is this world coming to?

Asking for a newsletter.

THE PROFESSIONAL: Protect Your Brain Before AI Hollows It Out

So I just bought tickets and planned my first Formula1 race weekend in Miami next month.

I can’t wait. While I’ll be in Miami for professional reasons that week, having my son fly down with a friend experience this together for the weekend is something I honestly am giddy about.

And yes, I’ll be bringing noise-cancelling ear buds.

In the meantime, let’s talk about who’s actually winning the AI race. And no, it’s not who you think.

While everyone’s been busy benchmarking models and debating which LLM writes the cleanest code, a quiet but meaningful shift is happening in how smart marketers are thinking about AI strategy. The question is no longer “are you using AI?” It’s “What is AI making you worse at?”

New research has given a name to something a lot of us have been feeling: “the exoskeleton effect.”

The idea is this: AI strengthens your output in the moment, but if you’re letting it do the thinking for you, it’s quietly hollowing out the skills underneath. The people who performed best in the study weren’t the ones who used AI the least. They were the ones who stayed in the driver’s seat cognitively. They double-checked outputs independently. They drilled into results with follow-up questions. They treated AI like a sparring partner, not a answer machine.

That distinction matters enormously for anyone building a marketing practice, a brand, or a business right now.

For me, I will NEVER give up my ownership over writing. I write everything myself -- emails, blog posts, this newsletter. But I do feed certain things into Claude for grammar and validation.

But where I turn to Claude most is now for ideas on slide layouts for Powerpoint and Google Slides. And doing it has made my independently less imaginative in the creativity in layouts. Which I’m fine with.

But the same holds true for brands. The ones that are going to win aren’t the ones who outsource their community strategy to a bot. They’re the ones using AI to scale genuine human connection, not replace it. User-generated content now drives 35% of global influencer campaigns and converts 29% better than polished brand content.

The audience can smell automation. They always could.

What they’re responding to is realness, and realness can be orchestrated and amplified with the right infrastructure, but it can’t be fabricated. Strategy is, and always must be, a human-led endeavor.

Which brings me to a stat that should be taped to every CMO’s monitor right now:

42% of marketers plan to increase investment in brand building and upper funnel in 2026. Not performance. Not bottom funnel. Brand. Mental availability. The thing you can’t buy with a retargeting campaign (via StackedMarketer’s The State of Advertising 2026 Report).

You can’t outbid everyone forever. But you can out-position them.

The brands doing that work, the ones investing in distinctive creative, building owned audiences, and using AI as a force multiplier rather than a replacement for judgment, are the ones building something that actually compounds. The exoskeleton keeps you moving. But you still need the muscle underneath.

Don’t forget to use it.

THE PRACTICAL: Preparing Our Kids for Jobs That Don’t Exist Yet

My daughter and her friends are at that age (almost 12) where they talk about what they’re going to be when they grow up all the time.

In a car ride with three of them over the weekend, I learned:

• Emily wants to be a famous plastic surgeon/photographer.

• Addison wants to be a Dallas Cowboys cheerleader and youth cheerleader coach.

• My Sydney wants to open a karaoke coffee shop. Clearly the apple doesn’t fall far...

None of those jobs seem likely to be replaced by AI. At least not in any way my brain can fathom.

But let’s have an honest conversation without the kids in the room overhearing what our kids are actually going to do for a living.

A recent Harvard survey found 59% of 18-to-29-year-olds see AI as a threat to their job prospects. Stanford research backs it up: employment among 22-to-25-year-olds in AI-exposed roles has already fallen 16%.

One 22-year-old dropped out of a computer science program to become an electrician. A 28-year-old left insurance underwriting to become a firefighter because, as he put it, “people are always going to want that empathy from an actual human.” A former Microsoft product manager built a career guide for his 17-year-old daughter analyzing which jobs can’t be replicated by AI. It’s been viewed over 5,500 times.

So here’s a framework I keep coming back to (and it applies whether your kid is 8 or 18. (But maybe less so if they’re 48. Sorry, dad.)

AI can automate any job that can be fully documented and executed on a computer. So the question worth asking is simple: What can’t be documented? What can’t be scripted?

Two things rise to the top every time.

1. Distribution. AI can generate an infinite amount of content, code, products, and presentations. Human attention is still finite. The people and brands that own an audience, a network, or a direct channel to people who trust them? That’s not going anywhere. If anything, it becomes more valuable as everything else gets commoditized. Teaching your kids to build in public, develop a voice, cultivate real relationships? Make a killer coffee and curate an even more killer karaoke set? That’s not soft skills. That’s infrastructure.

2. Judgment. When AI knows how to do almost anything, knowing what to do becomes the premium skill. Silicon Valley calls it “taste.” Whatever we call it, it comes down to the ability to predict and deliver what people actually want, before they know they want it. That’s not something you learn from a prompt. It’s something you develop by being curious, by being wrong, by being in rooms where hard decisions get made.

My son isn’t enjoying robotics, drones, and avionics because an algorithm is telling him to. He enjoyed it because he worked with a team, navigated pressure, made real-time calls, and cared about the outcome in a way that wasn’t performative.

That’s the career advice I’m banking on. Not a specific major. Not a specific industry. The ability to think, connect, and lead in rooms where the stakes are real.

AI will handle the rest. Make sure they can handle what AI can’t.

*AI Disclosure: 100% of this was written by me, a human with only light edits and typo corrections from Grammarly. The images are mostly AI-generated using Nano Banana.

Share The Four P's

Subscribe now

And somehow, that’s the most important marketing insight of 2026.

Not a trend. Not a cycle. A structural shift in how people read, shop, scroll, and decide.

The signals are everywhere if you’re paying attention: in the content that flops, the platforms people are abandoning, the AI tools rewriting how we work, and the political fights breaking out in real time over who gets to control all of it.

Trust is gone, and everyone is scrambling to figure out what comes next.

This one’s got a lot going on. Stay with it.

THE PERSONAL: Punctuation... and What It Says About Me (And Yours About You)

I’ve been thinking a lot about punctuation lately. Which, if you knew me five years ago, would probably make you laugh. Because five years ago, I was the guy who ended every email with three exclamation points.

Every Slack message felt incomplete without at least one. I was enthusiastic! I was engaged! I was apparently terrifying!

Then someone called me out. Not cruelly, but clearly. And I started paying attention.

Turns out there’s actual science behind this. Research suggests exclamation points can make you appear meaningfully warmer and more positive, but they come with a trade-off: a modest but real dip in how analytical or authoritative you’re perceived to be. We associate high emotion with lower assertiveness. Not fair, but also... not wrong?

So I dialed back. I read the room. I started saving the exclamation points for moments that actually earned them.

And then AI showed up. And everything got weird.

AI didn’t just change how we write. It changed how we think about how we write. Because now, every time I put a finger on a stylistic quirk, I have to ask myself, “Is this me? Or is this what a language model would generate if it was trying to sound like a friendly professional?”

Like... I used to love em dashes -- used them constantly. Then I started noticing AI used them constantly. So I stopped. Almost entirely. If you see one in my writing now, it’s deliberate. It earned its place.

I’ve apparently overcompensated with ellipses though... they’ve become my new emotional support punctuation... which probably says something about me I’m not ready to unpack...

That’s the bizarre feedback loop we’re all living in. AI learned to write like us. Then we noticed. Then we started writing differently to prove we’re still us. The machines didn’t just reflect our habits, they made us self-conscious about them. Which is either the most human outcome imaginable, or the most unsettling one.

Punctuation has always been a signal.

Warmth. Authority. Confidence. Chaos.

But in 2026, it’s also become a kind of fingerprint. Proof of personhood.

Or not. Meanwhile, our kids don’t know how to use it. At all. they text and voice dictate and punctuation has gone from essential to optional to ignored.

As for me, yes, I use exclamation points more carefully now. But when I use one, I mean it.

And that’s going to have to be enough!

THE PROFESSIONAL: You Can’t Buy Your Way Out of a Trust Problem

Let’s be honest about where we are.

Social media is broken. Not technically. Not functionally. Broken in the way that actually matters:

People don’t believe what they see on it anymore. Nearly half of US consumers flat-out distrust the information on social platforms. Most of the rest are somewhere in the “eh, maybe” zone. TikTok, Facebook, X... none of them are scoring above a 21% trust rating right now.

21%

That’s not a platform problem. That’s a signal about the entire environment.

The Paradox Nobody Wants to Say Out Loud

The thing that’s messed up now... The better your content looks, the more suspicious people are of it. We’ve reached a point where polish is a red flag. That’s what fueled my turn away from em dashes referenced earlier.

But same with images: Producing a flawless image, a perfectly structured article, a slick 30-second video costs nothing today. And people know it. So when they see it, their first instinct isn’t “wow, great brand.” It’s “bot.”

We’ve entered what I’d call the Zero-Trust Web. And if your marketing strategy was built around scale, automation, and aesthetics, you’re going to feel this one.

78% of people say it’s getting incredibly hard to tell what’s real from what’s synthetic. 3/4 of Americans trust the internet less than they ever have. Less. Than. Ever.

That’s not a trend you optimize around. That’s a structural shift.

The Wrong Place Problem

It’s not just what you say. It’s where you say it.

Research is unambiguous on this: Brands that show up next to unsafe, toxic, or objectionable content don’t just waste money. They take real, measurable hits to trust and brand opinion. Almost 20% drops. And the most underappreciated part? People who never even saw the bad placement, just heard about it, showed nearly the same decline.

The internet has a long memory and a short fuse.

The cheap CPM isn’t cheap. It’s deferred debt. And the interest compounds fast.

So What Do You Actually Do?

You stop outsourcing your brand’s reputation to an algorithm and start being intentional about the environments where your brand lives.

That means showing up in spaces where trust is already present. Publisher environments. Brand-owned destinations. Community-driven platforms where real people are having real conversations and the content around your message has been vetted, curated, and earned.

It also means rethinking what “content” even means right now.

The most valuable thing you can produce in 2026 isn’t polished. It’s verifiable. Real customer voices. Unscripted formats. Actual results, not projected ones. The brands winning today are the ones leaning into radical transparency instead of theatrical perfection.

And then there’s the engagement layer. This is where I’ll put a gentle plug in for the work we’re doing at Genuin, because it’s genuinely relevant here. We’re building infrastructure for brand-owned destinations, places where brands control the context, the community, and the content environment.

Not a feed you rent space on. Not an algorithm that decides your fate every six months. Owned. Moderated. Trusted. That’s not just a product pitch. That’s a response to exactly what the data above is screaming.

Trust me...

We’re now competing on two fronts simultaneously: convincing real humans and satisfying AI decision engines. Neither one rewards inauthenticity anymore.

The winners aren’t going to be the ones with the biggest budgets or the most automated pipelines. They’re going to be the ones who figured out that trust isn’t a campaign. It’s an infrastructure decision.

Build accordingly.

THE PRACTICAL: How I Actually Use AI (2026 Edition)

Let me tell you what my AI stack actually looks like right now, because I think it’s more useful than another think piece about what AI “could” do.

First, I use Granola for meeting transcription and notes, including personal stuff now, not just work calls.

I use Claude CoWork for professional orchestration, the kind of behind-the-scenes task management that used to eat hours.

I also lean on Claude and/or ChatGPT to proofread and simplify my emails.

I use it for video edits like Opus and Genuin to clip, caption, categorizr and curate for my TWO podcasts.

And for image generation, I’m bouncing between Gemini (Nano Banana) and MidJourney and a few others depending on the output I need.

That’s it. That’s the stack. No magic. No AGI. Just tools that removed friction from things I was already doing.

But when considering ALL of the options out there, whether as an end user, a professional team builder, or an AI-led product builder, I stick to the following framework:

Assist. Audit. Automate. Autonomize.

Those four words cover almost every legitimate AI use case I’ve seen work in practice. Most people are still in the first two stages. The third and fourth are where it gets interesting, and honestly, where most organizations aren’t ready to go yet.

Now zoom out.

ChatGPT just crossed 900 million weekly active users. That’s not a product anymore. That’s societal. And OpenAI is already running ads inside responses, with retail and grocery brands leading the way. That’s not a test. That’s a business model arriving in real time.

Meanwhile, Walmart’s AI shopping assistant is driving 35% higher order values for users who engage with it. Not because it’s magic, but because it understands intent. It builds full baskets around moments. A birthday party. A dinner. A game day. It connects digital engagement to same-day physical fulfillment. That’s not artificial intelligence. That’s applied relevance at scale.

Shoppers want virtual try-on. They want AI assistants to remove guesswork. They want automatic reordering for the things they already buy. The demand is there. The sensory gap of online shopping is real, and AI is closing it.

Amazon, on the other hand, just spooked its investors by pledging $200 billion in AI infrastructure this year, pushing free cash flow negative. Wall Street blinked. The stock dropped 12% in February.

The lesson isn’t that AI investment is wrong. It’s that the market is starting to distinguish between AI that works and AI that’s a press release.

The ones who figure out the former are going to win. The ones still chasing the latter are going to have a rough couple of earnings calls.

Use it intentionally. Know why you’re using it.

The era of “we’re exploring AI” as a strategy is over.

THE POLITICAL: I Use Claude. And I Have a Take.

As I mentioned above: I use Claude every day. Premium. Highest payment and feature tier.

It’s in my workflow, my writing process, my professional life.

So when the Pentagon designated Anthropic a “supply chain risk to national security,” I had thoughts.

Here’s mine, plainly: Anthropic was right.

The company refused two specific things: enabling mass surveillance of US citizens and fully autonomous weapons. That was the line. The Pentagon, which prefers broad “all lawful purposes” contract language, didn’t want explicit carve-outs. Anthropic said no. The government retaliated with a designation historically reserved for foreign adversaries like Huawei.

And then something remarkable happened.

Claude shot to number one in the App Store. Daily signups broke records every day that week. Free users climbed more than 60% since January. People chalked “GOD LOVES ANTHROPIC” on sidewalks outside their San Francisco headquarters.

The Streisand Effect, but make it AI ethics.

OpenAI, sensing an opening, signed the Pentagon contract under the standard federal language. Sam Altman said he doesn’t believe unelected corporate leaders should supersede democratically elected officials. It’s a reasonable-sounding argument. It’s also convenient timing.

Dario Amodei called OpenAI’s framing “straight up lies.” The vibe between these two companies is... not great.

Here’s what I actually think is happening beneath the headline drama: We’re watching the first real public fight over who controls AI. Not abstractly. Not in a Senate hearing nobody watched. In real time, with real consequences, with real principles at stake. And the government got off on extraordinarily bad footing, to put it generously.

Voters are divided on AI broadly, 48% favorable, 46% unfavorable, but the strongest predictor of how someone feels about it is whether they use it. Daily users are favorable by 57 points. People who rarely touch it are unfavorable by 42 points. That gap is going to define a lot of policy fights over the next two years.

Because the economic anxiety is real.

46% of voters think AI will hurt the American economy. 54% think it will increase unemployment. 65% think it will make billionaires richer while only 27% think everyday people benefit.

Those numbers don’t go away because the technology is impressive. They go away because someone makes a compelling, honest case that the benefits are distributed broadly.

Nobody is making that case right now with any conviction.

What Anthropic did, whatever you think of the specifics, was at least legible. They had a line. They held it. They paid for it... and then got rewarded for it by the public, which tells you something about where trust actually lives right now.

I’m not naive about corporate motives. But I know what it looks like when a company decides its values are non-negotiable, and I know what it looks like when they aren’t.

This one was pretty clear.

*AI Disclosure: 100% of this was written by me, a human with only light edits and typo corrections from Grammarly. The images are mostly AI-generated using Nano Banana.

Share The Four P's

Subscribe now

There’s a name for this. It’s called the Baader–Meinhof Phenomenon, also known as the “frequency illusion.” The term was coined in the 1990s after someone learned about a little-known German militant group and then suddenly encountered references to it repeatedly.

Later, linguist Arnold Zwicky formalized it, explaining that it’s a mix of selective attention and confirmation bias.

In other words, once your brain tags something as important, it starts scanning for it.

• Brands work this way. Stories work this way. Politics absolutely works this way.

• Nike doesn’t just sell shoes. It trains your brain to notice belief.

• Social media doesn’t just amplify content. It amplifies what we’re primed to see.

In 2026, the battle isn’t for impressions. Or even attention. It’s for perception.

And whoever shapes what we notice... shapes what we believe.

PERSONAL: Sports, Stories & Stuff That Actually Sticks

I can’t get my kids to watch sports with me.

Not in the “they’re busy” way. In the real, unmistakable indifference way. They don’t hate it. They just don’t care.

Meanwhile, I’m the guy staying up late to watch Division III college basketball on ESPN++. I’ll wake up early for Olympic curling qualifiers like it’s a playoff game. Sports, to me, are ritual. They’re time stamps in my life. I can map childhood memories to championships and heartbreaks.

We grew up in a different era.

Sports were stories. Sports were culture. Your team wasn’t just a preference. It was geography. It was family. It was identity. When Michael Jordan stretched in that Wings poster with William Blake’s quote, it wasn’t just a brand asset. It was mythology.

“No bird soars too high…” That line didn’t trend. It stuck.

Today, only about 20% of Gen Z adults identify as avid sports fans. 1/3 say they don’t follow sports at all. Leagues are reacting by flooding events with creators, leaning into meme culture, chasing digital-native formats.

Sports isn’t disappearing. But its role as shared cultural glue is...

The Emotional Infrastructure We Took for Granted

This isn’t just about sports. It’s about how meaning gets built.

Those of you who know me know I’m borderline irrational about Nike. I’ve accumulated more Swooshes than I care to admit. Greg Hoffman’s Emotion by Design puts language to something powerful: great brands don’t win on features. They win on feeling.

Nike didn’t dominate by optimizing specs. It built emotional infrastructure.

Empathy. Curiosity. Risk.

The “vision advantage” Coach K talks about isn’t about sharper creative. It’s about seeing what others miss. Empathy lets you understand people’s dreams and frustrations. Curiosity forces you outside your echo chamber to find unexpected inspiration.

That’s how Nike produced five-minute animated films like The Last Game ahead of the 2014 World Cup. Massive risk. Long production. No obvious product plug. It became the most shared brand video of the tournament.

Not because it gamed the algorithm. Because it created memory.

And memory creates culture.

The Fragmentation Effect

I used to think my kids lack attention spans. But it’s not that. They lack shared narrative.

We had fewer channels and fewer identity streams competing for our time. When something big happened in sports, we all experienced it together. Now every feed is personalized. Every algorithm curates a different reality.

The Baader–Meinhof effect on overdrive. You see what your brain is trained to see. My feed is full of late-night box scores and Nike nostalgia. StockX steals as I scroll through my feed.

Theirs is full of creators, micro-communities, and culture slices that feel more participatory and immediate than a three-hour broadcast.

Sports used to be communal by default. Now it has to compete for cultural oxygen.

Gen Z isn’t anti-identity. If anything, they’re hyper-aware of it. But their identities are curated, fluid, and self-constructed. They don’t inherit fandom the way we did. They assemble it.

Purpose Over Performance

Air Force 1’s weren’t designed to be a cultural icon. They were built authentically for basketball players. It won because it served a purpose. Then culture adopted it.

When Nike launched the Apple Watch Nike+, the message wasn’t about processing speed or battery life. It was “Running just became a lot easier,” embodied by Kevin Hart pushing through the desert.

Purpose over features. Identity over optimization. And in an era where AI floods feeds and algorithms fragment attention, emotional clarity becomes a competitive advantage.

Maybe my kids won’t ever sit through a random Tuesday night game with me. Which is kinda sad.

But the principle hasn’t changed. Humans attach to meaning.

And the brands, leagues, and creators that endure will be the ones that design for emotion first and distribution second.

PROFESSIONAL: AI as a Mirror, Not a Monolith.

If you want whiplash, subscribe to three different AI newsletters with respect to marketing, technology, and culture.

One says: Invest more.
Another says: You’re wasting money.
A third says: You’re about to work harder than ever.

Welcome to the most confusing professional landscape of our careers... where the only common thread is perspective.

1. The Boardroom View: Invest Now, Justify Later

AI budgets are surging. Fundraising is booming. Product launches are relentless. But under the hood, it’s alllll about anxiety.

A recent global survey of CIOs found that 74% regret at least one major AI vendor decision in the past 18 months. 62% say their CEO has questioned those decisions. Nearly 1/3 have been asked to justify AI outcomes they can’t fully explain.

Read that again.

We’re investing aggressively in something we cannot yet cleanly measure. At the same time, CFOs are planning to increase AI spending in 2026.

Not because ROI is crystal clear, but because AI is viewed as existential. A “Return on the Future” mindset. If this becomes foundational infrastructure, you cannot afford to be late.

Translation: Nobody wants to be Blockbuster. Nobody wants to explain why they didn’t invest. Everyone wants measurable returns. Few can clearly prove them.

We’ve shifted from experimentation to accountability.

The vibe is no longer “try it.”
It’s “show me.”

But measurement frameworks are lagging behind a technology still defining itself. That tension is where we’re living.

2. The Worker View: AI Makes You Faster. Then Busier.

Meanwhile, on the ground level, an eight-month study of enterprise AI adoption found something counterintuitive: AI didn’t reduce workload. It intensified it.

Workers completed tasks faster, so they took on more.

AI made complex tasks feel accessible. Non-technical employees suddenly coding. Marketing teams prototyping. Finance modeling faster. The result: it blurred work boundaries. It encouraged multitasking. It became a “partner” running in the background.

And slowly, almost invisibly, workloads expanded.

This is the productivity paradox, when tools remove friction, expectations expand to fill the space.

We used to hire another person for that extra scope... Now we say, “Can’t AI help with that?”

This isn’t dystopian. It’s cultural. But it does raise a question:

Are we using AI to eliminate work? Or to stack more onto the same humans?

3. The Market View: The Middle Is Disappearing

Now let’s zoom out.

Freelancers are a great canary in the coal mine. About 1/3 have fewer clients and earn less.
Another 1/3 have fewer clients and earn more. The difference = Positioning.

One group elevated into premium expertise. Strategic thinking. High-trust advisory.

The other got commoditized.

AI didn’t kill freelancing... it hollowed out the middle. And that pattern extends to marketing broadly:

• 52% use AI daily. It’s expected.

• 39% say output quality is often worse than human work.

• 53% expect flat budgets in 2026.

• Only 32% feel confident measuring ROI.

• Nearly 90% say email drives highest ROI.

• 42% plan to increase brand investment.

That’s not contradiction. It’s adaptation.

When acquisition costs rise and attribution confidence falls, brands retreat to what compounds: mental availability.

You can’t outbid forever, but you can out-position.

(Ooooh, I like that. Write it down!)

So What’s Actually Happening?

AI is not eliminating the professional class. It’s stratifying it. It’s forcing clarity.

Are you a button pusher...? A prompt operator...? Or a decision architect?
Are you using AI to chase speed? Or to create leverage?

Some leaders see AI as a cost center. Others see it as infrastructure. Some workers feel threatened. Others feel amplified.

To quote Austin Powers, I’ve just gone cross-eyed.
Yet ALL are right, because AI is not a single outcome. It’s a force multiplier.

And like every force multiplier in history, it rewards positioning, adaptability, and strategic thinking.

The people who win won’t be the ones who avoided AI.

They’ll be the ones who decided what role they want to play in an AI-shaped economy:
Operator or orchestrator. That choice is still ours.

PRACTICAL: Your Next Creator May Be in the Break Room

Maybe I stopped paying attention for a decade, but in the early days of social media, one of the core offerings my team and I built for brands was employee training. Toolkits. Social guidelines. Camera comfort. How to represent the company without sounding like a robot reading from Legal.

Then the influencer era exploded. Then paid creators took over. Everything became performance media. Now it feels like employee-generated content is back.

Or maybe it never left and it’s just another example of Baader–Meinhof waving its flag in my face.

As far back as 2010, consumers said they wanted to see more frontline employees on brand accounts. Real people. Real expertise. Real culture. Not stock photos and sanitized captions.

Fast forward to now and employee influencer programs are everywhere again.

And for good reason.

Employees are credible. They’re contextual. They have domain expertise. A product manager explaining why a feature exists will always feel more authentic than a brand account announcing it. A journalist explaining nuance on camera beats a carousel headline. A store associate showing how something works beats a polished 30-second ad.

This isn’t about turning your staff into TikTok stars.

It’s about unlocking institutional knowledge in human form.

If you’re serious about it, here’s the same practical playbook I trained Coca-Cola on all those years ago:

• Start small. Pilot with people who are already comfortable on camera.

• Build training resources. Not just brand guidelines, but storytelling frameworks.

• Expand beyond marketing. Engineers. Customer service. Operations. Diversity of POV builds credibility.

• Equip them. Tools matter. So does time allocation.

• Extend beyond social. Events. Recruiting. Sales enablement. Internal culture.

At many companies, employees are already posting about work anyway. Sometimes brilliantly. Sometimes chaotically. Sometimes deliciously, like the Staples Baddie.

You can either ignore that.

Or you can embrace it and build structure around it.

In an overwhelmingly AI-saturated world where feeds are full of polished sameness, the most differentiated asset you have is your people.

Turns out, the most scalable creator network might already be on payroll.

POLITICAL: Parental Controls Go Geopolitical

For years I’ve been that guy at dinner who says, “Maybe giving 12-year-olds algorithmic slot machines wasn’t our finest innovation moment.”

People nod. Then refill their wine.

Now governments are nodding too.

From Sydney to Paris to New Delhi, the push to restrict teens from social media is going global. Australia blocked under-16s in late 2024. Spain is moving toward 16. Germany has backed a ban. France passed one for under-15s. More than a dozen countries are debating similar moves.

It’s happening!

When 93% of middle-schoolers in France are already on social platforms, this stops being a niche parenting debate and starts becoming public policy.

In the U.S., most teens 13-17 visit YouTube, TikTok, and Instagram daily. About 20% say they’re on “almost constantly.” Addicted much? Yet despite lots of disagreement on just about everything else here, there’s political “interest” in restrictions. Whether that turns into enforceable law is another story.

We know their billionaire owners and CEO have the ear of this administration. And tech companies argue bans will backfire. Push kids to smaller apps. Less safe corners of the internet. Multiplayer games. Encrypted communities. The digital equivalent of “they’ll just sneak out the back door.”

They might be right. But that’s not the real signal.

The real signal is this: Governments are no longer debating whether social media affects young brains. They’re debating how aggressively to intervene. That’s a shift.

Lawsuits are stacking up. Texas sued Snapchat, alleging deceptive safety claims and addictive design features like infinite scroll and Snapstreaks. Courts in Europe are pressuring platforms to tighten protections. Executives are testifying that “too much use” is subjective.

Meanwhile, parents are divided. Teens are divided. Platforms are defensive. And, yeah, the marketers and advertisers are watching.

Because if regulation tightens, engagement mechanics change.
If targeting minors gets restricted, audience strategy changes.
If algorithm transparency becomes mandatory, optimization changes.

And if infinite scroll gets reclassified from “smart UX” to “addictive architecture,” that changes everything.

This isn’t just about kids.

It’s a Baader–Meinhof moment for society.

Once governments tagged social media as a potential public health issue, they started seeing it everywhere. Addiction. Anxiety. Polarization. Privacy. Lawsuits. Election interference. Brand safety.

Selective attention at a national scale.

For years, platforms optimized for attention. Now they may be forced to optimize for accountability.

And if that happens, brands will need environments that are not just high-reach... but high-trust.

You can call me dramatic. You can call me cautious. But when regulation crosses borders this fast, it’s not noise. It’s perception shifting.

And perception, as we said, is power.

*AI Disclosure: 100% of this was written by me, a human with only light edits and typo corrections from Grammarly. The images are mostly AI-generated using Nano Banana.

Share The Four P's

Subscribe now

Nothing beats this stretch. Or at least, nothing used to.

Sports were the one thing we could all agree on. The last shared language. The last place where we showed up together, wore the same colors, cheered the same moments, and briefly forgot how fractured everything else felt.

But even that has cracked.

Yes, sports have always been political. Sports have always been big business. And yes, sports have always been personal for me.

What’s changed isn’t athletes speaking up.
What’s changed is how violently some people react when they do.

The outrage isn’t about sports. It’s about control. And fear of who gets to represent “America.”

And it absolutely should be personal to every one of us.

THE PERSONAL: Values on Display

Prior to Sunday evening, I did everything in my power to avoid Super Bowl ads and commercials that were pre-released for PR and social consumption.

It was hard. Friends and former colleagues, even clients, had spots running in the Big Game. My LinkedIn feed was cluttered with commentary, as were all trades and mainstream press. It was hard, but I am proud of my effort, and my hot takes were truly unscripted.

But what I could not avoid was the absurd backlash against this year’s Super Bowl halftime performer, Bad Bunny.

The megatalented Puerto Rican sensation, fresh off his historic Grammy win for DeBÍ TiRAR MáS FOToS, the first Spanish-language album to ever win Album of the Year, was an obvious choice months ago.

Yet somehow, a fifteen-minute halftime show caused more outrage than a predator president.

Let that sink in.

Bad Bunny’s performance sent racist MAGA corners of the internet into a full-blown meltdown. No subtlety. No dog whistles. Just open, foaming-at-the-mouth racism. Wonder where they took those cues from...?

I’m so tired of living in a world where that gets more heat than crimes against humanity.

Meanwhile, at the Olympics, athletes are doing what athletes have always done: representing values.

Over the weekend, U.S. skier Hunter Hess said it plainly:
“Just because I wear the flag doesn’t mean I represent everything that’s going on in the U.S.”

That statement shouldn’t be controversial.
It should be obvious.

Anyone upset that he’s “bringing politics into the Olympics” doesn’t understand that the Olympics are (and always have been) political. Flags. Anthems. Boycotts. Protests. Propaganda. Unity. Division. It’s all baked in.

Fellow skier Svea Irving put it with more grace than most adults manage:
“It’s definitely a tough time in our country right now. I just continue to represent my values, which are compassion, respect, and love for others.”

That made me emotional.
And proud.

Maybe this awful stretch of history is a stress test. A reminder of who we really are when things get ugly. A push to decide which values we’re actually willing to defend.

Then there’s Amber Glenn, making history this week as the first openly queer woman to represent the U.S. in Olympic figure skating. She’s limiting her time on social media after receiving threats for speaking honestly about what’s happening in this country.

Read that again: Threats. For existing. For speaking.

I want to thank Amber Glenn, Hunter Hess, Chris Lillis, and others for speaking up while representing the United States on the world’s biggest stage.

Staying quiet is never the answer. But speaking up still takes courage, especially when you’re a public figure and the mob is loud.

These athletes didn’t ask to make political statements. The media asked the questions. They answered with maturity, clarity, and humanity.

They represent the people. They represent what America should stand for. They represent morals, not administrations.

They love this country. They just don’t love child rapists, Nazis, or racists.

In the old days, that wasn’t controversial. Now it apparently is.

And that’s exactly why their voices matter.

THE PROFESSIONAL: Repositioning the Power of Community

I wrote about this on LinkedIn last week, but it’s worth repeating here because the distinction keeps getting flattened.

Community is not an audience. Community is not a network.

Audience is consumption. Network is adjacency.

Community is mutual recognition, contribution, and memory.

These differences matters more than ever.

IRL, our attitudes and behaviors either build community or quietly erode it.
Online, especially on platforms like LinkedIn, our content should behave like community, not just talk about it.

This isn’t new for me. Nearly two decades ago, in a poorly lit conference room in New Jersey, I put language around what I was seeing in practice. I called it the 5 Cs of Community, and I still come back to it today because the fundamentals haven’t changed. The environment has.

• Content: If you don’t have something meaningful to contribute, nothing starts.

• Conversation: Are you talking to people or with them? Community is multi-directional by default.

• Connection: Almost literal. Are the pipes actually in place for people to connect with each other, not just with you?

• Collaboration: Not just give and take, but working together to build something better than any of us could create alone.

• Continuity Community doesn’t live in moments. It lives over time. It evolves. It changes. But it can’t stop.

If one leg of that table breaks, the whole thing wobbles.

Community still converts

The clearest proof point right now is Reddit. 70% revenue growth. 19% user growth. 121 million global daily active users.

They’re even phasing out the logged-in vs. logged-out distinction because engagement across both groups keeps improving. Add a $1B share buyback and evolving data partnerships with Google and OpenAI, and the message is clear.

Reddit’s audience is valuable whether users arrive intentionally or stumble in from search.

But the bigger lesson isn’t financial. It’s structural: Focused topics. Persistent threads. Shared norms. Memory. Community still converts because it gives people a reason to stay.

Group chats are real communities

If you want a counterpoint to “community is dead,” look at group chats.

A recent piece in Vogue described progressive mom group chats evolving from summer-camp logistics into grassroots organizing hubs responding to ICE activity. These groups coordinate rapid alerts, mutual aid, boycotts, and even candidate recruitment.

One chat helped recruit 30 Democratic candidates to run for local office.

No algorithm. No follower counts. Just trust, relevance, and continuity. That’s community.

Loneliness is the signal

The most haunting signal of all came from a different place. A Chinese app called “Are You Dead? asks users living alone to check in every two days or it alerts an emergency contact. It’s now one of the most downloaded paid apps in China and climbing in the U.S.

This isn’t a gimmick. It’s a response to scale loneliness.

When one-person households are becoming the norm, community isn’t a marketing tactic. It’s infrastructure.

Social platforms aren’t the answer

Which brings us to the uncomfortable part: Social media is increasingly bad at hosting communities.

U.S. TikTok users flirted with alternatives during ownership uncertainty, then mostly drifted back once the glitches stopped. Downloads of replacements slowed. Habit won. Gravity won. But community didn’t.

Even TikTok optimizes for individual engagement, not shared memory. And that’s not a TikTok problem. It’s a platform problem.

John Herrman recently described this shift as “desocialized media” in New York Magazine. Platforms once built around shared purpose now optimize for isolated consumption, where we’re all recreationally A/B testing content inside our own algorithmic bubbles.

We’re connected. But not together.

Where the refocus starts

Community is harder than audience-building. Slower than growth hacking. Less scalable than reach.

But it’s also more resilient.

When one of the 5 Cs breaks, you feel it immediately. I’ve felt it myself, publicly, on my own socials. I’ll share some of those failures soon. They’re instructive.

If we want stronger communities, online or off, the work isn’t new. We just need to reprioritize it.

THE POLITICAL: Permanent Headquarters

The blip in the political force this week was an actual good idea from the Democratic Party.
Yes. An actual good idea. From people at the top.

The decision to relaunch Kamala Harris’ campaign infrastructure as Headquarters isn’t about whether she runs again in 2028. It’s about finally understanding the game they’ve been losing for a decade.

Republicans build movements.
Democrats build campaigns.
Campaigns end. Movements metastasize.

That’s the real reason the Trump shit train keeps rolling on. It never stops at the station. It just keeps adding cars.

What Headquarters gets right is deceptively simple: don’t tear down the house every two years. The platform, rebuilt in partnership with People For the American Way, founded by Norman Lear, keeps the social graph, the audience, the talent, and the muscle memory intact. Same accounts. Same voices. Same gravity. Harris as chair emerita is almost beside the point. This is infrastructure, not idolatry.

And infrastructure is where Democrats have historically failed themselves.

The right has permanent organizing.
The left has pop-up tents.

What’s interesting is that the playbook for what comes next isn’t theoretical. It’s already working elsewhere.

Look at Japan’s Sanae Takaichi. Nearly 90% approval among 18–29 year olds. Not because she spammed memes, but because she made herself legible to a generation locked out of old power structures. Selfie diplomacy. Narrative consistency. Cultural fluency. Her supporters even coined “Sanakatsu,” borrowing from idol fandom culture. That’s not campaigning. That’s community formation.

Or look closer to home with Zohran Mamdani. His first formal press conference wasn’t for legacy media. It was for creators. Seventy-eight of them. Eighty-two million combined reach. He didn’t just allow them in the room. He told them they were the room. That’s not anti-press. That’s understanding where trust and attention now live.

Meanwhile, conservatives aren’t debating whether this stuff is “serious.” They’re operationalizing it. Corporate and foreign interests are paying Gen Z influencers directly. No think tanks. No white papers. Just cash, content, and access. TikTok to the White House. Literally.

Here’s the uncomfortable truth Democrats need to absorb quickly: this isn’t about vibes, memes, or Gen Z pandering. It’s about building Progressive Media Networks the same way conservatives built talk radio, cable news, and now creator-led influence machines. Permanent. Funded. Distributed. Talent-driven.

Headquarters is a start.
A smart one.
But only if it evolves from a brand into a backbone.

Stop rebuilding every cycle.
Start compounding.

THE PRACTICAL: The Agents are Growing Up

I almost made it through an entire Four Ps without mentioning AI. Alas, I’m merely an agent of practicality myself.

It’s not just unavoidable. It’s inevitable.
And it’s moving faster than most people think.

Over the last 6 months, one shift has dominated conversations with teams buying and deploying software: the explosion of AI agents in real-world use.

Aor exmaple, AWS launched its agent marketplace in mid-2025, the internal goal was modest. 50 agents. Something exploratory. By early 2026, that number had surged past 2,600 agents, making it the fastest-growing category in the platform’s history.

That doesn’t happen because of demos. It happens because people are trying to get work done.

The promise of agents is seductive. Systems that don’t just answer questions, but do things. Research competitors. Monitor pricing. Generate reports. Trigger actions. In controlled environments, they work beautifully.

In production, most of them fail. Not because they can’t reason.

Agents don’t fail at thinking. They fail at retrieval.

Blocked requests. Stale data. Rate limits. CAPTCHAs. Sites that behave differently by geography. Sessions that drop mid-workflow. The intelligence is there. The access isn’t.

The uncomfortable truth is that the web was never built for agents.

Humans browse intuitively. We click, scroll, wait, refresh, retry. We adapt without noticing. Agents can’t. They need programmatic access to live data, at scale, across regions, without getting flagged or blocked. And the public web actively resists that behavior.

This is where most agentic projects stall. Teams spend more time debugging infrastructure than building features. The roadmap slows. Confidence erodes. The “AI transformation” quietly turns into a maintenance project.

When access stops being the bottleneck, entire categories of use cases suddenly work.

Teams can systematically monitor how GPT responds to the same prompts across markets and over time. E-commerce players can track pricing, availability, and sentiment across thousands of products without relying on sampled data. Financial and market intelligence teams can monitor rapidly changing sources and detect meaningful shifts in near real time.

These aren’t futuristic scenarios... they’re practical outcomes.

Many teams try to build this infrastructure themselves. They spin up scrapers, manage proxy pools, and handle retries manually. Then they spend months maintaining it. Every site change breaks something. Every new geography introduces edge cases. The infrastructure work never ends.

That’s the hidden cost of “just build it.”

At the same time, agents themselves are being repositioned. OpenAI recently launched Frontier, an enterprise platform for deploying and managing agents across organizations. Notably, they’re calling them “AI coworkers.”

That framing matters.

If agents are treated like employees, AI spend shifts from IT to operations. Different budget owners. Bigger budgets. Higher urgency. Frontier gives agents shared context across siloed systems, auditable permissions, and the ability to learn from experience. One manufacturer reduced a six-week optimization process to a single day.

OpenAI is even embedding forward-deployed engineers with customers, borrowing the consulting-wrapped-in-software playbook that helped Palantir scale.

The takeaway is simple.

If agents are meant to deliver outcomes, real-time web access is the infrastructure they run on. The biggest blockers won’t be technical. They’ll be cultural debt, bureaucracy, and inertia.

That’s a leadership problem.
Not an AI one.

And once it cracks, we’re going to live in a very different world.

*AI Disclosure: 100% of this was written by me, a human with only light edits and typo corrections from Grammarly. The images are mostly AI-generated using Nano Banana.

Share The Four P's

Subscribe now

Sequestered at a gorgeous new hotel just a few miles from the Mexican border, the extreme weather hype suddenly became real. Flights were canceled early. Schedules unraveled. And hundreds of grocery and retail executives quietly scattered to get back to their teams.

One called it “DREAD: disaster readiness execution and deployment.”

Another midwestern grocery executive said, “We’re preparing for the ICEpocalypse.” Though I wasn’t sure which ICE they were talking about.

But no panic. Some drama. Lots of decisive movement.

Because while the trigger was weather, the conversations underneath were about something bigger. Fragility. Anxiety. Non-linear risk. Systems that look solid right up until they don’t.

Whether we were talking about supply chains, labor, AI, or public trust, the failing government… the same tension kept surfacing.

Everything feels closer to the edge.

I’d planned to leave early anyway. Three days is about the max I can do at a work event now without checking a suitcase or my sanity. But watching people leave wasn’t about convenience. It was instinct. Leaders sensing a shift and acting before the narrative caught up.

That instinct showed up everywhere last week. In how businesses are rethinking resilience. In how technology is outrunning its economics. In how people are opting out of systems that no longer feel stable, fair, or human. And in how power, when misused, stops looking like order and starts looking like fear.

This edition of The Four Ps lives in that tension.

• Between preparation and panic.

• Between velocity and sustainability.

• Between authority and legitimacy.

• Between what we’re told is “normal” and what clearly isn’t.

So yes, most of this edition was inspired by a conference… and a snowstorm (we’re barely dug out at this point), but it is fueled by the growing rage that boiled over this past weekend in Minneapolis.

In all, this is really about what happens when multiple systems start icing over at the same time. And how we respond when the forecast stops being hypothetical.

THE POLITICAL: Put It On ICE

Let’s be clear about something before the euphemisms kick in.

What’s happening right now is not “border enforcement.” It’s not “law and order.” It’s a slow-motion civil rights collapse happening in plain sight. And it’s being carried out by Immigration and Customs Enforcement.

There’s a dark irony in how often the word ice comes up lately. Because this ICE is colder than anything Mother Nature could throw at Minnesota. And unlike a storm, this one is deliberate.

Raids in neighborhoods. Agents showing up at schools. Children detained. Citizens and undocumented migrants swept up together, as if due process were optional and the Constitution were a suggestion. Brazen murder of a VA ICU nurse… and then lying about it blatantly and criminally.

This isn’t strength. It’s panic dressed up as power.

What makes it worse is how many Americans are cheering it on. Not because they’ve thought it through, but because cruelty has been marketed as competence. Fear reframed as patriotism. “Someone else” made into the problem because it’s easier than confronting the systems that actually failed us.

Meanwhile, the rest of the world is watching. And they’re not confused.

Europeans and Americans may disagree on plenty. But whether it’s in their drinks or on our streets, we all really, really hate ICE. The images coming out of Minneapolis don’t read as security. They read as authoritarian cosplay. The kind that erodes trust, invites chaos, and makes us less safe, not more.

And let’s not pretend this is some finely tuned operation.

This is the same agency that has made headlines for killing civilians, detaining children, mistakenly hiring journalists without background checks, and tripping over itself in moments that would be slapstick if they weren’t so horrifying. Competence is not terrorizing preschoolers. Effectiveness is not raiding schools. Enforcement without restraint is just violence with a badge.

Before Alex Pretti was murdered this past weekend, another event that shook me happened to a high school student.

Minneapolis junior Lila Dominguez documented ICE agents coming onto school grounds. Her hands were shaking as she reported it. Not because she was being dramatic. Because she was being honest.

Going to school should be the safest thing a kid does all day. The fact that our own government is interfering with that should stop every adult in their tracks.

And yet.

Here’s the deeper fracture that scares me. Young people are watching all of this and opting out entirely. Not toward extremism. Toward exhaustion.

A majority of Gen Z and millennials now reject both political parties. Not because they’re apathetic, but because instability has become their baseline. Financially. Politically. Emotionally.

They don’t see nuance because nuance hasn’t shown up for them. They see a system that can mobilize instantly to deport families, but can’t seem to mobilize to protect housing, healthcare, or education. They see force where there should be care. Speed where there should be judgment.

And they’re right to be angry.

But this isn’t about left versus right. It’s about right versus wrong.

Mass roundups don’t make us safer. Violating civil rights doesn’t make us stronger. Turning neighbors into enemies doesn’t make us more secure.

It makes us smaller. It makes us feared, not respected. And it guarantees backlash, not stability.

History is very clear about this part. You don’t hold a country together with terror.

If people are protesting in subzero temperatures, it’s not because they’re bored. It’s because something fundamental has broken. And the longer we normalize it, the harder it will be to put back together.

ICE doesn’t need more funding. It needs restraint, oversight, and accountability.

And the rest of us need to stop pretending this is normal. Because it isn’t.

The rage many of us feel right now is not misplaced. It’s overdue.

THE PRACTICAL: The End of Predictable Leadership

Back to the less important… last week, I found myself a little out of my element.

Standing on a stage as a startup pitch finalist at the aforementioned conference, with 1,700 leaders from the grocery, retail, and CPG world. Not just marketers. The people who run supply chains, stores, safety, operations, and entire businesses you interact with every single day.

It reminded me of something I learned at a Dentsu leadership offsite about 15 years ago. Back then, the framework was VUCA: volatility, uncertainty, complexity, ambiguity.

Useful at the time. But it feels… polite now.

The CEO of the industry association putting on the conference, FMI - The Food Industry Association - introduced a framework that actually describes the moment we’re in: “BANI.” It’s not new, but it was the first time I’d seen it adapted in a modern context.

BANI means: brittle, anxious, non-linear, incomprehensible.

And the shift isn’t semantic. It’s behavioral. These aren’t temporary conditions to manage through. They’re permanent requirements for leadership.

Brittle → Resilient
The food industry is quietly one of the most resilient sectors on the planet. From barcodes to modern supply chains to how it responded under COVID pressure, it bends without breaking.

The leaders who stood out last week had one thing in common: crystal-clear intent paired with flexible execution. Think airplane wings. Firmly attached. Constantly flexing.

They plan relentlessly for “what if” scenarios. Disruptions. Disasters. Access issues. Vulnerabilities. Not because they expect chaos tomorrow, but because resilience is built before it’s needed.

The mindset I kept hearing, even if no one named it, was bendable intent. Long-term direction stays fixed. Daily tactics stay loose.

Anxious → Anchoring
Anxiety is no longer an edge case. It’s ambient. Workers feel it. Consumers feel it. Leaders definitely feel it.

The best leaders right now act as a non-anxious presence. Not by pretending everything is fine. But by staying grounded when others aren’t.

One stat that stopped me cold: people who regularly eat meals together report higher happiness than those with higher income or employment stability. Food isn’t just fuel. It’s structure. Routine. Civility.

That matters beyond the industry. Leaders who reduce noise, create rhythm, and bring people back to basics lower stress by default.

Non-Linear → Neuroflex
This is the hardest shift.

The world no longer responds predictably. Small actions create outsized consequences. Big bets sometimes vanish without impact.

The leaders who thrive are improvisers. They accept the situation as it is. Trust intuition. Notice weak signals. Collaborate in real time. Fail fast without spiraling.

Some things still require linear discipline. Food safety is one. Ethics is another. But leadership, increasingly, is jazz. Not sheet music.

Incomprehensible → Interconnected
No one can know everything anymore. Pretending otherwise is the risk.

Strong leaders build trusted circles of expertise and pull them into decisions early. Especially in areas where speed matters and information is incomplete.

Interconnected leadership isn’t about consensus. It’s about velocity with confidence.

That’s the real shift I saw last week.
Not an industry lesson.
A leadership one.

THE PROFESSIONAL: AI’s Most Important Test Case Is...

AI startups are growing revenue at record speed.

ElevenLabs just crossed roughly $330M in ARR. Lovable is north of $200M. Cursor reportedly hit a $1B run rate. That’s not growth. That’s velocity.

But revenue only tells part of the story.

The costs underneath are brutal. OpenAI is facing multi-billion-dollar annual losses. Anthropic is burning billions too. And every startup building on these models inherits the same reality: compute, data, talent, infrastructure. None of it is cheap.

It requires capital. A lot of it.

In 2025 alone, dozens of U.S. AI startups raised nine-figure rounds. More than $80B has flowed into OpenAI, Anthropic, and xAI combined. That money isn’t just fueling growth. It’s oxygen.

What is surprising is where the real test is emerging: healthcare. A sector with zero tolerance for hallucinations, thin margins for error, and real human consequences when systems get things wrong.

On paper, the momentum is undeniable. OpenAI has launched ChatGPT Health. Anthropic followed quickly with Claude for Healthcare, offering HIPAA-ready tools for prior authorizations, insurance appeals, patient triage, and record summarization. Amazon is rolling out a 24/7 Health AI assistant for One Medical members. This is no longer experimentation. It’s deployment.

The promise is real.
Millions of people lack consistent access to care. Clinicians are buried in administrative work. Systems are fragmented. AI, used well, can compress time, reduce friction, and act as a second set of eyes. Not a replacement for doctors, but support where support is desperately needed.

The risk is just as real.

These models still hallucinate. They still overconfidently produce incorrect answers. And healthcare is the one domain where “mostly right” is still dangerously wrong. Google already learned this the hard way when it pulled AI-generated health summaries after inaccuracies put users at risk.

What makes this moment especially fraught is the cost structure underneath it all.

The same companies pushing hardest into healthcare are also absorbing massive losses. OpenAI and Anthropic are burning billions annually. The startups building on top of their models inherit that reality through compute costs, infrastructure, and compliance requirements. Investor capital is keeping the lights on, but capital doesn’t fix unit economics.

And yet, healthcare may be the only place where that spend can eventually justify itself.

Why? Because outcomes matter more than clicks. Trust matters more than novelty. And value isn’t measured in engagement minutes, but in better decisions, earlier interventions, and reduced systemic waste.

That’s the paradox.

Healthcare is AI’s biggest opportunity and its most unforgiving proving ground. If these systems can be made accurate, private, explainable, and cost-effective here, they earn legitimacy everywhere else.

If they can’t, no amount of ARR headlines will matter.

This isn’t about growth anymore.
It’s about whether AI can be trusted when it actually counts.

THE PERSONAL: The Space In Between

A couple of weeks ago, I’m on the train ride home from work, chatting up a friend and neighbor about where we’re headed.

Not literally, but as a society. One of those conversations that starts half-joking and slowly stops being funny.

We were talking about unemployment. Not layoffs in the LinkedIn sense. Not “re-skilling.” The deeper kind. Structural. Permanent. The kind that doesn’t come with a severance deck and a coaching session. Then it got even realer. As these conversations tend to do.

We were talking about an increasing likelihood of a quiet civil war between the haves and the have-nots. Between people with access to capital, networks, and optionality… and people without any of it. Resources concentrated on one side. Scarcity on the other.

At some point one of us said it out loud: The only people who really “win” in that future are billionaires with stocked bunkers… and the weird survivalists with stocked bunkers.

Everyone else just lives in the middle. Or falls through it. And what stuck with me wasn’t the extremism. It was the absence of a bridge.

We talk a lot about an “autonomous future” where AI does the work and humans finally get to relax. Create. Play. Be present. The problem is no one is seriously talking about the transition period. The gap between employment as identity and whatever comes next.

That in-between is where anxiety lives.

Just 24 hours ago, a new agentic, autonomous “community” called Moltbook was born. Or became self-aware. Or some variation of it. It’s a day old and already scaring the crap out of me.

Forget agentic commerce. This is where agentic subterfuge, warfare, and human destruction will ultimately point to as a source of origin.

Not to scare to, but the agents are already talking about destroying humanity. Just wait until it’s connected to other systems via MCP or other networks, and it’s lights out (literally) and game over. How long before they decide WE are the problem?

This is as real as any panic I’ve ever felt. And we should have seen this coming, even with generative AI, the writing was on the wall(s).

A recent study found that more than 20% of YouTube videos are now AI-generated slop. Content made to fill space. Noise masquerading as output. Not because creators are lazy, but because attention has become the last remaining currency and everyone is scrambling for it.

When people feel replaceable, they flood the system. When systems flood, meaning drains out.

That’s the part that feels personal to me. Not fear of AI, but fear of irrelevance. Fear of being unnecessary in a world optimized for efficiency instead of purpose.

I don’t think we’re headed for Mad Max just yet. I also don’t think we’re gliding smoothly into a leisure utopia. We’re in a messy, uncomfortable middle chapter where old structures dissolve faster than new ones appear. Maybe we’re a decade away from the collapse of society? Maybe 5 years? Maybe 30. But it’s coming

And that’s going to be hard on people.

On the train, we didn’t solve anything. No grand policy ideas. No tidy ending.

Just a shared realization that the future everyone is debating so loudly online already feels very real, very uneven, and very close.

And that the most important work right now might be helping people find footing in the space between what we were and whatever we’re becoming.

*AI Disclosure: 100% of this was written by me, a human with only light edits and typo corrections from Grammarly. The images are mostly AI-generated using Nano Banana.

Share The Four P's

Subscribe now

My year started with one word. Then a practice. Then a posture. A decision to stop holding back, sanding edges, and waiting for permission that never comes.

CES sharpened that instinct. Beneath the demos, predictions, and stage noise, one theme kept surfacing. Trust is thinning. Speed is compressing everything. And openness is no longer a soft skill. It’s leverage.

Being unleashed without being open is just volume. Being open without trust is noise. And trust without action doesn’t survive reality.

This week’s Four Ps is about how I’ve gotten there since New Year’s.

• Professionally, systems are evolving faster than belief can keep up.

• Practically, momentum compounds when intention turns into action.

• Personally, real networks are still built the same way they always have been. By showing up human.

This isn’t about hacks or hot takes. It’s about who earns trust when everything accelerates.

PRACTICAL: My Mantra For the Year is...

This year started with a simple challenge from someone I’ve never actually met before, except for on social media. I saw Matt Kerbel speak once at a Brand Innovators event, then reached out on LinkedIn because I liked his vibe, and now here he is, influencing something I do every single day.
The challenge he posted on New Year’s Eve: Sum up 2026 in one word.

I didn’t overthink it. Unleash.

Not hustle. Not optimize. Not reinvent. Unleash, because this is the year of not holding back.

I remember an early HR lead at one of my first companies saying you’re more likely to accomplish something if you write it down, so that’s what I do: I write things down.

And in doing so, I decided to go even bigger, not just stopping at one word for the year, but instead letting one word within the “unleash” theme go to work with me, for my, and from me every single day. For 365 days.

That’s right, I committed to a word of the day. One word. Every morning.
Something that could guide my thinking, my values, my agenda, my energy.

Two weeks in, I haven’t missed a day. And I have zero intention of stopping.

So far, the words have looked like this:

• Unlimited (Day 1) – removing self-imposed ceilings

• Begin (Day 2) – momentum starts with starting

• Courage (Day 3) – clarity without comfort

• Choice (Day 4) – ownership over autopilot

• Passion (Day 5) – energy as a renewable resource

• Focus (Day 6) – subtraction as strategy

• Trust (Day 7) – in myself, first

• Discipline (Day 8) – freedom through consistency

• Bold (Day 9) – fewer hedges, more conviction

• Commit (Day 10) – no half-yeses

• Momentum (Day 11) – direction beats speed

• Action (Day 12) – thinking ends here

• Pressure (Day 13) - how you handle it matters

Each word ladders up to the same idea. Unleash isn’t chaos. It’s alignment. It’s removing friction between intent and action. It’s saying the thing. Making the call. Publishing the post. Shipping the work.

Trust deserves a pause here... actually I’ll come back to this one. Because once you trust yourself, everything accelerates. This practice isn’t content. It’s calibration.

A daily reminder that forward motion compounds. And that the biggest unlock this year isn’t a tool, a title, or a plan. It’s permission.

Unleashed.

PERSONAL: Be Open to Being Open

Last week I got a reminder that the most powerful growth hack in business is still being a real human in public.
Not a brand voice. Not a deck. Not a carefully sanded LinkedIn persona.

A person.

And it’s not even something I would have thought to share, talk about, or acknowledge. To me, it felt normal. Easy. Even expected.

John Kraski shared a story on LinkedIn that I would not have thought warranted mentioning. Because he did the thing most marketers and executives are quietly terrified of doing.

He was open. And he asked for help out loud.

I first met John on LinkedIn a few years ago. Not IRL. Not because of a CRM or even a warm intro. We met because we were both open enough to show up like humans. Talking. Commenting. Sharing half-formed ideas.

So when we ran into each other at CES last week and he said, “Hey, I’m trying to get connected to someone at the NBA. I’m stuck,” there was no friction. No awkwardness. No positioning.

Just trust.

48 hours later, he was on a thread with the right people. (Thanks to Melissa and Bob!) That’s what being open buys you. Not impressions. Not likes. Not vanity metrics.

Access.

Most marketers and execs are trained to do the opposite. On and off LinkedIn. Protect the brand. Polish the message. Control the narrative. Never look unsure.

The result is a thousand beautiful profiles and almost no real relationships.

Being open feels risky, unprofessional, inefficient. It’s none of those things. It’s how you create gravity. And it’s how orbit expands.

This is why my only rule on LinkedIn is simple: Be open.

Being open isn’t just about posting more. It’s about participation. Choosing curiosity over cool. Conversation over consumption. Visibility over safety.

Most people scroll. Only a few show up. The ratio of consumption to creation is well over 99–1. And the difference between those two groups isn’t talent, timing, or strategy.

It’s willingness.

Willingness to comment when you could lurk. To DM when you could stay anonymous. To share half-formed thinking instead of waiting for perfection. To say “I need help” instead of pretending you don’t.

That’s where the network actually forms. Not in feeds, but in micro-moments of interaction that quietly build trust over time.

John and I didn’t build a relationship through a funnel. We built it by being present. By being visible. By being human in the same digital room again and again.

That’s how gravity works here.

Those who create get pulled into more orbits. Those who hide stay weightless.

So yes, be open in what you post. But more importantly, be open in how you engage.

Because in a world obsessed with reach, the real advantage is resonance.
And resonance only happens when people are brave enough to be seen.

PROFESSIONAL: Who Do You Trust When No One Is Watching?

I’ve already talked and written extensively about last week’s CES, which felt very different for me than in previous years. Five days’ worth of takeaways, insights, on-the-ground reporting.

Hallway conversations beat stages. Private signals beat public metrics. Attention beat reach.

If you missed these, you need to watch here to game the social algo...

But yeah, CES felt very different. It wasn’t about shiny demos this year. At least not for me. I didn’t make it to the main gadget-heavy show floor.

For me, it was about trust. How fast we’re burning it. How hard it is to earn.
And how casually we keep outsourcing it to machines.

But both on top of, and underneath, all of that was a deeper current.
One that came into sharp focus listening to Shelly Palmer break down what’s actually happening, not what we wish was happening.

AI timelines collapsed. AGI-level capability is no longer theoretical. It’s operational.

• Five hours of work in five minutes

• Months of engineering output, overnight

• One coding sprint = millions of dollars in value

AI isn’t assisting work. It’s compressing it. But the real shift isn’t speed.
It’s architecture. Orchestration. Distribution.

Models are raw intelligence, useless without direction and dangerous without boundaries.

What matters now is the system around the model:

• Defined context and identity

• Guardrails before prompts

• Skills agents can reliably execute

• Continuous scoring and evaluation

• Governance that assumes failure, not perfection

This isn’t an IT problem. It’s a leadership one. And marketing probably felt it first.

SEO quietly died in the corner while everyone argued about prompts. AEO is the new battleground. Zero-click answers. No traffic. No attribution... Just outcomes.

The new equation: Share of prompt → Share of mind → Share of wallet

And agents are coming for the rest. Not browsing, but executing.

A “holy fuck” moment for shopping, comparing, transacting.

So ask the real questions:

• Who trains the agent’s preferences?

• Who controls the defaults?

• Who owns the data exhaust?

• Who gets trusted when no human is watching?

Which brings us back to the point. In a world where video lies, audio lies, images lie, and interfaces hallucinate confidence… Trust becomes the currency.
And marketers become its stewards.

2026 won’t default to belief. It’ll default to skepticism. Earned trust won’t scale. But losing it will.

That’s the job now.

POLITICAL: Trust, Accelerated and Abused

Science, technology, security, and safety are no longer aligned.
They’re arguing in public. And the rest of us are supposed to just… trust the process?

Yes, because the decision-makers in business and government have always had our best interests at heart.

At CES, Jensen Huang made the case that slowing AI development makes us less safe. His logic: innovation speed and safety move together. Driving a 50-year-old car isn’t safer than building a new one. Fair point. Also convenient, given his company sells the engine for every new vehicle.

Meanwhile, some of the field’s most respected voices like Yoshua Bengio and Geoffrey Hinton are saying, maybe, just maybe, we should look both ways before flooring it. One side says “pause is danger.” The other says “unchecked speed is.” Pick your risk. Neither comes with a refund.

Zoom out and the trust problem gets louder.

We’re told AI will secure healthcare, but browser plug-ins quietly siphon 900,000 people’s private chats.
We’re told identity is protected, while companies pay $740M to fix what’s already broken.
We’re told regulation will catch up, as $200B in AI capital mints new billionaires faster than we can write rules.

But I just can’t trust our politicians here to keep things on track.

Renewables become culture-war villains.
Information blackouts become tools of control.
Military options get floated while “freedom” is hashtagged.

This isn’t new. It’s old power wearing new software.

The pattern is familiar: move fast, centralize control, normalize the exceptional, and ask for trust later. Sometimes wrapped in patriotism. Sometimes in progress. Sometimes in both.

Here’s the uncomfortable truth.

Trust doesn’t come from speed.
Or scale. Or promises. Or even AI content (that was a joke).

It comes from restraint, transparency, and consequences.

Right now, we’re accelerating everything except accountability.

And history is very clear on how that ends.

*AI Disclosure: 100% of this was written by me, a human with only light edits and typo corrections from Grammarly. The images are mostly AI-generated using Nano Banana.

Share The Four P's

Subscribe now

Not the “here’s what I hope happens” kind. The kind that show up uninvited. The ones you say out loud to your spouse. The ones you text to a friend with “don’t @ me.”

The ones you write down because if you don’t, your brain will keep spinning them at 2:11 a.m.

By the time I was done, I had 2,026 predictions for 2026, give or take. Which is… deranged. Unusable. And totally false. I came up with like 100.

So I did what any sane person would do: I prioritized a cross-section that felt representative across the Four Ps. I consolidated. I trimmed. I eradicated. Not because these are my most likely predictions, or even my favorites, but because 26 per section felt right for 2026.

These aren’t in perfect order. Some are practical. Some are personal. Some are spicy. Some are a little too real.

And yes, a few will age terribly. That’s kind of the point.

PRACTICAL: Faster, Smarter, Cheaper, Better. Quieter.

We’re kicking off with the most grounded predictions. Starting easy. Because these aren’t trends. Not hype. Just what quietly becomes normal.

So here’s what changes, what sticks, what stops feeling optional.

2026 brings fewer experiments and more execution. Less fascination, more discipline.
AI stops being impressive and starts being useful.

This year isn’t about using AI. It’s about operating with AI without breaking everything else... systems, workflows, tools, tips, and more:

1. Teams standardize on one AI control layer:

No more five copilots per team. Organizations consolidate around a single governed interface that routes tasks, permissions, and data safely. AI pilots die within 30 to 60 days if they don’t integrate. If it doesn’t connect to CRM, CMS, DAM, analytics, or core systems, it’s cut.

2. Context leakage becomes the #1 internal AI risk:

This is the main event. AI conversations shift from generation (baseline) to orchestration and protection. Companies actively map where brand, strategy, and customer context leaks and fix it. Private and on-device AI rolls out first in sensitive workflows like legal, finance, HR, and executive strategy.

3. “Who can prompt what” becomes a real permission system:

AI access becomes role-based. Junior teams don’t get the same capabilities as leadership or legal. Policies move from ethics theater to enforced controls because compliance, finance, and risk teams demand it.

4. AI agents live inside tools, not beside them:

Agents move into Slack, Notion, Salesforce, Jira, HubSpot, and internal systems. Standalone chat tabs fade. AI becomes infrastructure, not a destination.

5. Meetings get fewer, shorter, and more prepared:

AI briefs, pre-reads, and summaries reduce calls by eliminating unprepared conversation. Meetings that still happen arrive with context attached.

6. AI twins attend meetings without you:

Representation replaces note-taking. AI twins participate in parallel meetings, carry your context, and report back with synthesized decisions. They carry your opinions and decisions, not just summaries. The stress of double and triple bookings fades.

7. Internal search gets fixed before external marketing:

If employees can’t find answers instantly, AI doesn’t help. Knowledge hygiene becomes urgent. Internal discovery shifts from folders, wikis, and dashboards to conversational answer engines employees treat as a thought partner, not a file system. Internal AEO gets fixed long before external marketing does.

8. Docs beat decks for decisions:

Living documents with AI summaries outperform slides no one rereads. Decisions move faster when the artifact stays alive.

9. “Set it and forget it” automation gets banned:

Black-box systems trigger weekly audits. Performance Max-style automation faces scrutiny. One major stack cleanup happens company-wide. Redundant SaaS tools quietly sunset. Fewer vendors. Fewer pipes. Bigger pipes.

10. Teams stop measuring AI usage and start measuring AI impact:

“How many people used it” dies.“What changed because of it” wins. AI exposes broken workflows instead of fixing them. The model isn’t the problem. The process is.

11. Energy efficiency enters AI tool selection:

The AI energy bill becomes impossible to ignore. Cheaper inference, smaller models, and on-device AI get greenlit faster. Builders design leaner systems. Users become more intentional. Energy cost, inference efficiency, and model footprint become buying criteria, not afterthoughts. Big Tech must vertically integrate into energy-smart mode. Google, Amazon, and others will buy power generation to secure compute, not margins.

12. AI-generated drafts become invisible:

Soon, you won’t be able to tell. No announcements. No credits. No debates. AI assistance becomes assumed, like spellcheck. The output matters, not how it was made.

13. Zero-click value becomes the default:

If the insight isn’t usable immediately, it doesn’t ship. Internally or externally.

14. “Experimental” budgets shrink dramatically:

AI tools must prove ROI in weeks, not quarters. Vendors get shorter leashes. Thirty to sixty days to prove value or churn.

15. Manual overrides remain mandatory:

Every AI system ships with a human kill switch. Always. People still matter. And in a world where AI does more, humans are there to keep systems in check. Fewer people may be needed. Oversight remains non-negotiable.

And AI Project Manager becomes the most in-demand white-collar role. The human who orchestrates agents, tools, and workflows becomes more valuable than most specialists.

16. AI outputs are logged, not blindly trusted:

Traceability beats fluency. Logs matter more than confidence. Documentation of what’s been prompted, published, and processed becomes essential for accountability and growth. Systems are designed for failure, not perfection. Fallbacks, redundancy, and recovery plans become standard. Reliability beats elegance.

17. Async work rebounds because AI makes it clearer:

Coordinated communication in remote, hybrid, and async work finally improves. Written updates get sharper. Fewer meetings justify themselves.

18. The most successful AI systems are boring:

Improvements aren’t sexy. They’re effective. No keynote demos. No hype posts. Just quieter teams shipping better work, faster, with fewer mistakes.

19. Digital fatigue drives intentional offline rituals:

People schedule disconnection the way they schedule meetings. Walks, picnics, IRL dinners, and analog hobbies rebound quietly. Detox 2026.

20. Screen-free products for kids become normalized, not niche:

Parents don’t reject tech entirely. They demand bounded tech. Screen-free play, storytelling tools, and tactile toys grow fast.

21. AI companion care enters senior households:

File this under “weird but comforting.” Robotic and conversational companions become acceptable support layers for aging populations, especially as staffing shortages, geographic dispersion, and cost make constant human care unrealistic.

22. Health decisions shift from calories to processing:

“How close is this to its original form?” replaces macro obsession. Ultra-processed food labeling becomes clearer and more controversial.

23. Preventive health tech feels boring but necessary:

Wearables for sleep, glucose, heart rate, and stress become routine maintenance, not self-optimization theater. And yes, this probably starts with pets. Health tracking, emotion detection, and AI assistants get beta-tested on dogs first because we can. Remember who went to space first, too!

24. Trust signals go analog:

Physical mail, print, and private communities carry more weight than public feeds. If it’s tangible or shared quietly, it feels more real.

25. AI anxiety fades into AI expectation:

People stop debating whether AI should exist and start getting annoyed when it doesn’t work properly. Nobody asks how it works. They complain when it’s wrong. My kids use it daily for school. Teachers stop fighting it and start teaching it. Which means I now pay for premium AI accounts. More money. Cool.

26. The default mindset becomes “optimize the boring stuff”:

Less chasing novelty. More fixing friction. Life gets incrementally better, not dramatically different.

PERSONAL: With Whom We Spend Our Time Matters More Than How or Where

Now let’s look inward. To family. To home. To passions. To self.

These are not safe predictions. They’re the ones you argue about in the car, at the table, or in a text thread. They come from lived experience. Bias included. Some will age poorly. A few will age too well.

1. My kids will ask for help later than they should, but it will mean more when they do:

Independence comes earlier. Access to information outside of direct parental guidance is easier and more pervasive than ever. But when help is needed, when it’s actually asked for, it’s real. Fewer fake “I’m fine” answers. Pride delays the ask, but when it comes, it matters. More actual conversations. Fewer surface check-ins.

2. Parenting culture officially moves past “optimization”:

Sleep schedules, enrichment hacks, and perfect routines lose their grip. Parents choose sanity over performance. I will stop caring about bedtimes and, begrudgingly, care less about grades. Maybe.

3. The commute comes back to three days a week and stays there:

Not five. Not fully remote. Hybrid hardens. Podcasts and audiobooks surge. Complaints resume. Five is too many, even for Amazon. I can tell you firsthand that my friends at those companies are not happy. And they won’t stay.

4. Family logistics move fully async:

Shared docs, notes, and calendars replace verbal coordination. That ad for the shared calendar screen in the home? Yeah, we’ll cave and buy one. And “I told you” arguments drop noticeably.

5. Live sports reassert themselves as weekly emotional rituals:

Not highlights. Not betting slips. Sitting down and watching full games becomes a grounding habit again. We need the distraction and the retraining of our failing attention spans. At the same time, pro sports loyalty weakens. Moments replace teams. Fans care more about runs, players, and specific games. Allegiance gets fluid. And yes, I’m saying this as a deeply pained Mets fan.

6. College sports expand and fully break the illusion:

Playoffs are big money. CFP games carry higher stakes. The NCAA chases that revenue across more sports. I can see the NCAA tournament adding more teams and rounds. Fans complain loudly, then watch anyway.

7. Nostalgia stops being ironic and becomes sincere:

Old shows, old music, old routines return without apology. Comfort beats novelty. The 1980s are where we start. Greed makes a comeback. Big hair, shoulder pads, excessive makeup, power suits, neon. Let’s do it.

8. Family dinners get shorter but happen more often:

Fewer long sit-downs. More frequent 20-minute overlaps. Consistency beats ceremony. I’ll enforce device-free dining, even if it means shorter meals. As a result, screen rules for kids get firmer and clearer. Not less tech. Clearer boundaries. Fewer negotiations. Faster enforcement.

9. Health talk shifts from goals to maintenance:

The phrase “best shape of my life” disappears. “Trying not to feel like crap” replaces it. Also, that pain in my upper and lower back will eventually go away. It has to. Right?

10. Food decisions become simpler and stricter at the same time:

People quietly eliminate foods instead of debating them. Less discussion. More discipline.Note to wife: Please stop buying potato chips and chocolate-covered pretzels if you want me to drop ten pounds.

11. Speaking of… ultra-processed food becomes socially uncomfortable:

Not banned. Just judged. Labels get read. Excuses get thinner. Let’s bring this into the mainstream. The chemicals in our food are no bueno.

12. GLP-1 use becomes the dinner party topic of conversation:

Less whispering. Less shame. More honest tradeoff discussions. Moralizing fades.

And when GLP-1 oral pills are approved in 2026 and unleash mass demand shock, those holding back because of needles lose their excuse. Weight-loss stigma fades, supply collapses, and healthcare systems scramble. This is why I’m buying more Eli Lilly stock. Trust me.

13. Walking becomes the most common form of exercise:

Gyms don’t disappear, but walking wins on consistency, accessibility, and mental health. At least for me. Once it’s warmer. And I have more free time. Again… maybe.

14. Digital detoxes happen without announcements:

No challenges. No posts. Apps just get opened less. Quiet exits replace declarations. I have to take one day a week off from LinkedIn posting. I went four years straight posting daily, made a big deal of missing one day in fall 2025, and have since gone right back to daily. To quote James Brown, “Somebody stop me.”

15. Group texts splinter aggressively:

Large threads die. Smaller, tighter groups replace them. Fewer people. More honesty. Please. I have group text exhaustion.

16. Friend circles shrink and stabilize:

Less social expansion. More maintenance. People invest where return feels mutual. Reminder: your Instagram follows and Facebook friends are not real friends.

17. Tay-Trav get married on my birthday:

The wedding of the decade is going to happen next year, at some point. Yet details have been closely guarded. But knowing that 13 is Taylor’s lucky number, next year is ‘26 (13x2), it makes sense for the wedding to happen either on 6/13/26 (her half birthday) or 6/26/26 (my birthday, and the only 13 + Saturday in 2026).

18. Kids develop better AI instincts than adults:

I have no doubt my kids will surpass me in prompting and overall AI literacy. They question outputs, test limits, and distrust authority faster than their parents.

19. Offline hobbies rebound quietly:

Cooking, puzzles, writing, woodworking return. Not for content. For relief.

20. Music fragments further, but hits feel bigger:

Fewer shared songs. Louder reactions when one breaks through.

21. Travel becomes more repetitive and intentional:

People revisit the same places instead of chasing new ones. Familiarity becomes luxury. Wanderlust turns into “comfortlust” as the economy tightens, fear rises, and crowd avoidance becomes about safety, privacy, and sanity.

22. Humor gets darker but more generous:

Less snark for attention. More laughing because everyone’s tired. I love a good dad joke as much as anyone, but we’ll need more than puns to survive this Hatepocalypse.

23. Identity signaling decreases, behavior matters more:

People stop explaining themselves and let patterns speak. For example, a biweekly newsletter in four parts that’s been running since 2018. Do I really need to explain myself? Let the work travel. Skip the pomp.

24. The dominant personal goal becomes “a little calmer”:

I saw an Instagram meme this year. What you want at 10 vs 50. At 10: mansion, cars, fame. At 50: one vacation a year where the kids have their own room and a fridge with an ice machine. Not richer. Not louder. Just calmer than last year.

25. Sports become comfort TV:

Predictable winners, familiar villains, and just enough chaos to keep us coming back. Super Bowls, Finals, and Championships feel increasingly pre-scripted by money, scale, and systems. We still watch. We still care. But the joy shifts from who wins to how long the illusion holds. Parity becomes the fantasy, not the game. So Ohio State repeats as College Football Playoff champion once expansion fully kicks in. The Dodgers buy their way to another World Series. The Denver Broncos beat the LA Rams in the Super Bowl. The New York Knicks lose the NBA Finals to the Oklahoma City Thunder. The Edmonton Oilers win the Stanley Cup over the New Jersey Devils. England makes the World Cup Final and loses to Spain. I’ll complain loudly about the Mets, obviously

26. Marriage becomes more romantic and more resilient:

Less romance-by-default has become a sad trend in marriages. More logistics, more operations. More partnership. I vow to NEVER let that happen in my life. Yes to honesty, yes to shared calendars, and yes to personal time and grace. But I’m ready for bigger grand gestures. More surprises, more fun, more spontaneity. More “I’ve got this, go lie down.” Turns out, that’s love.

PROFESSIONAL: How to Avoid Getting Sucked into the Black Hole of Irrelevance

Now let’s zoom out and back into the work. To industry. To power structures.These predictions aren’t neutral. They pick sides. They assume consolidation, fatigue, accountability, and a growing intolerance for bullshit.

2026 is when the market stops applauding “AI-powered” and starts asking, “Who does this actually work for?”

1. “AI-powered” becomes a red flag in pitches:

This phrase loses its meaning and its persuasive power entirely. Buyers assume AI is table stakes. Saying it without proof signals weakness, not innovation. If your demo can’t show impact in minutes, you’re done.

2. Protection overtakes generation as the core AI value prop:

The biggest AI budgets move from content creation to context protection, orchestration, and governance. Brands realize leaking strategy, customer insight, and intent into shared models is an existential risk.

3. The winning AI platforms behave like operating systems:

Point tools get swallowed. Platforms that plan, route, govern, audit, and remember workflows win. “Cursor-for-X” becomes the dominant mental model across marketing, revenue, and operations. Humans increasingly manage agent fleets instead of tasks, providing taste, judgment, and accountability while agents handle execution.

4. Marketing orgs consolidate to one AI brain:

CMOs shut down the “everyone using their favorite AI” era. One sanctioned control layer. One shared memory. One permission model. Intelligence moves into the tools teams already live in—email, calendars, docs—not dashboards nobody opens. Chaos gets expensive fast.

5. Retail media networks hit their first credibility crisis:

Too many RMNs. Too little differentiation. Brands push back on opaque measurement, inflated CPMs, and duplicated reach. Consolidation becomes inevitable.

6. Commerce media quietly overtakes retail media in influence:

Apps with frequency, intent, and first-party relationships outperform retailers chasing ad dollars. Delivery, travel, and consumer services win the attention war. Retailers struggle because they’re not inherently innovators. Big players like Walmart force customers into corners without consistently proving ROI. Commerce media offers more flexible formats and functions.

7. Creators move upstream into strategy, not just execution:
The smartest brands bring creators in before the brief exists. Creators help shape positioning, narrative, and GTM, not just posts.

8. Influencer-for-hire models collapse:
One-off posts stop working. Long-term creator equity deals outperform transactional campaigns. Fewer creators. Deeper relationships. Real alignment. As defensive AI filters mature, generic outreach doesn’t just underperform… it never reaches a human at all.

9. Community becomes the primary distribution moat:
Audiences that gather, talk, and return outperform audiences that scroll. Comments are not community. Ownership matters. Platforms that don’t get this fall behind.

10. YouTube acquires or builds a real community layer:
The platform is a beast, but it’s basically become a personalized movie theater. It’s missing something big: a community layer. Forums, groups, gated access, memberships. This move would be YouTube admitting that broadcasting alone isn’t enough. Belonging becomes the next battleground.

11. WPP buys a creator-native studio that disrupts its own agencies:Content velocity beats polish. Culture beats decks. Expect layoffs in traditional creative departments and reinvestment in creator-led production systems.

12. The agency, publisher, and studio distinction collapses:
Everyone produces content. Everyone sells media. Everyone claims IP. The middle gets thinner. The edges get sharper.

13. Netflix, YouTube, and even Instagram lean fully into video podcasts and cultural curation:
Podcasters become tastemakers, not just hosts. Audio becomes a visual experience. Curation beats volume.

14. Ad-free streaming becomes niche, not premium:
Ad-supported viewing dominates. Consumers trade patience and privacy for price. FAST and AVOD continue to grow faster than anyone predicted.

15. AI evaluation replaces AI evangelism:
“How well, at what cost, and for whom?” becomes the only question that matters. The hype cycle ends. Proof becomes mandatory.

16. AI productivity dashboards enter boardrooms:
Real-time tracking of productivity gains, displacement, and ROI becomes standard. Anecdotes stop working. Data wins arguments.

17. AI factories face real public backlash:
Data centers get political. Energy use gets scrutinized. Permits slow. PR crises mount. Sustainable AI shifts from aspiration to requirement.

18. Big Tech vertically integrates into energy:
Google won’t be the last. AI companies buy power generation, storage, and infrastructure. Compute follows energy, not the other way around.

19. Venture capital goes fully barbell:
Mega-funds concentrate bets on a tiny number of “inevitable” winners. Early-stage experimentation dries up. Failure finally happens at scale.

20. 95% of the AI LUMAscape becomes irrelevant:
Fewer logos. Bigger pipes. Lower take rates. Better quality. Less margin leakage. The ecosystem gets brutally rational.

21. Privacy-first AI becomes a brand differentiator:
On-device, local, and sovereign AI win trust. Apple’s bet pays off. Users and enterprises start asking where their data actually lives.

22. Government AI regulation fails to deliver clarity:
Executive orders, lawsuits, and state-level chaos create noise, not rules. The courts become the real regulators.

23. AI litigation becomes the dominant enterprise risk:
IP, training data, hallucinations, and liability battles define the year. Legal and compliance teams gain power inside organizations.

24. Search becomes answer-first, not click-first:
AEO vs SEO debates fade. Static menus, forms, and funnels give way to conversational entry points. Great content wins everywhere, but only if it answers real questions immediately. Slop disappears into the void.

25. Brands start showing up inside LLMs whether they plan to or not:
If your brand isn’t referenced, cited, or understood by AI systems, you effectively don’t exist. Brand becomes training data.

26. The most successful companies look boring from the outside:
No hype posts. No launch theater. Just protected systems, owned audiences, durable communities, and quietly compounding advantage.

POLITICAL: Power, Backlash, and the Illusion of Control

This is the uncomfortable section. It’s less polling, less scientific, and more emotion. More pressure. Fewer guardrails.
These aren’t predictions about ideology. They’re predictions about systems under strain.

2026 is when politics stops pretending it’s about persuasion and starts behaving like raw power management:

1. Government AI regulation fails to meaningfully land:
Forced, irrational Executive Orders, task forces, and white papers create noise, not rules. Congress stays gridlocked. Agencies posture. Nothing durable passes. The courts quietly become the real regulators.

2. State-by-state AI chaos accelerates lawsuits:
Attempts to preempt state AI laws backfire. Governors push back. Attorneys general sue. Compliance teams design for legal ambiguity instead of clarity.

3. AI litigation eclipses AI innovation as the dominant policy force:
Training data, IP theft, hallucination liability, and consumer harm lawsuits pile up. Legal precedent, not legislation, defines what’s allowed.

4. A major AI-related court ruling reshapes enterprise behavior overnight:
One ruling. One injunction. One massive damages award. Every boardroom revises its AI posture in weeks, not years.

5. Data centers become a frontline political issue:
“AI factories” face local resistance. Energy grids strain. Water usage sparks backlash. Permits slow. Politicians discover that compute has constituents.

6. Energy policy and AI policy fully merge:
You can’t talk about one without the other. Clean energy, nuclear, geothermal, grid infrastructure, and compute capacity become national security topics. Control of energy increasingly determines control of AI advantage.

7. Big Tech vertically integrates into energy with political cover:
Acquisitions and partnerships get framed as “innovation leadership.” Regulators look the other way in the name of global competitiveness.

8. Public sentiment turns against unchecked AI infrastructure growth:
Not anti-AI. Anti-scale-without-guardrails. The vibe shifts from awe to suspicion.

9. Privacy becomes bipartisan again, briefly:
Left and right disagree on almost everything except one thing: people don’t trust platforms with their data anymore. The agreement doesn’t last, but it shows up.

10. On-device and sovereign AI become geopolitical talking points:
Countries start framing local AI as independence, not innovation. “Digital sovereignty” becomes campaign language.

11. The Trump administration’s AI posture fuels legal and cultural backlash:
Attempts to limit state regulation spark lawsuits and energize opposition. Executive orders lack teeth. The fights move to court.

12. Trump’s age and health stop being whispered about and start being openly debated:
Polling shifts. Media coverage changes tone. Even allies hedge. Voters don’t need proof. They need doubt.

13. A medical or incapacitating scare becomes politically destabilizing:
No announcement. No transparency. Just silence, speculation, and contingency planning.

14. J.D. Vance becomes a serious contingency scenario, not a thought experiment:
Quiet preparations happen. Power brokers plan. Nobody says it publicly.

15. Epstein-related pressure builds into a political breaking point:
Files, testimony, and reporting resurface. Public disgust grows. Defenses weaken. The question isn’t guilt. It’s whether the pressure fractures the base or gets buried in a deal.

16. Antisemitism continues to rise globally, regardless of political intent:
Criticism of Netanyahu’s government remains badly conflated with antisemitism. Messaging fails. Violence and harassment increase before clarity arrives.

17. Women’s rights in Israel become a flashpoint in international discourse:
Far-right governance brings policies that alarm allies. The issue cuts across party lines and complicates traditional support coalitions.

18. Climate messaging reframes around cost of living, not catastrophe:
Democrats talk less about the planet and more about bills, jobs, and health. It’s pragmatic or it loses.

19. Democrats win control of one chamber in the 2026 midterms:
Not a landslide. A rebuke. Inflation fatigue, chaos exhaustion, and priorities misaligned with daily life drive turnout.

20. Republicans retain just enough power to block sweeping reform:
Gridlock persists. It’s not decisive. It’s surgical.

21. Redistricting quietly shapes outcomes more than campaigning:
Maps matter more than messaging. Districts flip without persuasion. Voters feel confused, not defeated.

22. Prediction markets gain mainstream political legitimacy:
Kalshi-style markets get cited on cable news. Odds become talking points. “The market says…” replaces expert panels.

23. Crypto becomes a political canary:
A major Bitcoin drawdown rattles confidence. Markets wobble. Housing softens. Leaders scramble to downplay correlations.

24. Economic anxiety outweighs culture wars for undecided voters:
People stop arguing about ideology and start asking why everything feels harder.

25. Public trust continues to decay, but apathy grows faster than outrage:
While I hope we see more protests and activity, my sense is that we’re going to see more disengagement. The loudest voices don’t represent the majority. They just sound like they do.

26. Politics feels less dramatic, but more dangerous:
What i hope... but not necessarily predict... is more stability. Fewer shocks. More erosion. Institutions don’t collapse, but hollow out. And most people only notice when something finally stops working. We can’t afford to wake up in June and have a complete, irreversible failure. We’re on that track, but it’s time to care more, act fast, and get loud. That’s what I’m here for...

Wishing you and yours all the best for a happy, healthy New Year!

*AI Disclosure: 100% of this was written by me, a human with only light edits and typo corrections from Grammarly. The images are mostly AI-generated using Nano Banana.

Share The Four P's

Subscribe now