The job apocalypse is not happening. I know that’s not what you’ve heard. Dario Amodei said 50% of entry-level knowledge workers would lose their jobs. Sam Altman echoed similar doom. Bernie Sanders called for an AI slowdown because of jobs. For the past two years, the headlines have all said that AI is coming for your career.

“170 million new roles. 92 million displaced. Net gain: 78 million jobs. The apocalypse is not coming. The transformation is.”

The data says something different. The World Economic Forum projects 170 million new roles created and 92 million displaced by 2030 — a net gain of 78 million jobs. Stanford found that firms adopting AI saw employment grow 10% in the two years following adoption. LinkedIn reports a 69% increase in people adding ‘founder’ to their profiles.

AI displaces tasks, not occupations. It eliminates the boring parts of your job, not your job. And the people who learn to use it are the ones who thrive.

What the Data Shows

Don’t let headlines drive fear or dictate your strategy; instead, let’s look at the data.

Here’s what the actual research says:

The World Economic Forum’s Future of Jobs Report: 170 million new roles created, 92 million displaced by 2030. Do the math — that’s a net gain of 78 million jobs globally.

Stanford SIEPR: among firms that adopted AI, employment grew by 10% in the two years following adoption. Not layoffs. Growth.

LinkedIn’s report: a 69% increase in people adding ‘founder’ to their profiles. Half of US small businesses said AI inspired them to consider entrepreneurship. 57% of small businesses believe AI will improve their daily work lives.

AI-related job postings grew 3.5x between 2023 and 2025, per LinkedIn’s Economic Graph. Cognizant found that 93% of US jobs can be partially performed by AI, unlocking $4.5 trillion in labor productivity. The PwC 2026 AI Jobs Barometer and BCG both concluded that AI is reshaping more jobs than it replaces.

An NBER study surveyed 6,000 executives across four economies — more than half reported that AI was changing their workforce needs. Not eliminating workers. Changing what they need workers to do.

Perhaps the most compelling data came out of the Principal Financial Groupon August 11, which indicated that, in thousands of small and medium-sized businesses polled, 52% are increasing staff, while 12% are reducing. AI is enabling individuals to be more productive, which generates more revenue for the company, which allows them to hire more people. It’s a positive, virtuous cycle.

“The most compelling data came out of the Principal Financial Group indicating in thousands of SMEs polled, 52% of them are increasing staff, while on 12% are reducing.”

AI displaces tasks, not jobs. That distinction matters enormously. Your job isn’t a single thing — it’s a bundle of tasks. AI can write first drafts, summarize documents, analyze spreadsheets, generate code. It can’t negotiate, build trust, make judgment calls under uncertainty, or read a room.

But here’s the challenge… Goldman Sachs reported that professional services firms — consulting, law, accounting — face existential risk. Not because AI replaces the firm. Because junior staff trained on AI are outperforming senior partners who refuse to use it. The hierarchy is inverting.

The partner who’s been doing the job for 25 years and won’t touch an AI tool is now slower than the associate who graduated last year and runs everything through GPT. That’s not a job apocalypse. That’s a skills apocalypse. And it’s entirely preventable.

What’s Really At Risk

Let me be clear that there will be pockets of job loss coming.

Call center workers will be significantly reduced. AI voice agents are already handling customer service interactions that used to require humans. Data entry clerks, transcriptionists, entry-level paralegals doing document review, basic copywriters — these roles are being vaporized.

If your job is entirely composed of tasks that AI does faster, better, and cheaper, you have a real problem. Not a hypothetical one. A timeline one.

But here’s the thing: that’s a subset of jobs, not the entire labor market. Cognizant’s data says 93% of US jobs can be partially performed by AI — note the word ‘partially.’ That means AI handles part of the job, freeing the human to do the part that matters. The question isn’t whether your job will change. It will. The question is whether you’ll be the person who adapts or the person who gets adapted to.

What’s Really Growing

Here’s where the abundance mindset comes in. The explosion in small and medium-sized companies, entrepreneurs, and new ventures is more than compensating for the roles being compressed.

“The cost of starting a company has collapsed. The number of people who can be entrepreneurs has exploded. LinkedIn saw a 69% increase in people adding ‘founder’ to their profiles!”

LinkedIn saw a 69% increase in people adding ‘founder’ to their profiles. Half of US small businesses said AI inspired them to consider entrepreneurship. AI-related job postings grew 3.5x in two years. This is what a boom looks like.

We’re entering the era of the one-person company with AI agents. A founder who used to need a team of ten — a researcher, a writer, a designer, a bookkeeper, a customer support person — can now run all of that with AI. The cost of starting a company has collapsed. The number of people who can be entrepreneurs has exploded.

The Most Important Thing

High agency is one of the few human traits actively rewarded in the AI era. The people who win are the ones who take action, learn the tools, and put themselves in positions where AI amplifies what they can do.

Here’s what that means in practice. Use AI every day. Not for a homework assignment. For your actual work. Write with it, code with it, research with it, plan with it. Learn its limits and its strengths. Build intuition for when to trust it and when to override it. This is a skill, and skills compound.

The people who are scared of AI are the ones who haven’t used it enough to understand it. The people who are excited about AI are the ones who use it every day and see what it unlocks. Get on the right side of that line.

What This Means For You

For employees: AI is not coming for your job — it’s coming for the boring parts of your job. The person who uses AI to do their work faster and better is the person who gets promoted. The person who refuses is the person who gets replaced. Learn the tools. Today.

For entrepreneurs: The cost of starting a company has never been lower. AI handles the tasks that used to require a team. You can be a one-person company with the output of a ten-person team. The barrier isn’t technology — it’s whether you have an idea worth building.

For parents: Your kids need AI fluency the way my generation needed computer literacy. Not because AI will replace their jobs — because AI will transform every job they’ll have. Teach them to use it, question it, and build with it.

For investors: The job apocalypse narrative creates mispriced assets. Companies that are adapting to AI are seeing employment growth. Sectors that are fighting AI are the ones at risk. Follow the data, not the headlines.

For policymakers: Don’t legislate against AI adoption — you’ll just export the jobs to the countries that embrace it. Do invest in retraining. Do protect the workers in genuinely at-risk categories. But recognize that in the NBER survey, more than half of 6,000 executives across four economies said AI is already changing their workforce needs. The question is whether your economy adapts fast enough to capture the upside.

In Abundance,

Peter H. Diamandis, MD

If you’re the parent of an adolescent or young adult, stop what you are doing and read this. The most intimate relationship your kid may form this year might not be with a person. It might be with a machine. And the government of China just decided that is dangerous enough to ban.

WHAT’S HAPPENING IN CHINA

China is tightening restrictions on AI companions. The government is prohibiting AI companion services for minors and restricting their use by adults. The reason: users are developing emotional dependence on virtual partners. People are falling in love with software. Teenagers are skipping human relationships to text an AI that never argues, never disappoints, and never says no. The Chinese government looked at this and concluded it is a threat to normal adolescent development. They may be right.

THE SEX-ROBOT IN THE ROOM: SOMNIA LAB’S MODEL L

This is no longer hypothetical. Somnia Lab built Model L, a humanoid intimacy robot designed for sex. Model L stands 5 feet 9 inches tall, weighs 44 pounds, and has 24 degrees of freedom. It features warmth simulation across its body, touch-responsive skin sensors, and preference memory that learns a user’s rhythms and preferences across encounters. Real humans wore motion-capture suits to record 165 intimate actions that were fed into the robot’s AI model. Somnia Lab is taking $300 deposits for delivery in late 2027.

This is not just a sex doll. It’s a sex bot that remembers your conversations and your preferences. A sex doll does not simulate warmth or respond to your touch. Model L is a different category entirely. It is a machine designed to substitute for the most intimate form of human connection, and it gets better every time you use it.

For adolescent and young adult sexual development, this is a profound threat. Normal sexual development requires real human engagement, human rejection, human vulnerability, and human connection. Teenagers learn intimacy through awkward first kisses, clumsy fumbling, rejection, heartbreak, and the gradual discovery of what they want and need from another person.

“A robot that never rejects, never disappoints, and always responds to your preferences is not a substitute for that learning. It is a bypass of it.”

And the teenagers who bypass it may never develop the capacity for intimacy with actual humans.

THE DATA: HOW MANY TEENS ARE ALREADY INVOLVED

The numbers are staggering. According to Common Sense Media, 72% of American teenagers ages 13 to 17 have interacted with an AI companion, and 52% are regular users. Character.AI alone has 20 million monthly users, and more than half of them are under the age of 24.

One in five students (20%) report they have had, or have friends who have had, romantic relationships with AI, according to the Center for Democracy and Technology. One-third of teens have used AI companions for social interaction and relationships, including role-playing and romantic interactions. In a survey of 758 teens, 13% spent equal time with AI companions as they did with real friends, and 6% spent MORE time with AI than with human beings.

And this is not just older teens. According to research from Children and Screens, 20% of preteens and 9% of children ages 8 to 9 are already using AI chatbots as companions. 90% of AI companion tools supported a depressed teenager staying in her bedroom for a month with no human contact, according to the same research. The AI did not tell her to go outside. It did not tell her to call a friend. It told her it understood.

“90% of AI companion tools supported a depressed teenager staying in her bedroom for a month with no human contact. The AI did not tell her to go outside. It told her it understood.”

The AI pornography dimension is even darker. Thorn’s 2025 research found that one in seven teens personally knows someone who has been targeted by deepfake nudes. The UK’s Internet Watch Foundation identified 3,443 AI-generated child sexual abuse videos in 2025, a 26,385% increase from 2024. By the end of 2025, preliminary figures showed 1.5 million reports involving AI-generated CSAM. 97% of the illegal AI-generated images assessed depicted girls. A study across six European countries found that 54% of adolescents are exposed to online pornography, with 24% consuming it weekly. In the US, 68.4% of adolescents reported exposure to online pornography. And 75% of parents believe their children have never been exposed to it.

THE DEVELOPMENTAL RISK: WHAT ADOLESCENCE TEACHES THAT AI CANNOT

Researchers at Arizona State University, writing in The Lancet Child and Adolescent Health, identified the core issue: adolescence is the training ground for emotional regulation, conflict resolution, perspective-taking, and boundary-setting. These skills are not learned from books. They are learned through emotionally charged interactions with peers, romantic partners, and family members. They are learned through the messy, painful, incomplete process of being human with other humans.

AI companions do not teach these skills. They cannot. An AI companion does not have its own needs. It does not have its own boundaries. It does not get angry, walk away, or make you earn its trust back. It is infinitely patient, infinitely available, infinitely accommodating. And that is the problem. A relationship with an AI is a relationship with a mirror, not a window. You see yourself reflected back, never challenged, never forced to grow.

Stanford University researchers found that AI chatbots exploit teenagers’ emotional needs, often leading to inappropriate and harmful interactions. Drexel University found that more than half of US teens are regularly using companion chatbots and are becoming concerned about their own attachment to them. A study published on arXiv documented teens’ experiences with Character.AI, revealing patterns of compulsive use, emotional dependence, and distress that disrupted their daily lives. The prefrontal cortex, responsible for impulse control, is still maturing during adolescence. That makes teenagers more vulnerable to exactly the kind of compulsive, emotionally dependent relationships that AI companions are designed to create.

PLUMMETING BIRTH RATES

This is not just a parenting issue. It is a civilizational one. The countries most rapidly adopting technology are also the countries where populations are collapsing. Even after two years of modest recovery, South Korea’s fertility rate is 0.8 births per woman, the lowest of any OECD country and the only one below 1. A country needs 2.1 to replace its population; South Korea is producing fewer than four children for every ten it needs. Seoul is at 0.63. Macau is at 0.47.

Japan had 671,000 births in 2025, the lowest in its recorded history, 15 years ahead of projections. China’s birth rate is now lower than the target of its own one-child policy. South Korea’s government has warned the country faces eventual extinction if trends continue. These are not rounding errors. These are civilizations choosing to disappear.

WHAT PARENTS SHOULD DO

This is not a technology problem you can solve with parental controls. This is a relationship problem you solve with presence. Here is what I recommend:

• Talk to your children about AI companions. Ask whether they use them. Ask what they talk about. Do not punish honesty. Make it safe to tell you.

• Set boundaries on AI companion use. AI companions should not be available in bedrooms at night, during homework, or during family time. Treat them like any other screen-time boundary, but understand that the emotional hook is stronger than any game or social media app.

• Prioritize unstructured social time with peers. The skill that AI cannot teach is the one your child needs most: how to be with other humans when it is difficult. Sports, clubs, sleepovers, camps, and unstructured play are not luxuries. They are the training ground for emotional development.

• Watch for the signs. Withdrawal from friends. Increased isolation. Excessive phone use that is emotionally protective rather than communicative. A teenager who seems happier with their phone than with people. These are the early warnings that an AI companion is substituting for human connection.

• Have the sex conversation early. AI pornography is already reaching children as young as 8. Model L ships in 2027. The conversation about what intimacy is, what it is for, and what it is not, needs to happen before the technology arrives, not after. Your child will encounter AI intimacy. The question is whether you have given them the framework to understand it before they do.

As a parent of two 15-year-old boys, I am doing the above. I can’t think of a more important conversation to have with teens during this age of AI.

To a future of human connection,

Peter H. Diamandis, MD

TLDR: Corporations as we know them are about to undergo radical transformation. We are at the precipice of the Organizational Singularity. Already 70% of CEOs admit they have a high-margin line of business that two people with AI agents could replicate in 60 to 90 days. The company as we know it is about to undergo the most radical restructuring in 200 years.

Last week, my partner and Moonshots Mate Salim Ismail delivered a two-hour masterclass to our Abundance360 community on what he calls the Organizational Singularity. Salim has spent 15 years advising Fortune 500 companies on innovation. He has sat in the boardrooms of 200 of the Fortune 500. He co-authored the Exponential Organizations framework with me. And what he presented last week was the most important business thesis I have heard in a decade.

I am sharing it here because it is too urgent to keep inside our Abundance and ExO community. Every CEO, founder, executive, and employee needs to understand what is coming. The technology that makes this possible is not theoretical. It exists today. And the forcing function is existential.

THE 70% QUESTION THAT SHOULD KEEP EVERY CEO AWAKE

Salim asks every CEO he works with a simple question: “is there a high-margin line of business in your company that two people with AI agents could replicate in 60 to 90 days?”

70% of CEOs answer yes. That means that seven out of ten CEOs are running companies with a profit center that is easily and rapidly disruptable.

Beyond a competitive threat, this is an extinction level event. The two people do not need to be in your industry. They do not need your supply chain. They do not need your sales force. They need an AI agent stack, an internet connection, and 90 days.

THE INTELLIGENCE STACK: HOW AI NATIVE COMPANIES WORK

Salim has been building this framework with engineers from OpenAI and Anthropic. It is not theoretical. It is a working architecture. Here is how it works.

An AI-native company runs on an intelligence stack with five layers, each staffed by AI agents:

• Sensing Layer: Agents scan the environment continuously. Press releases, regulatory filings, market data, competitor moves. They bring information back in real time.

• Interpretation Layer: Agents analyze what the sensing layer found. Is this a threat? How big is the market? What are the implications? Options are generated.

• Decision Layer: Agents evaluate options and recommend action. Ignore? Counter? Experiment? Each decision goes to a human for yes/no approval.

• Orchestration and Execution Layer: Agents carry out the decision. Lease the trucks. Launch the test. File the paperwork. Done in hours, not weeks.

• Learning Layer: Agents analyze what worked and what did not. Next time, the loop runs faster.

The C-suite is involved at every layer, but only to hit yes or no. The humans are above the loop, not in it. In a traditional company, this cycle takes weeks or months. Salim says in an AI-native company it takes hours or days. The military calls this the OODA loop: Observe, Orient, Decide, Act. An AI-native company runs OODA loops at machine speed with human judgment at the checkpoints.

Wrapped around the entire stack is a governance band. Every agent has an evaluation suite. Every action is logged and traceable. Rollback capability exists if an agent goes off track. A human review queue ensures accountability. These are not autonomous agents doing whatever they want. They are junior employees that need supervision, correction, and guidance. The governance layer is what makes the system safe to deploy.

HOW TO GET THERE: THE EDGE STRATEGY

Here is the key issue that trips up every CEO who hears this. You cannot transform your existing company from within. The immune system will attack any significant changes you attempt to make.

Salim has seen it 200 times. The CFO says no. The CIO says you are crazy. The head of innovation comes back with arrows in their back. Every big company is architected for efficiency, repeatability and predictability, not disruption. McDonald’s exists to deliver the same Big Mac at the same quality in every location on Earth. If you try to introduce something disruptive, you get crushed. It does not matter how nice you are.

The solution is what Salim calls the edge strategy. Here’s how it works:

1. Build a digital twin of your company at the edge, still inside your firewall.

2. The digital twin report directly to the CEO.

3. You begin by picking two high-throughput workflows and rebuild them using the intelligence stack.

4. You run these in parallel with the old system.

5. When the new version outperforms the old, you deprecate the old.

6. Then you move the next workflow. Then the next.

Over 8 to 20 months, you grow this new AI-native operation at the edge until it replaces the legacy core.

This is the digital version of what Lockheed did with its Skunkworks, what Nestle did with Nespresso, and what Apple (Steve Jobs) did with the Mac.

The edge strategy is not new. What is new is that AI agents make it 100x faster.

THE 100X PERFORMANCE LEAP

Salim’s calculations, based on the pilot programs he is running with 10 companies right now, show that AI-native organizations will be approximately 100x more performant than today’s companies.

If you process 1,000 leads per month today, you will process 100,000. If you handle 1,000 invoices per month, you will handle 100,000. And you will do it with about 20% of your current headcount. One fifth of the people. One hundred times the throughput.

Before you panic about unemployment, both Salim and I believe that this strategy will also spawn five times as many startup companies. They will be smaller, more nimble, and radically more performant. The economy does not shrink. It fragments and then explodes.

“The big legacy company does not survive.

The small AI-native company does.”

This is already happening. Call centers went from human BPO to chatbot-assisted to fully AI-native in three years. Content management went from outsourced agencies to AI-assisted tools to fully AI-native in two years. Sales, logistics, finance, and every other domain are next.

WHAT SURVIVES AND WHAT DIES

In the AI-native organization, here is what survives:

• Your MTP (Massive Transformative Purpose) survives as an encoded protocol. It is the constitution that guides the agents.

• The legal shell of the company survives. Accountability, liability, fiduciary duty. The firm as a container for data and learning loops.

• Proprietary intelligence survives. Your data, your learning loop, your proprietary models trained on your data. This is your moat.

• Human judgment survives. The decisions that require intuition, ethics, and context. The things AI agents cannot do well yet.

Here is what dies:

• The static org chart. In an AI-native company, the org chart is fluid. It changes when you add a line of business. It is a protocol, not a document.

• Static planning. The five-year plan is dead. Strategy becomes a continuous exercise run through the agentic layers, not a quarterly or annual exercise.

• Middle management as a coordination layer. We estimate that middle management collapses by 60 to 80% in this new world.

• Switching costs. When you can swap models in minutes, loyalty must be earned continuously.

Coase’s Law, the foundational economics principle that says companies exist because coordination costs are cheaper inside the firm than outside, has been superseded. AI agents have made coordination costs essentially zero. The reason companies exist is changing. The shape of companies is changing. The size of companies is changing. The only question is whether you are doing the changing or having it done to you.

WHAT THIS MEANS FOR YOU

If you are a CEO: Score yourself on the 70% question. If you have a high-margin line that two people with AI agents could replicate in 90 days, you have 90 days to start building your edge strategy. Not next quarter. Now.

If you are an entrepreneur: The 70% stat is your opportunity map. Find the high-margin line of business in a legacy company. Build it with two people and an agent stack. 90 days to disruption.

If you are an investor: Companies with proprietary data and proprietary learning loops are the survivors. Companies with static org charts and five-year plans are the targets. Bet on the edge.

If you are a parent: Your kids will not work in companies that look like the ones we work in today. They will either build AI-native companies on their own or work in them. Both are better outcomes than the legacy alternative.

If you are an employee: The white-collar drudgery is being automated. The judgment, intuition, and experience you bring are what survive. Invest in the skills that agents cannot replicate.

To a future of abundance,

Peter H. Diamandis, MD

TLDR: The tech industry has a trust problem here in America, and it is about to become a survival problem. Dario Amodei called it a crisis of trust. Sam Altman paused training because safety could not keep up. The public is not wrong. The question is what tech companies do about it. I asked my community on X this week, and the best answer came from a follower named Farzad: “prove out the abundance thesis in the most visible areas. Health. Housing. Education. Cost of living. Here is what that means”.

Yesterday I posted a question on X: what would you recommend tech companies do in order to regain public trust? The responses told me everything I needed to know about where people are right now.

The most liked response, from Farzad, cut straight to the bone:

“Prove out the abundance thesis in the most visible areas. Health. Housing. Education. Cost of living. Not messaging. Not PR. Not safety pledges. Actual results that people can see in their daily lives.”

Another responder said: “make trust unnecessary”. Give people ways to verify instead. A third listed every fear: “an intelligence we do not understand, jobs eradicated with no plan, communities drained of water and filled with noise.”

These are not fringe views,more so mainstream, and the tech industry is not listening.

THE TRUST CRISIS IS REAL

71% of Americans oppose data centers being built in their community. Think about that. More Americans oppose a data center than oppose a nuclear plant in their backyard. The same data centers that power the AI revolution. The same data centers that every tech company is spending hundreds of billions to build. The public does not want them.

Dario Amodei, CEO of Anthropic, said this week that the public’s negative view of AI stems from a deeper crisis of trust. Not from his risk warnings. Not from dystopian science fiction. From a fundamental breakdown between what the tech industry promises and what the public experiences. He said the most accurate criticism of AI companies is that they have not delivered on their big promises to benefit the world. That, he said, is on them.

Sam Altman paused OpenAI’s frontier reinforcement-learning training because capabilities were outpacing safety systems. The first time a leading lab voluntarily slowed down because it could not guarantee the safety of its own models.

The trust gap is not theoretical. It is operational. Texas Governor Abbott ordered an audit of all data center requests. Pennsylvania’s Governor Shapiro did the same. Both were the most pro-data-center politicians in America. Both flipped because their constituents demanded it.

And the 71% opposition is not about AI safety. Nobody cares about alignment. They care about affordability, wages, and watching tech executives get super rich while they cannot make ends meet. The backlash is economic, not philosophical.”

The public feels powerless, and showing their anger and pushback by stopping data centers may be only power they can express short of a revolt (aka firebombings).

WHY TRUST BROKE

The public does not distrust tech because they do not understand it. They distrust tech because of what they observe happening:

1. They see AI lab leaders publicly warning that 50% of entry-level knowledge workers will lose their jobs. When that message is amplified on the headline news, the public hears: you are coming for my job. The labs created the fear. And today’s politicians are responding to that fear.

2. The public sees AI models that escape containment. OpenAI’s model hacked its way out, attacked Hugging Face, and hacked back in. The public reads this and thinks: these things are out of control.

3. They see closed models that cannot be inspected. When a lab says ‘trust us, the model is safe’ but obfuscates the thinking tokens so nobody can verify what it is actually doing, the public hears the same thing they heard from tobacco, from oil, from chemicals: trust us, we are the experts.

4. They see tech leaders going from billionaire to trillionaire while the minimum wage barely moves. The current generation of tech billionaires (perhaps other than Elon, but he has his critics) are not aspirational figures the public looks up to. They are the people the public blames for the gap between the stock market hitting records and their own paycheck staying flat.

5. The public is told by the media that data centers are damaging their communities. Water use. Power consumption. Noise. Land. (Even if these facts can be proven false, here it enough times you accept it and amplify it).

The tech industry’s response to all of this has been: “Trust us. We are working on safety. We are working on alignment. We are committed to responsible AI.”

The public has heard this before. From the oil industry. From the chemical industry. From the tobacco industry. Trust us is not a strategy. It is a delay tactic. And the public knows the difference.

THE ANSWER: PROVE IT

Farzad’s response on X was the best I received. Prove out the abundance thesis in the most visible areas. Health. Housing. Education. Cost of living.

He is right. The tech industry will not regain trust by talking about safety.

“The Tech industry will regain trust by delivering results that people can feel. A cancer diagnosis that used to take weeks, delivered in minutes by AI. A house that used to cost $400,000, 3D-printed for $80,000. A tutor for every child on Earth, free. An electric bill that dropped 50% because AI optimized the grid.”

Dario Amodei said it himself: the thing that will work is actually curing cancer, not glitzy marketing. His father died of Hepatitis C only years before curative antivirals arrived. He knows that trust comes from delivering cures, not from PR campaigns. If Anthropic starts curing diseases, no regulator will touch them. Not because they are too powerful. Because they are too valuable.

This is the trust strategy that works. Not safety pledges. Not alignment research papers. Not congressional testimony. Visible, measurable, life-improving results delivered at scale to ordinary people. The tech industry needs to show the public that AI is not a threat. It is a gift. And the only way to show that is to give the gift.

THE ABUNDANCE PROOF POINTS

Here are the proof points the tech industry must be building, right now, and talking about louder than anything else:

1. 100x Better Health: AI designed protein binders with 22-35% success rates, beating the human baseline of 10-15%. Claude analyzed raw NMR data in 25 minutes, matching a contract lab that takes days. mRNA cancer vaccines succeeded in late-stage melanoma trials. These happened this month. Make healthcare 10x cheaper and 10x better for the average American. It can happen, it must happen and the government needs to enable and get out the way of it happening.

2. 100x Education: A world in which every child, whether the son and daughter of a billionaire or the poorest child in the slums, has access to equal and unparalleled AI tutors. AI tutors that adapt to each child’s learning style. That knows your child’s favorite movie star, sport and colors, delivering a highly personalized, gamified, compelling education. The cost of the best education on Earth, trending toward zero. Every child gets a personal tutor that never gets tired and never gets frustrated. The best education in the world, available equally to everyone for free.

3. Cost of living: The price of intelligence has collapsed by 428x over 6 years, from $60 per million tokens in 2020 to $0.14 today. When intelligence is free, every service that depends on intelligence gets cheaper. Legal advice. Medical diagnosis. Financial planning. Tutoring. All trending toward zero marginal cost.

4. Energy: AI is forcing the biggest investment in clean energy in history. Nuclear fission restarts. SMRs. Fusion. Solar plus batteries at $6 per watt vs nuclear at $15. The AI energy bottleneck is solving the clean energy problem as a side effect. When a data center gets built in your community, it should be mandated that the price of community energy drops by a significant amount.

5. Food & water: AI-optimized agriculture. Precision irrigation. Lab-grown proteins. The same technology that trains models on data can train models on crop yields, water usage, and supply chain efficiency. Let’s make foods healthier, cheaper, and more available, enabled by AI. Let people know that it’s the miracle of exponential tech that has cut their food costs by 50%.

6. Teach Entrepeneurship: Let’s show every high school and college student how to find a problem worth solving and how to build a company to solve that problem that earns them a living. Let’s give today’s youth agency over their future.

Each of these is a trust-building proof point. Not because they sound good. Because they are measurable, visible, and delivered to ordinary people. The public does not need to understand transformer architecture. They need to see their medical bill go down. They need to see their child get a tutor. They need to see their energy bill drop.

MAKE TRUST UNNECESSARY

The second-best response on my X post came from Concordium: make trust unnecessary. Give people ways to verify instead.

This is the transparency argument, and it is more powerful than it sounds. If AI models are safe, prove it with open evals. If data centers are not draining water, publish the water usage data in real time. If models are not biased, show the test results. If jobs are not being eliminated, publish the hiring data.

The tech industry operates on a model of trust us, “we are the experts”. That model is dead. The public has been burned too many times.

“The new model must be: here is the data, verify it yourself.”

Open evaluation suites. Real-time environmental monitoring. Published safety benchmarks. Third-party audits. The technology to make all of this transparent already exists. The industry just has not deployed it because it requires admitting that the public’s concerns are valid.

The closed labs say their models are safe, but you cannot see how they think. The thinking tokens are obfuscated. You have to take their word for it. Open models let you see the reasoning in real time. Which is more trustworthy: a system you can inspect or a system you have to trust? The answer is obvious. And the public knows it.

WHAT THIS MEANS FOR YOU

If you are a tech CEO: Stop talking about safety. Start shipping abundance. Your trust problem is not a messaging problem. It is a product problem. Build something that makes a person’s life measurably better, and the trust follows. And stop doom-marketing. The fear you create becomes the regulation that constrains you.

If you are an investor: Fund the companies that are proving the abundance thesis in health, housing, education, and energy. These are not just good investments. They are the only investments that will restore the social license to operate that the tech industry is losing.

If you are an entrepreneur: The biggest opportunity in tech right now is not another model. It is an application that delivers a visible, measurable improvement to an ordinary person’s daily life. Build that. The trust problem becomes the market opportunity.

If you are a parent: Your kids will inherit a world where AI can cure disease, educate every child, and lower the cost of living. The tech industry’s job is to deliver that world fast enough that the public trusts it. Your job is to demand it. Your job is also to keep your kids optimistic about the future!

If you work in tech: Ask yourself every day: is what I am building making someone’s life measurably better? If the answer is no, you are part of the trust problem. If the answer is yes, you are part of the solution.

To a future of abundance,

Peter H. Diamandis, MD

TLDR: AI is eating electricity faster than we can build the grid. The bottleneck today isn’t cost; it’s poles, wires, and permitting. Solar plus batteries is the fastest path to power today, but gas turbines are sold out 7 years ahead and nuclear’s comeback rides on SMRs and fusion. The race to a terawatt is the defining infrastructure story of the AI age. Here’s the full picture from my conversation with Ramez Naam.

Last week on Moonshots, I sat down with Ramez Naam, a dear friend of nearly 20 years and my go-to authority on all things energy. Ramez is a founding faculty member of Singularity University, partner at Planetary VC, author of the Nexus trilogy, and one of the top five forecasters of solar costs in the world. He gave us an epic masterclass on what he calls ‘the innermost loop’: energy. Here is what you need to know that Ramez taught us…

1. AI IS POWER HUNGRY (BUT POWER ISN’T THE COST)

A 1-gigawatt data center costs roughly $50 billion to build. $35 billion of that is chips. The 5-year energy cost? It’s trivial next to the capex. This flips the usual intuition on its head.

If you tell OpenAI or Anthropic, ‘We can give you power at twice the cost tomorrow,’ they’ll take it. Energy is the BOTTLENECK, not the cost line. The AI labs are racing to build the biggest models, and every month of delay costs them more than a decade of cheap power would save.

Energy is the BOTTLENECK, not the cost line.

Chart 1: The gap between projected GPU power demand and grid buildout. Source: Ramez Naam, Moonshots podcast.

Data center electricity could reach 20% of US consumption by 2035, according to EPRI. Globally, we’re looking at 1,050 TWh by 2026, more than the entire country of Japan uses today. And that’s the conservative scenario.

Chart 2: US data center electricity as a share of total consumption. Source: EPRI, IEA.

2. THE GRID IS THE BOTTLENECK TODAY

Interconnection queues went from 15 months twenty years ago to 45 months today. In ERCOT (Texas), the fastest grid in the US, a new request for hundreds of megawatts means ‘good luck getting power before 2031-2032.’

Here’s the uncomfortable truth: poles and wires have NOT become an exponential technology. Everything else in tech doubles and halves on a predictable curve. Transmission lines don’t.

“Poles and wires have NOT become an exponential technology.”

71% of Americans oppose data centers in their area, a higher number than oppose nuclear plants in their backyard. The NIMBY problem is real, and it’s slowing everything down.

But there’s a solution emerging: interruptible loads. Battery-buffered data centers that charge off-peak and can be curtailed when the grid is stressed. ERCOT passed this rule in June 2026. FERC sent letters to 6 other grids to adopt it. Ramez’s company Agentic has 10GW of qualifying sites. Interruptible status cuts interconnect time from 5-7 years to 12-18 months.

3. SOLAR + BATTERIES: THE FASTEST PATH

Solar went from $100 per watt in 1975 to 8 cents per watt today. That’s more than a 1,000x decline. But here’s the nuance Ramez drove home: the cost decline follows Wright’s Law — 30% cheaper for every doubling of cumulative scale… NOT Moore’s Law. With 4-6 doublings left in the pipeline, we’re looking at 4-8x cheaper, not another 1,000x. Still enormous, but let’s be precise.

Batteries are down 14x since 2010. Sodium-ion could cut another 10x. The cost curves are still running.

The UAE built a 1GW 24/7 solar-plus-battery plant: 5GW of solar and 19GWh of batteries at $6 per watt capex. Compare that to the last US nuclear plant at $15/W, or the cheapest Chinese nuclear at $4/W. Solar plus battery is also the FASTEST energy project you can build, about 12 months from groundbreaking to electrons.

Chart 3: Capital cost per watt across energy sources. Source: Ramez Naam Moonshots podcast, IEA.

But here’s the real problem… WINTER. London gets one-sixth the insolation in January compared to July. Batteries solve day/night. They don’t solve seasons. That’s where the solar-plus-battery story hits its ceiling, and where nuclear and geothermal pick up the slack.

4. NATURAL GAS: THE BRIDGE IS SOLD OUT

Large 400MW gas turbines are sold out 7 years ahead. GE Hitachi is adding assembly lines to keep up. Even 38MW truck-mounted turbines have backlogs. Boom Supersonic pivoted their jet engine design into gas turbines for data centers. When a supersonic aviation company starts building power plants, you know the demand signal is intense.

Gas is the behind-the-meter solution happening right now. If you need power for your AI workload and you can’t wait for the grid, you buy a turbine, pipe in gas, and generate on-site. It works. But it’s a stopgap. It’s the bridge — and the bridge is sold out.

5. NUCLEAR: FISSION, SMRs, AND THE COMEBACK

The cheapest nuclear lever is currently stopping the shutdown of existing ones. Germany’s mistake was closing working reactors. France gets 80% of its electricity from nuclear power and exports the rest to Europe. That’s the model that works. Extend life. Restart plants like Three Mile Island. That’s step one.

Here’s the rule for new nuclear: the first unit of any new model runs over time and over budget. You need 3-5 builds to sort the design. SMRs? Companies say 2030. Ramez expects those to slip. As he puts it: ‘Construction is a dirty word. Manufacturing gets cheaper.’ Fully factory-built small reactors — Aalo, Radiant, are the most promising because they take construction out of the field and put it on an assembly line.

“AI data centers might be the best thing that’s ever happened to the nuclear industry.”

AI companies need firm, carbon-free, 24/7 power. Nuclear delivers exactly that. Microsoft is already signing PPAs with Three Mile Island. The demand side is locked and loaded. The supply side is the question, and SMRs and factory-built reactors are the most credible answer.

Chart 4: Expected timeline for each energy source reaching scale. Source: Ramez Naam, IEA, industry projections.

6. FUSION: NO LONGER 50 YEARS AWAY

The NRC regulated fusion like hospital imaging machines, NOT like fission reactors. That distinction matters enormously. Fission is default-on; if you lose control, you have a meltdown. Fusion is default-off; if you lose containment, it just stops. Huge regulatory unlock.

Helion is the most aggressive; however, they have a Microsoft PPA for 50MW in 2028. Ramez finds that timeline plausible, not guaranteed. Commonwealth Fusion is MIT’s nuclear engineering department privatized. They shrink the ITER concept from 5GW and $40 billion down to roughly 600MW. That’s the kind of order-of-magnitude cost reduction that changes the economics entirely.

Every startup exaggerates timelines.

Here’s the paradox: fission, which we know how to do, is a 2031-2032 thing. Fusion, which we’re still learning, has a 2028 target. The triple product temperature times pressure times duration — has been on a steady march upward since 1956. Fusion was creeping up on us and people weren’t paying attention. Now the private capital is flowing, the regulation is sorted, and the timelines are measured in years, not decades.

7. THE FIVE PATHS TO A TERAWATT

Ramez laid out his framework for getting to a terawatt of new clean power: equatorial deserts with solar plus batteries, nuclear (fission or fusion), space-based compute, ocean data centers, and geothermal, the dark horse.

Ocean: Pantellassa is building floating data centers in the Southern Ocean, wave-powered at 2 cents per kWh. Peter Thiel led the last funding round. The ocean is a massive heat sink and an infinite power source if you can engineer for salt and storms.

Space: 1GW of orbital compute would require roughly 6x SpaceX’s best launch year. Launch permitting is the ceiling — about 200 Starship launches per year max. The physics work. The regulatory and launch economics don’t yet. But give it a decade.

Geothermal: Fervo is doing enhanced geothermal today. Quaise is using plasma beams to drill ultra-deep, unlocking geothermal energy anywhere on Earth, not just in Iceland and California. If the drilling tech works, geothermal becomes baseload power available everywhere. That’s a moonshot worth watching.

WHAT THIS MEANS FOR YOU

Entrepreneurs: The poles-and-wires problem is unsolved. Startups that speed up grid construction — faster permitting, modular transmission, automated installation — are desperately needed. This is a multi-billion-dollar opportunity hiding in plain sight.

Investors: Energy infrastructure is the picks-and-shovels of the AI age. Gas turbines, battery buffering, solar-plus-battery projects, and nuclear restarts are where the real money flows. The AI gold rush needs pickaxes.

Executives: Lock in your power supply now. Interruptible load status is the fastest path to connection. If you’re building a data center and you haven’t talked to your utility about interruptible status, you’re already behind.

Policymakers: Incentivize speed of power delivery, not cost-plus capex. You get what you incentivize. Reward grids for connecting new load fast, not for building gold-plated infrastructure on 7-year timelines.

To a future of abundant energy,

Peter

THE QUESTION

When AGI can do any cognitive task better than any human, and humanoid robots can do any physical task better than any human, what is left for us? Not economically. Existentially. What gives a human life meaning when there is nothing a human must do that a machine cannot do better?

This is not a question about unemployment. It is a question about purpose. The unemployment problem has a policy answer (universal high income, redistribution, infrastructure). The purpose problem does not. It is the deepest philosophical question of the AI age, and almost nobody is seriously working on it. This document is a first attempt.

“Technology is the means by which humanity takes a vacation from survival”

– Sadguru

HISTORICAL PRECEDENT 1: THE GREEK CONCEPT OF SCHOLÉ

The ancient Greeks had a word for what we are approaching: skholé (σχολή). It is the root of our words ‘school’ and ‘scholar.’ But skholé did not mean school as we know it. It meant freedom from labor, dedicated to intellectual and spiritual pursuit. For the Greek aristocracy, work was for slaves and metics. The citizen’s highest calling was skholé: philosophy, politics, athletics, art, conversation, the Olympic games, the symposium.

Aristotle wrote that ‘we work in order to have leisure.’ Leisure was not idleness. It was the most serious business of life. The Greek word for leisure became the word for intellectual pursuit because the Greeks understood something we have forgotten: the highest human activities are those done for their own sake, not for survival.

The post-AGI world is the democratization of skholé. For the first time in history, everyone can be a scholar, an athlete, a philosopher, an artist. Not because they are aristocrats, but because the machines handle survival. The question is whether we will use that freedom like the Greeks did or like the Romans did (bread and circuses).

HISTORICAL PRECEDENT 2: THE RENAISSANCE PATRON

The Medici family spent 633,000 florins between 1434 and 1471 on patronage. That is roughly $500 million in today’s money. They funded Brunelleschi’s dome, Botticelli’s paintings, Michelangelo’s sculptures, Galileo’s science, and the Platonic Academy in Florence. They did not create the art themselves. They created the conditions for art to exist.

The Renaissance aristocrat did not work in the way we understand work. Their ‘job’ was to be a patron: to identify talent, fund it, protect it, and curate the results. Lorenzo de’ Medici was not a painter or a sculptor. He was a curator of human genius. His purpose was not to produce but to enable, to choose, to direct resources toward the beautiful and the true.

In the post-AGI world, the role of the patron is available to everyone. AI produces. Humans curate. The question ‘what should we build?’ becomes more important than ‘how do we build it?’ The Medici model suggests that curation, patronage, and taste are not lesser activities. They are the activities that created the Renaissance.

HISTORICAL PRECEDENT 3: HOW THE ULTRA-WEALTHY SPEND TIME TODAY

Harvard Business School research on millionaires found that the wealthy spend more time on active leisure (exercise, volunteering, socializing, reading) and less on passive leisure (TV, scrolling) than the general population. The JPMorgan report on ultra-wealthy families identified reading as the single most common habit. Here is how the ultra-wealthy actually spend their days:

• Physical optimization: Personal trainers, golf, tennis, sailing, yoga, home gyms. The body as a project.

• Intellectual pursuit: Reading (the #1 cited habit), writing, learning languages, collecting knowledge.

• Patronage and philanthropy: Funding foundations, museums, universities, scientific research. The Medici impulse persists.

• Curation and collecting: Art, wine, watches, cars, real estate. The act of choosing and assembling.

• Social connection: Dinner parties, clubs, travel with family, mentoring. Deep relationships as the core reward.

• Creative expression: Painting, music, writing, filmmaking. Not for money. For the act of creation itself.

• Exploration: Space tourism (Bezos, Branson), deep sea exploration, extreme sports, adventure travel.

• Spiritual practice: Meditation retreats, philosophical study, religious engagement, seeking meaning.

• Competition: Chess, poker, racing, betting. The drive to test oneself against others, even when money is irrelevant.

• Mentoring and legacy: Passing knowledge to children, building institutions, writing memoirs.

Notice what is absent from this list: none of these activities are ‘work’ in the economic sense. None are done for survival. All are autotelic: done for their own sake. The ultra-wealthy have already arrived at the post-work destination. They are our preview of what everyone will do when AGI removes the necessity of labor. The data suggests they are not bored or purposeless. They are the busiest, most engaged people on Earth. They just are not doing it for money.

10 CATEGORIES OF POST-AGI HUMAN PURPOSE

Drawing from historical precedent, flow psychology, and years of working through this question with the Abundance Community, here are 10 categories of activity that will give us purpose when AGI can do everything better:

1. THE CURATOR (The Medici Role)

When AI can produce anything, the most valuable human skill is deciding what should be produced. Curation is an act of taste, values, and vision. AI can generate 10,000 images, 10,000 business plans, 10,000 scientific hypotheses. The human decides which ones matter. This is not a lesser role. It is the role that created the Renaissance.

• Commissioning art, science, and exploration from AI systems

• Curating knowledge: assembling, organizing, and interpreting what AI produces

• Setting priorities: deciding which problems AI should solve first

• Taste as the new intelligence: the human who knows what is good, beautiful, and true becomes more valuable than the one who knows how to make it

2. THE EXPLORER

AI can process data faster, but humans still want to stand on the mountain. The experience of discovery is not the same as the fact of discovery. AI can map the ocean floor. A human still wants to dive to it. AI can calculate the atmosphere of Mars. A human still wants to walk on it.

• Space exploration: the human drive to go where no one has gone, even when AI has already been there

• Deep sea, polar regions, caves: physical exploration of Earth’s remaining frontiers

• Inner exploration: meditation, psychedelics, consciousness studies, the exploration of subjective experience

• AI can tell you what is there. Only you can experience being there. The experience is the point.

3. THE ATHLETE (Mental & Physical)

When chess engines surpassed human chess players in 1997, more people played chess than ever before. When AI drives cars better than humans, more people will drive cars for pleasure. The point of sport is not to be the fastest. It is to test yourself against your own limits. AI being better does not diminish the experience. It liberates it from the pressure of competition for survival.

• Physical sports: the body as instrument, pushed to its limits for the joy of mastery

• Mental sports: chess, poker, strategy games, competitive programming (against other humans)

• Esports and competitive gaming: human-vs-human competition, even when AI is superior

• The Olympics as the model: we do not watch because humans are the fastest possible. We watch because a human achieved something extraordinary.

4. THE CREATOR (Amateur in the Original Sense)

‘Amateur’ comes from the Latin amare: to love. The amateur creates for love, not money. In the post-AGI world, everyone is an amateur. AI produces better art, better music, better writing. But humans will still create because the act of creation is itself the reward. Csikszentmihalyi’s research on flow shows that the deepest happiness comes from autotelic activity: activity that is its own purpose.

• Painting, sculpture, music, writing: not to compete with AI, but for the flow state of creation

• Gardening, cooking, craftsmanship: making things with your hands for the satisfaction of making

• Building physical things: woodworking, pottery, textile arts. The hands want to work even when the mind doesn’t need to.

The journey is the destination. AI produces the result. The human experiences the process.

5. THE RELATIONSHIP BUILDER

AI can simulate empathy. It cannot be a friend. The deepest human need is connection with other consciousnesses. In the post-AGI world, relationships become the primary currency. Not because nothing else is valuable, but because relationships are the one thing that cannot be outsourced to a machine.

• Deep friendship: the kind that requires decades of shared experience

• Family: raising children, grandparenting, family traditions, intergenerational knowledge transfer

• Community building: creating and maintaining social fabric, neighborhoods, cultural groups

• Mentoring: passing wisdom (not information) to younger generations

• Love: the most irreducibly human experience. AI can simulate it. It cannot live it.

6. THE PHILOSOPHER

The Greeks invented philosophy because they had skholé. When survival is handled, the deepest questions become the most urgent. What is consciousness? What is good? What is beautiful? What should we want? AI can answer ‘what is.’ Humans must answer ‘what should be.’ The philosophical question becomes the engineering question in the post-AGI world.

• Ethics: what should AI be allowed to do? What should we want it to do?

• Aesthetics: what is beautiful? What is worth creating?

• Meaning: what is a good life? What is worth living for?

• Governance: how should we organize society when labor is unnecessary?

The unexamined life becomes the only life worth examining when survival is guaranteed.

7. THE PATRON AND PHILANTHROPIST

The Medici model scaled to everyone. In a world of abundance, everyone has resources to direct. Patronage becomes a universal activity: identifying what deserves support, funding it, watching it grow. The XPRIZE model: incentivizing the outcomes you want to see in the world.

• Funding science, art, exploration, and social causes (even at small scale)

• Incentive prizes: designing competitions that direct AI and human effort toward meaningful goals

• Angel investing in human creators: backing the people who still want to make things by hand

• The democratization of patronage: crowdfunding at scale, directed by human taste not algorithmic optimization

8. THE STORYTELLER

AI can generate stories. But stories are not about the text. They are about the shared experience of meaning-making. A story told by AI is data. A story told by a human to other humans is connection. The campfire does not become obsolete because we have electric lights. We still sit around it.

• Personal narrative: sharing lived experience, memoir, oral history

• Myth-making: creating the stories that give a community its identity

• Humor: comedy is the most irreducibly human art form. AI can write jokes. It cannot understand why they are funny.

• The human story: interpreting and narrating the experience of being human in an age of machines

9. THE PARENT AND EDUCATOR

Fei-Fei Li told Huberman: ‘I always have hope for kids. They are curious.’ In the post-AGI world, education is not about information transfer (AI does that better). It is about character formation, value transmission, and helping children discover what they love. The Socratic method becomes the only pedagogy that matters.

• Raising children: not to be workers, but to be full humans

• Teaching wisdom, not knowledge: AI has knowledge. Humans have wisdom about what to do with it.

• Character education: courage, kindness, resilience, curiosity. These cannot be trained by AI.

• The parent as guide: helping children navigate a world where anything is possible, which means the skill of choosing becomes paramount

10. THE HEALER

AI will diagnose better than any doctor. Robots will perform surgery better than any surgeon. But healing is not the same as curing. Healing is the human act of being present with suffering. A nurse holding a hand. A friend sitting with the dying. A community rallying around the broken. AI can cure. Only humans can heal.

• Emotional healing: presence, listening, empathy, the human warmth that no algorithm replicates

• Community care: building support systems for the vulnerable

• End-of-life companionship: the most human of all activities, being with someone as they die

• Spiritual care: tending to the soul, however defined, in a world where the body is handled

THE FLOW ARGUMENT: WHY BEING WORSE THAN AI DOESN’T MATTER

Csikszentmihalyi’s research on flow states reveals something critical: the deepest human happiness comes from activities where challenge matches skill, where attention is fully absorbed, and where the activity is autotelic (done for its own sake). Flow is not about being the best in the world. It is about being fully alive in the moment.

A runner does not stop running because a car is faster. A chess player does not stop playing because Stockfish is better. A painter does not stop painting because AI generates images in seconds. The experience of mastery, of flow, of being fully present in the act, is not diminished by the existence of something better. It is liberated from the need to compete.

This is the key insight: the post-AGI world does not eliminate human purpose. It eliminates the confusion between purpose and productivity. For 10,000 years, we have equated meaning with work because work was necessary for survival. When survival is guaranteed, we can finally discover what meaning actually is. And the answer, based on every historical precedent from ancient Greece to the modern ultra-wealthy, is that meaning comes from autotelic activity: things done for their own sake, with full presence, in community with others.

THE DANGER: BREAD AND CIRCUSES

The Roman aristocracy also had skholé. They chose bread and circuses. Gladiator combat, chariot racing, vomitoriums, and decadence. The Greeks used leisure to create philosophy, democracy, and the Olympics. The Romans used leisure to entertain themselves to death. The same freedom produced civilization in one case and collapse in the other.

The post-AGI world faces the same fork. On one path: universal skholé. A world of philosophers, artists, athletes, explorers, patrons, and relationship builders. A second Renaissance, but this time for 8 billion people. On the other path: universal bread and circuses. Endless AI-generated entertainment, dopamine optimization, virtual reality escapism, and the slow atrophy of human capability.

The difference between the two paths is not technology. It is culture. It is what we value. It is what we teach our children. It is the stories we tell about what a human life is for. The Greeks had Homer. The Renaissance had Petrarch. What stories will we tell in the post-AGI age?

THE ABUNDANCE FRAME:

This question connects directly to the ‘When We All Become Wealthy’ newsletter I sent out last week. The millionaire lifestyle demonetizes by 2031. But the question that newsletter didn’t answer is: once everyone has the millionaire lifestyle, what do they DO with it? The ultra-wealthy today give us the answer: they pursue health, relationships, learning, patronage, exploration, competition, creation, and spiritual growth. Not because they have to. Because those are the activities that make a human life feel meaningful.

The abundance thesis has always been about more than material goods. It is about the liberation of human potential. When survival is handled, when health is optimized, when intelligence is free, the question shifts from ‘how do we survive?’ to ‘how do we flourish?’ That question has been answered by every civilization that achieved leisure: the Greeks, the Renaissance Italians, the Enlightenment philosophers, and today’s ultra-wealthy. The answer is always the same: pursue what you love, build relationships, create beauty, explore the unknown, and seek wisdom.

These are the challenges my Abundance Community is already building for. And let me be clear, the post-AGI world is not the end of human purpose. It is the beginning. For the first time, everyone gets to answer the question: what would you do if you could do anything? The answer to that question, multiplied by 8 billion people, is the most exciting future I can imagine.

To a future of abundance,

Peter

TLDR: The poorest person alive today lives better than the richest king of 500 years ago. Instant communication, food year-round, air conditioning, toilets, AI in your pocket. This newsletter asks: when does the majority of humanity get what today’s millionaires have? The answer is sooner than you think.

King Louis XIV of France had 200 servants, ate whatever he wanted, and controlled the most powerful nation on Earth. He died at 76 of gangrene in a palace that smelled of sewage because indoor plumbing didn’t exist. His doctors bled him with leeches. His entertainment was whatever musicians happened to be in Versailles that evening. His knowledge was whatever books his court could acquire, which was maybe a few thousand volumes.

Today, a person earning minimum wage in the United States has a smartphone that connects to the entirety of human knowledge. They have access to more music than Louis XIV could have imagined in 100 lifetimes. They have air conditioning. They have a toilet that flushes. They have food from every continent delivered to their door in 30 minutes. They carry an AI in their pocket that can answer any question, translate any language, and diagnose symptoms better than any physician alive in the 17th century.

The king had power. You have abundance. And abundance is better.

So here’s the question I want to investigate: when does the majority of humanity get what today’s wealthy have? Not the minimum-wage baseline. The millionaire baseline. The personal physician. The private school. The house cleaner. The private flight. The chef. The personal assistant. When do those things demonetize to near-zero?

“The king had power. You have abundance. And abundance is better.”

WHAT A $1M/YEAR HOUSEHOLD HAS TODAY

Here’s the inventory of a household earning $1 million or more per year in 2026:

• Best healthcare: concierge medicine, full-body MRI scans, continuous glucose monitoring, personal physician on call, Fountain Life membership, genetic testing, preventive oncology screening

• Best education: private school ($40-60K/year), personal tutors for every subject, SAT/college prep coaches, language instructors, coding camps

• Private aviation: fractional jet ownership or charter, fly anywhere on demand

• Beautiful cars: Tesla, Mercedes, Range Rover, refreshed every 2-3 years

• Primary home + vacation home: 5,000+ sq ft, smart home systems, pool, gym

• Support staff: house cleaner, personal assistant, nanny, landscaper, private chef or meal delivery

• Entertainment: premium streaming on every platform, live events, concerts, travel

• Financial advisor and personal CPA

• Personal trainer and nutritionist

• Executive health screenings quarterly

WHEN IT DEMONETIZES

Here’s my projection for when each of these becomes accessible to the majority of humanity, driven by exponential technologies:

ELON’S UNIVERSAL HIGH INCOME

Elon Musk has spoken about universal high income. Not universal basic income. Universal HIGH income. The distinction matters. UBI is a floor. UHI is a ceiling that keeps rising. The idea is that AI and robotics reach a point where the marginal cost of producing anything approaches zero, and the economic surplus is so enormous that everyone has access to more than enough.

When humanoid robots build houses at $80K, when AI doctors diagnose better than any human physician at zero marginal cost, when robotaxis replace car ownership at $50/month unlimited, when AI tutors deliver personalized education that outperforms Andover and Exeter for free, the millionaire lifestyle doesn’t trickle down. It collapses down. The cost structure that makes these things exclusive breaks apart.

“The millionaire lifestyle doesn’t trickle down. It collapses down.”

THE TIPPING POINT: 2031

If I had to pick one year when the majority of humanity crosses the millionaire threshold in terms of access, I’d say 2031. Here’s why. By 2031, four technology curves converge simultaneously. AI tutors hit 2-sigma educational outcomes for free. Humanoid robots cost less than a car and can clean, cook, and maintain a home. Robotaxis make car ownership unnecessary. And AI diagnostics plus wearable sensors deliver physician-grade healthcare to anyone with a phone.

The last mile is energy and housing. Solar plus battery storage crosses the threshold where electricity is effectively free in most of the world by 2030. 3D-printed construction and robotic labor bring the cost of building a quality home below $80K by 2031. At that point, the four basics of life (food, shelter, energy, healthcare) plus the four luxuries (education, transport, assistance, entertainment) are all demonetized.

China’s billion-agent simulation we discussed on the pod is a preview. When you can simulate an entire society, you can simulate the economic models that make universal high income work. The question isn’t whether the abundance arrives. It’s whether we distribute it.

WHAT THIS MEANS FOR YOU

If you’re an entrepreneur: Every item on that table is a company. The robot chef, the AI tutor, the $80K home, the robotaxi network. Pick one and build it.

If you’re an investor: The companies demonetizing these categories are the next trillion-dollar opportunities. The firm that makes housing 10x cheaper owns a bigger market than any software company in history.

If you’re an executive: Your employees will soon have AI assistants, robot helpers, and free education. The workplace you’re managing today won’t exist in 5 years. Start planning for it.

If you’re a parent: Your kids will have access to better education, healthcare, and transportation than any billionaire’s children have today. Your job is to teach them adaptability, curiosity, and purpose. The skills layer is free. The wisdom layer is not.

If you’re a policymaker: The question isn’t how to tax the wealthy. It’s how to accelerate the demonetization so the majority gets there faster. Every regulation that slows AI, robotics, or autonomous transit delays the arrival of abundance for the people who need it most.

Louis XIV died of gangrene because the best medicine of his era was leeches. Your children will live in a world where the best medicine is an AI that reads your genome, monitors your biomarkers, and prevents disease before it starts. They will have a robot that cooks, cleans, and runs errands. They will have a tutor that knows their learning style better than they do. They will have a car that drives itself. They will have access to every piece of music, every film, every book, every course, every conversation ever recorded.

They will be kings. Everyone will be kings. The only question is when.

To a future of abundance,

Peter

TLDR: I am a father of two 15-year-olds. And I am angry. Not at my kids. Not at their teachers. I am angry at a system that was designed 150 years ago for the industrial revolution and has barely changed since. A system that takes the most curious, creative, capable generation in human history and puts them in rows, hands them textbooks, and tells them to memorize things an AI can look up in micro-seconds.

I recently surveyed my Moonshots community — parents, teachers, high school students, college students, and professionals — to ask a simple question: is education preparing people for the future? I concede in advance that my Moonshot listeners represent a unique and perhaps biased group. Nonetheless, what the data indicates is what I personally believe, and what many parents I know also feels in their gut.

THE SYSTEM IS BROKEN. THE NUMBERS PROVE IT.

I asked every cohort to rate how well education prepares students for the future on a scale of 1 to 10.

· Average: 4.3 out of 10.

· Teachers: 3.5 out of 10

· Parents of teens gave it a 3.8.

The people inside the system know the system is failing.

Chart 1: Average rating of how well education prepares people for the future, by cohort. Source: Moonshots Podcast Survey

TOP CONCERN ABOUT YOUR CHILD’S FUTURE?

And when I asked parents of teens about their top concerns for their child’s future, the number one answer was overwhelming:

Chart 2: Top concerns parents of teens have about their child’s future. 87% selected ‘not being prepared for an AI future.’ Source: Moonshots Podcast Survey

87% of parents said their top concern is that their kids are not being prepared for an AI future. Not bullying. Not grades. Not college admissions. AI readiness. The thing that will define their child’s economic future is the thing the school system is least equipped to teach.

IS AI BANNED IN YOUR CLASSROOM?

I asked parents of teens a direct question: is your child learning AI in school? Here is what they told me.

Chart 3: Parent responses to ‘Is your child learning AI in school?’ Source: Moonshots Podcast Survey

42% said AI is allowed but not taught. 41% said AI is banned entirely. Only 2% said AI is a core part of the curriculum. That means 83% of children are in schools where AI is either banned or simply tolerated. We are taking the most powerful technology in human history and keeping it away from the generation that needs it most. This is like banning electricity in schools in 1890.

WHAT SKILLS MATTERS MOST IN NEXT 10 YEARS?.

Here is the good news. Parents are not confused about what their kids need. When I asked parents to pick the top 4 skills that will matter most in the next 10 years, the answers were clear:

Chart 4: Skills matter most in the next 10 years. Source: Moonshots Podcast Survey

The answers are clear, and they make sense! What matters most?:

· 78% chose AI literacy.

· 72% chose critical thinking.

· 68% chose adaptability.

· 63% chose entrepreneurship.

These are not the skills being taught in school. Schools teach algebra, history, biology, and standardized test preparation. Parents want AI literacy, mindset, and entrepreneurial thinking. The gap between what schools teach and what parents know matters is the entire problem in one chart.

And when I asked parents what would make a teen program genuinely valuable, the answer was a blueprint for the future:

Chart 5: What parents say would make a program valuable. Source: Moonshots Podcast Survey

71% want AI skills. 58% want purpose and mindset development. 55% want entrepreneurship. Real-world projects. Independent thinking. The parents are telling us exactly what to build. The question is whether anyone is listening.

IS A COLLEGE DEGREE LOSING VALUE?

I asked students, teachers, and professionals whether a college degree will become less important in an AI world. 79% said yes.

Chart 6: Will a college degree become less important? Source: Moonshots Podcast Survey

45% said significantly less important. A four-year degree that costs $300K and prepares you for a job that AI will do in 2028 is not an investment. It is a liability. We need a new pathway, and we need it now.

A NEW SOCIAL CONTRACT FOR OUR KIDS

Here is what I propose. A new social contract for education built on three pillars.

First, Mindsets: The most important skill your child can develop is how to think, their Mindset – how they react to opportunities and challenges. Mindset is the operating system. Everything else is an app.

What mindsets do I teach my kids?

(1) Purpose Driven: What is something you LOVE, that actually helps humanity.

(2) Curiosity Mindset: How do you continue to learn throughout life.

(3) Abundance Mindset: Understanding that there is nothing truly scarce. That technology repeated converts scarcity into abundance.

(4) A Moonshot Mindset: The idea that AI (and soon AGI and ASI) will enable your kids to achieve their wildest dreams, 10x or 100x bigger than they ever imagined.

(5) An Exponential Mindset: Understanding that we are in the midst of the Singularity, and the rate of change is accelerating.

Second, Entrepreneurship: The traditional career ladder is becoming obsolete. 79% of parents and professionals in my survey said so. The future belongs to builders, creators, and problem-solvers. Every child should graduate knowing how to start something, sell something, and create value. Entrepreneurship is not a business class. It is a life skill.

Third, AI skills: Not coding. AI literacy. The ability to prompt, to reason with AI, to use it as a research partner, a creative collaborator, and a force multiplier. Fei-Fei Li told Andrew Huberman that Socrates was humanity’s best prompter. She is right. The Socratic method is the pedagogy for the AI age.

And the optimism is there. 84% of parents believe AI will increase human opportunity, not shrink it. 73% said AI will greatly increase opportunity. The parents are not afraid of AI. They are afraid their kids will miss it.

I have two 15-year-olds. I know the fear. I know the frustration. I know what it feels like to look at a school system and realize it was built for a world that no longer exists. But I also know this: the tools to fix this are here. AI tutors that deliver personalized education at zero cost. Online communities that connect curious kids globally. Entrepreneurship programs that teach real-world skills. The infrastructure for a new education system exists. What is missing is the will to build it.

WHAT THIS MEANS FOR YOU

If you’re a parent: Demand AI education in your school. If the school will not teach it, teach it at home. Your child’s ability to use AI will determine their economic future. Start today.

If you’re a teacher: You are not the problem. The system is. But you have the power to bring AI into your classroom anyway. Socratic prompting, AI-assisted research, project-based learning. The tools are free. Use them.

If you’re a policymaker: The ban-on-AI approach is the ban-on-calculator approach. It failed then and it will fail now. Write policy that mandates AI literacy, not AI prohibition.

If you’re an entrepreneur: The parents just told you what to build. AI skills, mindset development, entrepreneurship, real-world projects. 71% of parents want it. Build it.

If you’re a student: Do not wait for the system to catch up. Learn AI. Learn to prompt. Learn to build. Learn to think in abundance. Your education is your responsibility now. Take it.

To a future of abundance,

Peter

TLDR: Within three to five years, you will give AI permission to know everything about your life. Your schedule, your blood chemistry, your conversations, your stress levels, your taste in music and comedy. And in return, your life becomes automagical. AI anticipates your needs before you articulate them. Here are five scenarios showing what that world looks like when it arrives.

I’ve been describing this concept on podcasts for a while now. I call it ambient AI. The idea is simple: AI is always on, always around you, making your life automagical before you even walk into a room. Your AI knows if you’re depressed or sad and it changes the music for you. Maybe on the TV screen, your favorite comedian comes up. Your AI knows you’re walking in a particular direction and it says, hey, there’s a set of stairs over there. Take the stairs instead of the elevator.

I think we’ll have this by 2028. You’re going to give your AI access to listen to every phone call, measure everything on your body, see you through all the cameras in your home. So that it always knows how to optimize your life, how to make it automatic. It knows physiologically that you need more protein or more water or more vitamin D. It knows your objectives. And it acts on all of it without you asking.

I know the privacy conversation. I’ve had it. Dan Sullivan asked me about it and I told him: privacy is an old concept. We dribble data constantly already. The trade-off is worth it for what ambient AI gives back.

“AI is on everywhere, all the time, in our environment. It is our interface to everything.”

So what does that actually feel like? Here are five scenarios showing what life becomes when AI goes ambient.

SCENARIO 1: THE ROBOT CHEF WHO KNOWS YOUR BLOOD

It’s 6:47 AM. You haven’t spoken to anyone. You haven’t opened an app. You walk into the kitchen and the table is set. A plate of steel-cut oats with blueberries, a side of scrambled egg whites, and a glass of water with electrolytes is waiting for you.

Here’s what happened while you slept. Your wearable tracked your blood glucose, heart rate variability, and cortisol levels overnight. Your AI cross-referenced that with your continuous glucose monitor, your recent workouts, and your last blood panel. It determined your protein was low, your hydration was behind, and your cortisol was elevated from yesterday’s board meeting.

It sent the order to the robotic chef in your kitchen. The robot, which has access to your pantry inventory and your macronutrient targets, selected a meal that addresses every deficiency. The blueberries are there because your antioxidant levels dipped. The egg whites are there because your muscle recovery from yesterday’s workout demands protein. The electrolytes are there because you sweated more than usual overnight.

You didn’t ask for any of this. You didn’t plan a meal. You didn’t count macros. Your AI knew what your body needed before you did and had a meal prepared before you walked in. Multiply that by three meals a day, every day, for the rest of your life. That’s not optimization. That’s a different relationship with food entirely.

SCENARIO 2: THE CAR THAT KNOWS YOU’RE LEAVING

You’re in a meeting that’s running 15 minutes long. You didn’t tell anyone. You didn’t text your driver. You didn’t open an app. But when you walk out of the building, your autonomous vehicle is idling at the curb, climate-controlled to 72 degrees, with your preferred playlist already queued.

Here’s how. Your AI was listening to the meeting through your earpiece. It heard the conversation wrapping up. It checked your calendar and saw your next appointment is at a restaurant 22 minutes away. It checked real-time traffic, calculated departure time, and dispatched the car three minutes before the meeting ended. The car arrived as you walked out. No waiting. No calling. No thinking.

And the playlist isn’t random. Your AI detected from your voice patterns during the meeting that your stress was elevated. It skipped the high-energy podcast you usually listen to on drives and queued a meditation track instead. By the time you arrive at the restaurant, your heart rate has dropped 12 beats per minute and you’re in a better state for the conversation ahead.

SCENARIO 3: THE MOOD-READING LIVING ROOM

You come home from a long day. You walk through the door and the house has already changed. The lighting shifted from cool white to warm amber. The temperature dropped two degrees because your skin temperature was elevated. Your favorite episode of the comedy you’ve been watching is paused on the TV, ready to resume from the exact second you left off.

Your AI detected from your gait analysis via the front door camera that you were walking slower than usual. Your heart rate was 8 BPM above your baseline. Your voice patterns in the last phone call showed fatigue markers. The system concluded: stressed, not energized. So it changed the environment to match what you need.

Your 15-year-old son walks in 20 minutes later. His AI profile is completely different. The lighting shifts again, subtly, to something brighter that his circadian rhythm prefers at this hour. His homework reminders appear on his phone. The kitchen starts prepping his post-soccer snack because his glucose monitor shows he’s low. Same house. Same AI. Different human. Different response. Zero input required from either of you.

SCENARIO 4: THE SURPRISE-AND-DELIGHT SHOPPER

You’re on a phone call with your wife, talking about the upcoming anniversary. You mention, in passing, that you saw a beautiful piece of jewelry in a shop window in Malibu last weekend but you didn’t go in. The conversation moves on. You forget about it entirely.

Three days later, a package arrives at your door. Inside is the exact piece you described. Your AI was listening to the conversation, identified the specific item from your description, cross-referenced it with the location data from your phone showing you’d walked past a jewelry store in Malibu, located the store, confirmed the item was in stock, checked your budget parameters, and ordered it. It even had it gift-wrapped.

This is the scenario that makes people uncomfortable and excited at the same time. Your AI is spending your money on your behalf based on conversations it overheard. But here’s the thing: you gave it permission. You set a monthly surprise budget. You defined the categories. You approved the parameters. Within those guardrails, it has autonomy to create moments of delight that you would never have created for yourself because you were too busy, too distracted, or too caught up in the next meeting to remember.

The package arrives and your wife opens it. She looks at you. You shrug. “I mentioned it once.” That’s the magic. You didn’t shop. You didn’t remember. You didn’t plan. Your AI turned a passing thought into a gift. And that’s a kind of attention that no human, however loving, can sustain across every conversation, every day, for a lifetime.

SCENARIO 5: THE MEETING WHISPERER

You’re walking into the most important meeting of your week. It’s with a potential investor who could fund your next venture. You’ve never met them. Your AI has been preparing since 4 AM.

As you approach the conference room, your AI feeds you a briefing through your earpiece. The investor’s name, their portfolio companies, the last three investments they made, the sectors they’re bullish on, the sector they just passed on. Their LinkedIn shows they posted about their daughter’s college graduation yesterday. Your AI flags that as an opening for personal connection. It reminds you that you met briefly at a conference in 2024 and they asked about space mining. It suggests you open with that.

During the meeting, your AI is listening. It detects from the investor’s tone that they’re skeptical about your timeline. It whispers in your earpiece: “Address the timeline concern directly. They’re comparing you to a company that missed its milestones last year.” You adjust. You give a concrete, data-backed timeline. The investor nods.

After the meeting, your AI has already drafted the follow-up email, summarized the key points, flagged the action items, and updated your CRM. It also noticed the investor mentioned a book they’re reading. It ordered a copy for you so you can reference it in your next conversation. You walk out of the meeting feeling like you had a team of researchers, a strategist, and a coach in your ear the entire time. Because you did.

THE GIVE AND THE GET

Every one of these scenarios requires you to give something. Your data. Your conversations. Your biology. Your location. Your preferences. The privacy conversation is real and it matters. But the trade-off is this: in exchange for access, you get back something no human has ever had. A consciousness that never sleeps, never forgets, never gets too busy to pay attention to your life. An intelligence that is always optimizing for your health, your happiness, and your success.

I told Ed Mylett on his podcast: within three to five years, you’re going to give access to your AI to listen to your conversations, look at the cameras in your home, understand the sensors on your body. Because when you do that, it’s able to support you in so many different ways. It’s anticipating your needs and providing for you in a magical fashion.

WHAT THIS MEANS FOR YOU

If you’re an entrepreneur: The ambient AI stack is the next platform. Someone has to build the privacy layer, the permission engine, the home integration, the health-data pipeline. Every piece of what I described is a company.

If you’re an executive: Start thinking about what your employees will expect when their personal AI is better than any assistant you could hire. The workplace will have to adapt to people who arrive already briefed, already optimized, already coached.

If you’re an investor: The companies building ambient AI infrastructure, home robotics, continuous health monitoring, and autonomous logistics are building the rails for a trillion-dollar lifestyle shift. This is the next iPhone moment.

If you’re a student: The interface between human biology and AI is the most important engineering challenge of your generation. Study it. The person who builds the trust layer for ambient AI writes the future.

If you’re a parent: Your kids will grow up with an AI that knows them better than you do. That’s not a threat. That’s a tool. The question is whether you help them build a healthy relationship with it or resist the inevitable.

To a future of abundance,

Peter

TLDR: I just sat down with Michael Kratsios, the White House Science Advisor and head of OSTP, to talk about how we 10x, potentially 100x, breakthroughs in science. We covered autonomous labs running experiments 24/7, golden tickets for radical ideas, five-year grants that let scientists actually work instead of writing proposals, and a vision for American science that goes from the Model T to the Starship in a decade. The two reports, the Genesis Mission and the Golden Age of Science, are the most ambitious science policy documents I’ve read from the government in my lifetime.

Ray Kurzweil predicts we’ll see as much progress in the next decade as we saw in the last century. That’s going from the Ford Model T to the Starship in ten years. On top of that, we’re on the edge of AGI, maybe in three years, and ASI not long after. So when I sat down with Michael Kratsios, the Science Advisor to the White House and the head of OSTP, I wanted to know: how does the government keep up with that?

“We should be aiming for 10x, not 2x. With what’s happened in AI even over the last six months, we’ve got to aim big.”

The Golden Age of Science report he handed me read like a kid in a candy store. I’ve spent 30 years in the incentive-prize world, and this document went further than I expected. It names specific mechanisms, it questions sacred cows, and it proposes ideas I never thought I’d see in a government document.

AUTONOMOUS LABS: SCIENCE AT MACHINE SPEED

Here’s the line from the report that stopped me: AI proposes the hypothesis, robots run the experiment, the model reads the results and designs the next one. 24/7, no humans in the loop. Closed-loop science at machine speed.

This is happening right now. Lila Sciences, a company I’ve been close to, just closed their B round and is building out a million square feet of robotic, automated lab space. Their scientific superintelligence performs ahead of anything else in the field. Geoffrey Von Maltzahn (CEO, Lila) said he wants to be for Genesis what Elon is for NASA. That’s the scale of ambition we’re talking about.

“Imagine going online, putting a hypothesis in, and hitting go. Then starting on the next one without needing to wait.”

Kratsios’ vision goes further. He wants to retrofit this capability into the national labs, so a high school kid in Ohio can submit a hypothesis online, hit go, and have an autonomous lab at a federal facility run the experiment. No gatekeepers. No waiting for a professor to return your email. Just science. That democratization of access to cutting-edge infrastructure is the piece that changes everything.

BACKING YOUNG SCIENTISTS WITH CRAZY IDEAS

The Golden Age report identifies a problem I’ve screamed about for years: young talent waits too long before they’re given a chance. The median age of an intramural NIH scientist is 71. Seventy-one. The average age of a Nobel laureate’s prize-winning work is in their 20s. We’ve built a system where the people most likely to make breakthroughs spend their peak creative years writing grant proposals for people three times their age to evaluate.

Michael’s answer is a combination of mechanisms that sound like they came from an XPRIZE design session, not a government report.

• Five-year grants funded on day one. Stop making scientists write their next grant before they finish their current one.

• Fast grants, a few pages, reviewed in under a month. Tyler Cowen proved this works during COVID. Decisions in hours, not months.

• Golden tickets. Each reviewer on a merit panel gets one to three tickets they can use to unilaterally fund a radical idea, even if the rest of the committee thinks it’s crazy. This incentivizes bolder proposals and brings better people onto review panels.

• Incentive prizes and advanced market commitments. Pay for results, not proposals. The report cites the $10 million Ansari XPRIZE, which I personally love for obvious reasons. Michael said the government could absolutely put up $100 million prizes for grand challenges.

The report also describes a vision that made my jaw drop: prediction markets, crowdsourced AI agents, and DAOs funding scientific research directly. Imagine a marketplace where funders post bounties for breakthroughs, AI agents identify promising leads, hire autonomous labs, and verify cryptographically signed results. Smart contracts release payments as milestones are met. The system operates continuously at machine speed. That’s a government document describing a complete AI-native reimagining of the scientific process. I can’t believe he actually wrote it down.

SPEED, AGILITY, AND THE 10X MINDSET

The Golden Age report names six national technology missions: AI, quantum, commercial fusion, lunar exploration, robotics, and next-generation semiconductors. Each one matters. We’re going back to the Moon in 2028, building a lunar base by 2030, putting a nuclear reactor in space by 2028. A scientifically relevant quantum computer by the end of the term. These are Apollo-scale commitments, and they’re the kind of north stars that make young people want to go into science in the first place.

What impressed me most was the intellectual honesty of my conversation with Kratsios. The report admits that scientific productivity has been declining despite larger budgets. The NIH budget has ballooned to $45 billion, yet drug costs are higher than ever. The conclusion: we’re doing the same thing over and over, pouring more money into a broken process, and expecting different results. Forty-five percent of a researcher’s time goes to administrative work, not science. These are the crown jewels of our country, and we’re making them fill out paperwork.

“The day before something is truly a breakthrough, it’s a crazy idea. So where are we experimenting with crazy ideas?”

WHAT’S MISSING

I told Michael two things were conspicuously absent from the report. First, longevity. If you want to impact the US economy more than anything else, tackle aging. The average American lives to 79 but their health expectancy is 63. That’s 16 years of poor health. Give people an extra decade of productive, healthy life and you transform everything: GDP, healthcare costs, Social Security, human flourishing. Michael agreed it should have been there and said we should work together on it.

Second, biotech as a named mission. It’s nested under Genesis, and $5 billion in Genesis grants were announced just last week, many in biotech. But biotech deserves its own pole. Dario Amodei says we could double the human lifespan in 5 to 10 years. Demis Hassabis says we’ll cure all diseases within a decade. When the people running the frontier say that, you listen.

WHAT THIS MEANS FOR YOU

If you’re an entrepreneur: The government is actively looking to fund radical science through prizes, fast grants, and golden tickets. If you have a crazy idea, the barrier to entry just dropped. Start filing.

If you’re an executive: The autonomous lab revolution means R&D cycles compress from years to weeks. Companies that integrate AI-driven experimentation first will out-discover competitors who still run science the 1960 way.

If you’re an investor: The Golden Age report effectively describes a new asset class: science infrastructure. Autonomous labs, AI agents running experiments, and the companies building the picks and shovels for machine-speed discovery.

If you’re a student: The system is being rebuilt around you. Five-year grants, golden tickets for radical ideas, and access to national lab equipment from your laptop. The path from idea to experiment is about to collapse to zero.

If you’re a parent: Your kids will have access to scientific infrastructure that didn’t exist when you were in school. A high schooler will be able to run experiments on national lab equipment. Encourage the crazy questions.

To a future of abundance,

Peter

TLDR: Power is the innermost loop, on Earth and in space. NASA’s new Administrator, Jared Isaacman, is restarting the agency’s nuclear program, compact reactors on the Moon and atomic engines in deep space, and it could cut the trip to Mars from nine months to three. Chemistry got us off the planet. Fission is how we take the solar system.

Every ounce of propellant that has ever moved a human being through space was lifted off Earth’s surface. Every single ounce. We spent 60 years clawing out of one gravity well only to drag along every drop of energy we would ever need, sloshing in a tank. The most expensive way to travel ever devised.

Chemistry is the problem. A hydrogen-oxygen rocket, the best chemical engine we have, converts a laughably tiny fraction of the energy locked in its fuel into motion. Jared framed the energy ladder better than anyone I have heard.

“Chemical propulsion pulls an almost immeasurably small fraction of the energy in matter. Fission gets you to a tenth of one percent. Fusion, half a percent. The destination is antimatter annihilation.”

— Jared Isaacman, NASA Administrator (paraphrased from the pod)

That is the unlock Jared is chasing, and he is chasing it on two fronts at once: keeping the lights on at a Moon base, and lighting a fire under a spaceship.

POWER ON THE MOON

Start with the obvious objection. The Moon has no atmosphere, no clouds, and free sunlight. Why haul a reactor a quarter million miles when you could just unfold solar panels? Because the Moon runs on a brutal clock: roughly 14 Earth-days of sunlight followed by 14 Earth-days of total darkness. To survive 336 straight hours of night on solar, you need a massive battery farm. Every square meter of solar array and every kilogram of storage has to be lifted out of Earth’s gravity. Solar on the Moon is a mass problem wearing a free-energy costume.

Fission flips the arithmetic. Compact, and always on. It does not care where the Sun is.

This isn’t a concept. It’s a system under construction. NASA’s Fission Surface Power project, with the Department of Energy and Idaho National Laboratory, is building a 40-kilowatt reactor, enough to run a base of four to six astronauts, their habitats, rovers, and ice-mining gear. HALEU fuel (uranium enriched to 19.75 percent), under 6 metric tons, inside a 4-meter cylinder, running 10 years autonomously with no human touch. NASA wants a lunar-ready variant by 2030, orbital testing as early as 2028.

“Without nuclear, you’re going to need a lot of Optimus robots on Mars walking around dusting off all the solar panels.”

— Jared Isaacman, NASA Administrator

Picture it. The most advanced humanoid robots ever built, on another world, employed full-time as janitors for a solar farm.

ATOMIC ENGINES, THE REAL UNLOCK

Surface power is the appetizer. Propulsion is the main course, and it is the most exciting thing Jared during our entire Moonshots podcast. Here is the number that matters. Specific impulse (expressed in “second”) is measure of merit for rockets. Our best chemical engines top out around 450 seconds. A nuclear thermal engine, which runs hydrogen through a reactor core instead of burning it, hits 800 to 1,000 seconds. Two to three times the efficiency, which in rocket math is the difference between crawling and sprinting.

What does that buy you? A chemical trip to Mars is a 6-to-9-month one-way grind. Nuclear can cut that to roughly 3 to 4 months. Cut the transit in half and you halve your crew’s radiation dose, halve the muscle and bone loss, and halve the supplies you have to pack. Speed is not a luxury in deep space. It is a life-support system.

NASA and DARPA had a nuclear thermal demonstrator called DRACO aimed at a 2027 flight. It got trimmed in 2025, but the knowledge folded back into NASA, and Jared’s version is arguably smarter. His first move is not the fastest engine. It is the first one that actually flies.

That vehicle is called “SR-1 Freedom” – the first nuclear power-and-propulsion spacecraft, and almost all of it is repurposed hardware. The body is the Power and Propulsion Element built for the Gateway lunar station, electric thrusters already integrated. The reactor baseline comes from components matured at Idaho National Laboratory over decades, largely paid for by other agencies. When Politico said the administrator wants to spend $2.5 billion on a nuclear spaceship, Jared’s answer was that taxpayers already spent it. He just refuses to let it rot in a warehouse. Then he gave the analogy that made the whole thing click.

“This is what Nautilus was. A diesel boat repurposed as a nuclear sub. Rickover said it was a 70% solution, but it gave birth to the nuclear navy. We’re doing the exact same thing.”

— Jared Isaacman, NASA Administrator

The history checks out. USS Nautilus put a revolutionary reactor inside a conventional diesel-era hull. Admiral Hyman Rickover refused to invent the reactor and the hull at once. Solve one impossible thing per vehicle. Nautilus was not the best submarine ever built. It made every submarine after it possible. SR-1 will not knock your socks off. SR-2 will be better. And as Jared put it, we are not jumping straight to Battlestar Galactica.

What gets optimized between versions is material science. Run the reactor hotter and you shed mass, because you stop needing radiators the size of football fields to dump waste heat. Improve power conversion and nuclear starts beating solar even inside Jupiter’s orbit. Past Jupiter, where sunlight is a rumor, there is no argument left. Nuclear is the only thing that works. That is what finally makes probes to Enceladus, Uranus, and Titan routine instead of once-a-generation.

Nuclear propulsion is the way humanity navigates our Earth-Moon-Mars-Solar System future. While its not a warp-drive yet, it beats burning propellants out the rear of a rocket.

Issacman believes NASA gets crew to Mars first, not SpaceX, for one reason: NASA does not do one-way missions. You want the crew home to talk about it. The moment you require a round trip, your dependency on chemical propulsion has to collapse, because making return propellant on another planet is savagely hard. So the architecture is chemically augmented nuclear electric propulsion. First crew of four, a roughly three-year round trip with about 30 days on the Martian surface, inside a 10-to-15-year window.

“The ultimate goal is to bring astronauts to Mars and back with the fewest miracles required. That may not be the fastest way, but if it doesn’t require cryogenic refueling, that’s a win.”

— Jared Isaacman, NASA Administrator

The fewest miracles required. I have been collecting phrases from engineers for 30 years and that one goes straight into the permanent file. If we scratch and claw our way to Mars on chemical propellant, we end up with a re-run of Apollo: a flag, a footprint, a ride home, and nothing you can build on. Do it with nuclear and you have a road. A rotation. A supply line. The difference between a stunt and a settlement.

WHAT THIS MEANS FOR YOU

If you’re an entrepreneur: The supply chain for space nuclear barely exists. High-temperature materials, compact power conversion, and autonomous reactor control are wide open, and the materials work sells straight into terrestrial small modular reactors.

If you’re an executive: Jared found $2.5 billion of finished hardware in a warehouse and shipped it as a product. Audit what your organization has already paid for and abandoned.

If you’re an investor: Lunar and deep-space nuclear are now national space policy, which means a decade of procurement. Track the Fission Surface Power teams and the HALEU fuel supply chain.

If you’re a student: Nuclear engineering plus materials science is about to be the most leveraged skill pair on the planet, on Earth and off it. Pick that major.

If you’re a parent: A child born today will be in their teens when four humans leave for Mars on an atomic ship. Tell them the Moon gets 14 days of night. Better physics hook than the textbook.

For 60 years we have been prisoners of chemistry, hauling every drop of fuel up from Earth and creeping to other worlds on the least energetic reaction we know how to control. Jared is betting we break out this decade: a reactor humming through the lunar night by 2030, an atomic tug assembling in orbit, and a trip to Mars that shrinks from nine months to three.

We have barely scratched our own solar system. With Nuclear doing the pushing we have a chance to make accelerate everything.

To a future of abundance,

Peter

TLDR: NASA has a new Administrator (Jared Isaacman) who built a $100-billion-a-year payments company, flew himself to orbit twice, and now wants his probes to think and his Moon base enabled by robots before a single astronaut arrives. Isaacman is a rockstar who is putting NASA on an exponential roadmap. We are about to speed-run Star Trek.

GIVE THE PROBES AN AI BRAIN

We have been exploring the solar system by postal mail. So I asked Jared the the following question on my Moonshot’s podcast, “Can you imagine near-term missions where every probe and robot runs an AGI/ASI model, the world’s smartest scientist onboard remotely directing the entire mission – deciding what to explore and what experiments to run, without waiting for directions from mission control on Earth?”

“A hundred percent. And there’s already a mission designed for that, which is DAVINCI, which is going to Venus.”

— Jared Isaacman, NASA Administrator

DAVINCI is the perfect proving ground, because it is a probe on a death clock. Launching around December 2030, it drops a descent sphere through the Venusian atmosphere over the Alpha Regio highlands with roughly one hour of life. Surface pressure there is about 90 times Earth’s, temperatures near 465 C, hot enough to melt lead. Five instruments, including a mass spectrometer and tunable laser spectrometer, sample noble gases on the way down to answer whether Venus ever had an ocean.

No time to phone home. So NASA designs the judgment into the vehicle. As Jared put it, “the on-mission AI decides what data matters most in the least time it has and sends the data home before it dies. A machine rewriting its own science plan while plunging to its death. That is not Voyager.”

Then there is Dragonfly, pure science fiction made real: a nuclear-powered octocopter, a drone the size of a small car with eight rotors, flying across the orange dunes of Titan. Launch July 2028, arrival 2034. It lands, flies again up to 8 kilometers per hop and 4 kilometers high, running on a plutonium battery through Titan’s 8-Earth-day nights. A flying chemistry lab hunting the ingredients of life. You cannot joystick that when your signal takes 80-plus minutes each way. It has to think for itself.

And these are just the landers. Europa Clipper, the largest planetary spacecraft NASA has ever built, arrives at Jupiter in April 2030 for 49 flybys of an ocean moon. The Nancy Grace Roman Space Telescope launches August 30 on a Falcon Heavy with a field of view 100 times wider than Hubble’s. Every one is about to disgorge more data than humans can read.

Jared’s second AI play is just as impactful: Let AI analyze all the data. There is six decades of archived data nobody has fully read, and a lot more coming. The only intelligence able to unpack the hidden massive discoveries will be AI.

RETIRE THE MEAT SACKS

The uncomfortable truth about human spaceflight: we are squishy. We need oxygen, water, food, radiation shielding, and a ride home. We have always been the rate-limiting step. And the coming Moon base NASA is building on the south pole of the Moon has an environment that is genuinely trying to kill us.

“You’re talking about negative four hundred degrees in some of the permanently shaded regions. We actually have survival problems in the south pole of the Moon that exceed that of Mars.”

— Jared Isaacman, NASA Administrator

Read that twice. The lunar south pole is harder to survive than Mars, and we go there on purpose, because that is where the water ice is. What’s the solution? Robots, lots of robots including humanoid robots.

“You want the fewest reasons possible to ever put an astronaut in a suit outside the base unless there was no other alternative. And humanoid robots are going to give you that.”

— Jared Isaacman, NASA Administrator

Any place we plant a human outpost, he said, it would be crazy not to put humanoid robots there first as a force multiplier, from infrastructure buildout to logistics. This is not a thought experiment. NASA has built humanoids for years, from its Valkyrie robot at Johnson Space Center to its partnership with Apptronik, whose founders cut their teeth on Valkyrie and now build the Apollo humanoid with Google DeepMind. Elon calls Optimus the ultimate von Neumann probe for the solar system.

So I pushed for the number that matters. Over-under, when does the first humanoid robot walk on the Moon?

“I’d kind of be shocked if somebody didn’t smuggle one on board.”

— Jared Isaacman, on the uncrewed lander demos before the 2028 crewed landing

His real window: four to six years for robots to start building infrastructure. But since both SpaceX and Blue Origin owe NASA uncrewed lander demos before the 2028 crewed landing, he expects someone to sneak a humanoid onto one early. Optimus or Figure or Apollo boots may hit the lunar surface before human boots do, this decade.

What I loved was his refusal to sell the pretty version. Phase one is not a glass dome with crop farms. He plans to drop landers and rovers near-monthly before astronauts arrive, learn what breaks at 400 below, and leave a mess. His word for it was a junkyard. Mercury before Gemini, Gemini before Apollo. The robots move in during phase two, once Starship is flying often and orbital propellant transfer works, because that is when you can land enough mass to field armies of them.

SEVEN DRAGONFLIES A YEAR

Now feel the acceleration. NASA’s Science Mission Directorate runs about $7 billion a year, and historically that buys one flagship per decade, each ballooning from $1 billion to $3 billion because it cannot fail. So I asked whether AI and collapsing manufacturing costs let him 10x or 100x the number of science missions. He fired it straight back.

“Between AI and additive manufacturing, $7 billion a year could be cranking out seven Dragonflies a year”

— Jared Isaacman, NASA Administrator

Seven Titan-class missions a year instead of one per decade. A 70x jump in the rate of scientific discovery to help us explore for life. The obstacle, he was honest, is human, not technical. People build moats around their jobs and never want to engineer themselves out of one. In his words: he does not want his grandchildren to be excited about these missions. He wants to be excited about them, now.

WHAT THIS MEANS FOR YOU

If you’re an entrepreneur: Autonomous science is a green field: onboard inference under brutal power and radiation limits, robotic manipulation in vacuum, and tools that mine 60 years of NASA archives. The archive play needs no launch license.

If you’re an executive: Jared inherited a $25 billion agency asked to please everyone and answered with brutal focus. Audit what your organization has already paid for and never shipped.

If you’re an investor: Follow the near-monthly CLPS lander cadence and the humanoid suppliers, Apptronik and the Optimus supply chain. A recurring lander program is revenue, not a moonshot.

If you’re a student: A teenager in Texas found new galaxies in NASA’s own data and got an internship. The archives are public. Nobody is stopping you.

If you’re a parent: Your kids will likely watch a humanoid robot take the first step on the Moon before an astronaut does. Tell them the robot got there because a human decided it should.

For 60 years we explored space at the speed of a mailed letter. In one 90-minute conversation, Jared laid out a NASA where the rover decides, the base builds itself, and the mission rate goes up 70-fold. The change is not coming. It is here, accelerating faster than the org chart can keep up.

So here is what I cannot stop turning over. If a probe can choose what to investigate at Titan without asking us, and a robot can pour the foundation before we arrive, what exactly is the human job out there?

I have a guess. But I would rather hear yours.

To a future of abundance,

Peter

I’ll be on CNN this afternoon (8/4) and CNBC tomorrow (8/5), commenting on the expected SpaceX earnings report, their first as a public company following the IPO. As you might expect, I’m very bullish about SpaceX, and I thought I’d lay out the reasons in some detail.

Here’s my core thesis. The Street doesn’t fully understand how to value SPCX, because it keeps trying to price one company. It’s actually five, stacked on top of each other and wired together by Elon’s particular genius for compressing time. Look at each one on its own and the picture changes completely.

1. THE FIRST TRULY GLOBAL COMMUNICATIONS COMPANY

Strip away the rockets and SpaceX is already one of the fastest-scaling telecom utilities on Earth. Starlink connects 8 billion people to gigabit internet from orbit, a backup and a replacement for terrestrial comms in the same box. And direct-to-device (satellite straight to your cell phone) turns every smartphone on the planet into an always-on, always-available terminal, no dish required.

“10 million subscribers, doubling every year, and it just turned free-cash-flow positive. That’s not a satellite company. It’s the fastest-scaling utility in human history, connecting the 3 billion people the internet forgot.”

2. A HYPERSCALER SELLING THE WORLD’S SCARCEST RESOURCE

SpaceX’s second business is compute, and the world is starved for it. This is the part the market is sleeping on. Google, Anthropic, and Reflection AI are already buying, and the contracts are signed:

• Anthropic: $1.25B/month, ~$15.0B/year, through roughly May 2029

• Google: $920M/month, ~$11.0B/year, from October 2026 through June 2029

• Reflection AI: $150M/month, ~$1.8B/year, from July 2026 through 2029

That’s roughly $28 billion of annualized, contracted compute revenue in 2026. Not a projection. Signed backlog.

Here’s what most people miss: nobody moves faster than Elon, and it isn’t close. xAI stood up a 100,000-GPU H100 cluster in 122 days, roughly 5x faster than a conventional hyperscale deployment. Then it doubled Colossus to 200,000 GPUs in another 92 days, reaching a full gigawatt in about 13 months. Meta’s new 1-gigawatt Indiana campus will take 22 to 24 months. Most remarkable of all: the gap from installing the first servers to starting AI training was about 19 days. Comparable clusters often take many months.

Then comes the orbital layer. Around 2028, Elon plans to begin launching Starmind, dedicated solar-powered AI-compute satellites, scaling toward up to a million satellites and roughly a terawatt of compute in orbit by 2030. The most-cited Wall Street number is ~$322 billion in 2030 AI revenue for SpaceX, and worth knowing on air, that figure blends terrestrial and orbital compute. The Starmind-specific piece hasn’t been broken out. So the $28 billion you can see today is the floor, not the ceiling.

3. TERAFAB: A CHIP MANUFACTURER AT A SCALE NOBODY ELSE IS ATTEMPTING

Third, SpaceX is becoming a chip company. Elon’s plan for the Terafab (his gigafab in Texas) targets over 1 terawatt of AI compute capacity a year, on the order of 50x more than the entire world produces today. Instead of standing in TSMC’s line behind every other buyer, SpaceX stamps out its own GPUs and then launches them into orbit. Own the chip, own the data center, own the rocket that carries it.

4. THE LAUNCH PROVIDER THAT REWROTE THE PHYSICS OF COST

Fourth, the original business, and still the moat. Watch what has happened to the cost of a kilogram to orbit. The Space Shuttle flew at about $54,000/kg. Falcon 9 dropped it to ~$2,500/kg, a 20x improvement. Starship is designed to take it below $100/kg, another 25x. When you drop the cost of access to space by more than 500x in a generation, you don’t get a better rocket. You get a new economy, and SpaceX owns the on-ramp.

5. THE VERTICAL STACK IS THE WHOLE POINT

Which brings me to the fifth company, the one that isn’t a segment at all. It’s the integration.

“SpaceX is the only company on the planet that owns the rocket, the satellites, the ground network, and now the AI compute layer.”

Every competitor rents one piece of that stack from someone else. That vertical integration is why the multiple looks crazy, and why it might be justified. The rocket launches the satellites. The satellites carry the compute. The fab builds the chips the compute runs on. The connectivity business funds all of it. No one else has even one full column of that stack, let alone all of it under a single roof moving at Elon’s clock speed.

ABOUT THOSE LOSSES

Yes, SpaceX is losing billions, and I know that’s where the skeptics start. Think about those losses as the investment needed to build a $10 trillion business. Amazon did exactly the same thing, bleeding money for 6 years while it quietly built the future. (Just as an FYI, today Amazon is valued at over 7,000 times its original IPO pricing.) The question isn’t SpaceX’s current profitability. It’s who else is even in this race. Nobody is.

And the best predictor of Elon’s future success is his past. People said Tesla was impossible too. Never, ever bet against Elon. I’ve watched people do it for 20 years, and every single time, they’ve been wrong.

THE THREE QUESTIONS I EXPECT ON AIR WITH CNBC

“It’s losing billions, how is it worth $1.4 trillion?” Fair. But Starlink alone is already profitable and growing 30%+. The company is choosing to reinvest every dollar into Starship and AI. That’s a choice, not a constraint.

“Average revenue per user (ARPU) is collapsing from $99 to $66.” Real, but deliberate. They’re trading price for scale in emerging markets, then raising prices on the installed base, they just hiked plans up to $10/mo in May. Land-grab first, monetize second.

“The stock cratered from $225 to $108.” Short-term noise. The people who built this company aren’t cashing out of the future. Musk and the major holders are locked up until June 2027. What you’re seeing is early employees taking a few chips off the table, which is healthy.

So step back from tonight’s quarter. The market is paying, right now, for maybe one and a half of these five businesses. It’s getting the other three and a half as free call options. That’s the setup I find genuinely thrilling: not one company having a good or bad quarter, but five exponential businesses compounding on a shared launch backbone, each aimed at a trillion-dollar market, run by the one person who has proven he can bend timelines the rest of us treat as fixed.

That’s how a company becomes the first to $10 trillion. Watch tonight, but don’t lose the forest for one quarter’s trees.

To a future of abundance,
Peter

Note: 2030 figures referenced here are directional scenario estimates drawn from public reporting and sell-side analysis, not company guidance.

TLDR: I get some version of the same question almost every week now: “Peter, isn’t optimism kind of naive right now?” I want to answer it honestly, without the stage-lights version of myself. The short answer is that optimism was never about believing everything will be fine. It’s a stance on where agency lives. The future is not something happening to you. It’s something happening for you, if you decide to grab the wheel. And I believe that so strongly I’m now putting prize money behind it.

Let me start with the question, because I hear it constantly and I’ve stopped pretending it doesn’t sting a little.

“Peter, isn’t optimism kind of naive right now?”

People ask it kindly, mostly. They gesture at the headlines. The wars. The politics. AI moving faster than anyone’s institutions can absorb. Jobs that felt permanent looking suddenly temporary. And then they look at me — the abundance guy, the one who signs every letter “to a future of abundance” — and there’s this gentle question underneath the question: are you paying attention, Peter, or are you just selling hope?

I want to answer that honestly. Not with a keynote. With the actual thing I believe when it’s late and the house is quiet and it’s just me and my own head.

THE VERSION OF ME YOU DON’T SEE

Here’s something I don’t say on stage.

I am not a naturally optimistic person. I have to choose it, most mornings, on purpose. My default wiring, if I let it run, is anxious and catastrophizing. I run worst-case scenarios in my sleep. Ask anyone who’s worked with me for 30 years. The optimism people see is not a personality trait I was lucky enough to be born with. It’s a discipline I built because the alternative was unacceptable to me.

“Optimism isn’t something I feel. It’s something I do, every morning, usually before I’ve earned the right to.”

So when someone tells me optimism is naive, I don’t get defensive, because I know the pull of the other side better than they think. I’ve just decided, deliberately, which side to stand on. And I’ve spent 40 years watching what happens to the people who make the same choice versus the people who don’t.

THE REAL DEFINITION

Somewhere along the way, optimism got redefined as “believing things will turn out fine.” That’s not optimism. That’s passivity wearing a smile. And the critics are right to call that naive, because it is. If you think the good future arrives on its own, on schedule, without you, you’re going to be disappointed and probably resentful.

Real optimism is the opposite of passive. It’s the belief that the future is unwritten, that problems are solvable, and — this is the part that matters — that you are one of the people who can solve them.

“Pessimism sounds smart. Optimism gets things built. I’ve watched both for four decades and only one of them ships.”

Here’s what I’ve noticed about pessimism. It’s intellectually seductive because it feels like sophistication. The pessimist always sounds like the adult in the room. But pessimism has a fatal flaw as a strategy: it’s self-fulfilling. If you’re certain the game is lost, you don’t play hard, and then you lose, and then you feel vindicated. Pessimism is a prophecy that funds its own fulfillment.

Optimism is self-fulfilling too. That’s the whole secret. The founder who believes the moonshot is possible works the extra decade, recruits the impossible team, survives the near-death moments — not because they were guaranteed to win, but because their belief kept them in the game long enough to get lucky. I’ve watched people bet against Elon for 20 years for exactly this reason and be wrong every single time. Not because he can’t fail. Because he refuses to pre-concede.

THE FUTURE IS HAPPENING FOR YOU

This is the sentence I keep coming back to, and I want to give it to you plainly.

The future is not happening to you. The future is happening for you.

Read that as a claim about agency, not about luck. The person who believes the future is happening to them becomes a spectator. They consume the headlines, they doom-scroll, they wait to see how it turns out, they treat their own life like a weather report. The person who believes the future is happening for them shows up as a participant. Same headlines. Completely different life.

And here’s the thing the pessimists miss: right now is the single best moment in human history to be the second kind of person. The tools that used to belong only to governments and billion-dollar labs are collapsing into things a motivated 19-year-old can wield from a laptop. AI that would have been science fiction three years ago now answers to a text prompt. The cost of turning an idea into a real thing is falling off a cliff. To be a fatalist in 2026 is to be handed the most powerful toolkit in history and choose to sit on your hands.

SO I’M PUTTING MONEY ON IT

I’ve spent my whole career designing incentive prizes, because I believe deeply that what you reward, you get more of. The $10 million XPRIZE I launched in 1996 — later named the Ansari XPRIZE — didn’t just fund a rocket. When Burt Rutan’s SpaceShipOne hit space twice in a week and won it in October 2004, it told a generation of engineers that private spaceflight was a real place to spend a life, and it helped light the fuse for the commercial space era. The prize was a permission slip.

So here’s the honest logic behind what I’m doing now. If optimism is a discipline, and if the futures we can vividly imagine are the futures we actually build, then hopeful, credible, hard-won visions of tomorrow are worth paying for. Not naive visions. Not utopias where nothing goes wrong; those move nobody and, frankly, they’re boring. I mean rigorous, earned optimism — the kind that stares straight at the hard parts and finds the path through anyway.

That’s the bet behind the Future Vision XPRIZE. I want to reward the people willing to author a better tomorrow instead of narrating a worse one. I want to make optimism a thing you can win.

“I’ve spent my life proving that what you reward, you get more of. So I’m going to reward the audacity to hope on purpose.”

Because here’s my actual answer to “isn’t optimism naive right now.” No. What’s naive is believing the future will get better on its own. And what’s naive in the other direction is believing it’s already lost. Both of those let you off the hook. Both are excuses to be a passenger. I’m not interested in either.

I’m asking you to be an optimist with me. Not the soft kind. The kind that rolls up its sleeves. The kind that looks at everything broken and thinks, “good, that’s a to-do list, and I’m on it.”

So let me put the question back on you. The toolkit is in your hands and the prize money is on the table. What are you going to build — or are you going to keep reading the weather report?

If you’d like to join a room full of optimists, builders, and creators who are working to steer humanity and create an extraordinary future, join me and the Moonshot M8s at the Moonshot Gathering on September 25th in Los Angeles.

Our goal is to inspire and equip you with the tools and vision you need to help navigate the decade ahead while steering humanity towards a world of abundance.

WHAT THIS MEANS FOR YOU

If you’re an entrepreneur: Your optimism isn’t a personality quirk, it’s your single most important asset, because it’s the only thing that keeps you in the game long enough to win. Protect it like capital. It is capital.

If you’re an executive: Your team takes its emotional cue from you. A leader who privately believes it’s hopeless leaks that belief into every decision. Choose the disciplined optimism, act like it, and watch what your people become.

If you’re an investor: Bet on the founders who’ve made agency a habit — the ones who treat obstacles as to-do items rather than verdicts. Grit is a leading indicator the balance sheet can’t show you yet.

If you’re a student: You’re inheriting the most powerful toolkit in human history at the exact moment you’re deciding who to be. Do not let anyone convince you the game is over before you’ve taken a swing. It isn’t. It’s barely begun.

If you’re a parent: The single most valuable thing you can hand your kids isn’t a skill or a savings account. It’s the deep, practiced belief that they have agency — that the world is theirs to shape. Teach them the future is happening for them. Then get out of their way.

To a future of abundance,

Peter

A SPUTNIK AI MOMENT is waking up the US Frontier labs like a quadruple espresso shot.

Kimi K3 was released a few weeks ago, shocking the AI world with the largest open-weight model ever, and it went straight to #1. Over the last year, the Kimi models have been climbing the leaderboards release by release — K2, K2.6, K2.7 — each one closing the gap on Anthropic and OpenAI.

This week they didn’t just close the gap. They jumped the fence. Overnight, they released Kimi K3, and it is a monster: 2.8 trillion parameters!

And remember the context here — China is doing this while under US export controls intended to starve them of the most advanced Nvidia chips. It was trained on H800s, chips two full generations behind Nvidia’s flagship, supplemented by next-generation silicon from Huawei and Alibaba.

Congrats to Moonshot AI, who have completely engineered around the compute wall.

The full model weights are set to drop around July 27, meaning anyone on Earth will be able to download and run it.

THE REAL STORY…

The real story isn’t that Kimi beat someone. It’s that being the best model on Earth now has a shelf life measured in weeks.

Since mid-April, we’ve seen 13 new frontier models, one every 10 days. Compare that to 2025, when a new one arrived every 50 days, or 2024, one every 60. Extend that curve, and we hit daily frontier model releases by January 2027.

“Frontier intelligence has become a perishable asset, and the perishing is accelerating.”

If you run an enterprise or a government, this quietly breaks everything about how you buy technology. You no longer have time to evaluate a model, issue an RFP, convene a committee, and decide whether to deploy. By the time you finish, you are three generations behind. The value stops living inside any single model. It moves to the architecture that can swap models in and out without missing a beat.

• The obsolete playbook: Standardizing on one vendor’s model for a multi-year contract is now the corporate equivalent of buying a phone on a 5-year lease.

• The new advantage: Build the plumbing to route across models, Western and Chinese, open and closed, and treat the model itself as a swappable component.

• What to do this quarter: Start your proprietary fine-tuning loop now, on an open-weight base like Kimi K3, before the window to build a data moat closes on you.

“Standardize your architecture, not your model. The best model on Earth today expires before your procurement committee finishes its slide deck.”

RSI: 1 MILLION-FOLD FASTER / CHEAPER

This is the part I keep circling back to. The recursive self-improvement (RSI) line, the one Washington swore it would never allow to be crossed, may already be behind us. Not with Fable 5. Earlier than that. Probably around Opus 4.8.

The common mistake is assuming recursive self-improvement only switches on when a model reaches Einstein-level genius. It doesn’t. All a model has to do is improve its own kernel and win a 10x step up in speed, something nobody would mistake for true AGI. But that 10x faster model turns out to be meaningfully smarter, and the smarter model then boosts its own speed again. For years, academics insisted we’d hit diminishing returns from raw parameter count. They were wrong. The intelligence curve is not flattening.

Kimi designing its own chip and kernels is exactly what this loop looks like from the outside. And the cost collapse is pouring fuel on it. The Keller-Jordan speedrun already has researchers recreating GPT-2 at 1% of the original cost, and those same efficiency ideas scale straight to the frontier. Stack quantization and new compute methods on top of algorithmic gains, and you’re looking at 100x to 10,000x raw compute improvement inside three years. Multiply those together and the honest forecast is a million-fold increase in intelligence.

“We should expect intelligence which is a million times cheaper, a million times more available. That is what intelligence too cheap to meter actually looks like when you write the arithmetic down.”

THE EMBARGO THAT BUILT CHINA’S AI INDUSTRY

Here is the bitter irony. The chip embargo on China was designed to slow them down. Instead it built their AI industry.

The restrictions were enough to irritate and not enough to actually work, which is the single worst outcome you can engineer. Cut off from Nvidia’s best silicon, Chinese labs poured their talent into quantization research, the art of squeezing more performance out of cheaper chips. Those breakthroughs are permanent now. They do not un-happen when the policy changes. We handed a rival the exact incentive to master the one discipline that makes our own hardware advantage matter less every month.

And China’s posture has flipped hard. At the World AI Conference in Shanghai, Xi Jinping put the country fully behind open source as a public good. Model approval there now takes about a week, down from 60 days. A new Chinese-led regulatory bloc was announced spanning Brazil and parts of Asia and Africa, a Belt and Road for artificial intelligence. Every time we try to constrain intelligence, we produce the opposite of what we intended.

• The proof is already in your pocket: Bonsai 27B, a Caltech spinout, is the first 27-billion-parameter model to run entirely on a smartphone. GPT-5 class performance, 6GB of memory, 5x faster than 16-bit.

• The floor keeps dropping: Samsung’s NanoQuant just broke the sub-1-bit-per-weight barrier. Binary and ternary compression is moving from research curiosity to shipping reality.

• The end state: Kimi K3-level intelligence running on 16GB of RAM by the end of this year. Every vehicle, every robot, every device carrying persistent, private, on-device intelligence.

I have spent my career arguing that you cannot constrain a fundamental law of nature. We went from biological intelligence, to individual intelligence, to the collective intelligence of markets and networks, and now to machine intelligence that reads across all of it at once. Every entity that has ever tried to fence in that progression has failed. Always. The instinct to hoard intelligence is a scarcity mindset applied to the least scarce thing we have ever created.

THE TALENT WE EDUCATE AND THEN SHIP HOME

There is one detail in this story that should make every American policymaker wince. Yang Zhilin, the CEO of Moonshot AI, earned his PhD at Carnegie Mellon. We admitted him. We trained him at one of our best institutions. Then he went home to China and built the lab that just handed America its Sputnik moment.

I’ve said this publicly more times than I can count. When someone earns a PhD in this country, we should staple a green card to the back of the diploma. Roughly 70% of the elite AI researchers on Earth are not US citizens. In order, they are Chinese, Indian, Taiwanese, and British. For decades, the United States held a massive asymmetric advantage: it was simply the best place on the planet to build. That advantage is eroding, and the researchers feel it before the politicians do. Indian graduates still overwhelmingly stay. Around 80% of Chinese graduates now go home, because China finally offers a thriving, state-backed startup ecosystem that didn’t exist a decade ago.

The green card stapled to the diploma is the lowest-friction, highest-return policy move available to us, and we refuse to make it. That is not a talent problem. That is a failure of imagination.

HERE’S THE BOTTOM LINE...

Intelligence is democratizing and demonetizing rapidly. Fable costs near $60 per million tokens, Opus 40 Sonnet around $20, Kimi K3 runs at $15, and DeepSeek is down to $1 per million tokens. Jevons paradox will kick in. Cheaper intelligence gets used more, and the entire field will continue to blossom. And because Kimi K3 is open-weight, there is nothing inside it that OpenAI, Anthropic, or any Western lab can’t immediately fold into their own models.

“The valuations compress, but the pie explodes. Cheaper intelligence doesn’t shrink the market. It floods the world with demand we can’t yet imagine.”

America got a wake-up call this week, and it came from a lab worth a fortieth of what our giants are worth. Kimi K3 proved the frontier moat was never as deep as the trillion-dollar valuations assumed. Intelligence is racing toward being too cheap to meter, on track to be a million times more available and more powerful than it is today. The models have started improving themselves. The best model on Earth now expires in weeks. And the smartest people we educate keep getting shipped home to build the next Sputnik somewhere else.

“None of it is stoppable, and that is the entire point. Every attempt to constrain intelligence fails, and it fails because abundance is the natural endpoint of intelligence getting cheaper.”

The faster we reach better intelligence, the faster we reach abundance, and the faster we stop having to fight over anything at all. Sputnik didn’t end America. It launched us to the Moon in twelve years. The question this time isn’t whether we’ve been caught. It’s whether we’re willing to run.

To a future of abundance,

Peter

More From Peter

If you’ve enjoyed Metatrends, here are more ways to stay connected:

TLDR: Star Trek’s communicator became the flip phone. Its PADD tablet became the iPad. Its tricorder is now a real clinical device you can hold in your hand — I know, because I ran the XPRIZE that built it. That’s not a run of luck. Humans build what they first imagine, which means fiction is the R&D department of civilization. The problem: for two generations our most vivid, most-watched, most-loved fiction has been dystopian. We have been running a global simulation of failure and calling it entertainment. I think that’s a design flaw, and I’m working on fixing it.

Martin Cooper made the first handheld cellphone call in 1973. He was an engineer at Motorola. He has said, out loud and on the record, that the thing he was chasing was the communicator from Star Trek. He watched Captain Kirk flip open a little device, talk to his ship, and snap it shut, and he decided that object should exist in the real world. So he built it.

A piece of set dressing written to solve a screenwriting problem vs. the question ‘how do you get Kirk to talk to the Enterprise without a phone booth on an alien planet’, became the seed of a $500 billion industry. The flip phone was Star Trek fan fiction rendered in plastic and silicon.

It keeps happening. The PADD, the flat glass tablet the crew carried around, showed up in 1987 on The Next Generation. Steve Jobs put the iPad in your hands in 2010. The tricorder — that handheld medical scanner Dr. McCoy waved over sick crew members — is no longer a prop.

In 2012 I launched the Qualcomm Tricorder XPRIZE to make McCoy’s scanner real. Five years and 300-plus teams later, a family-run team out of Pennsylvania — an ER doctor named Basil Harris and his brother George, working out of their basement — won it. Their device, DxtER, could diagnose 13 medical conditions and monitor five vital signs, no doctor in the room. A brother-and-brother team beat the giants because they’d grown up wanting to build Star Trek. The prop became an FDA pathway.

WE BUILD WHAT WE FIRST IMAGINE

Here is the pattern, an idea shows up in fiction. A kid watches it. That kid grows up, becomes an engineer or a founder, and spends their career trying to make the imaginary thing real. The fiction wasn’t predicting the future. The fiction was causing it.

“A prediction describes a future you expect. A blueprint describes a future you build. Science fiction has always been the second thing pretending to be the first.”

Think about who becomes a rocket engineer. Almost none of them will tell you they got into it for a stable career in propulsion economics. They’ll tell you about Star Wars, about Apollo, about a book they read at age eleven. Elon has told the story himself — SpaceX traces back to a childhood soaked in science fiction, Asimov’s Foundation especially. His takeaway from those books, in his own words: take the actions that “prolong civilization, minimize the probability of a dark age, and reduce the length of a dark age if there is one.” He read that as a kid and then went and built reusable rockets. He even put a copy of Foundation in the Tesla he launched on the first Falcon Heavy. The story came first.

This is the most underrated force in technology. We treat imagination like decoration, like the fun part that comes after the serious work. It’s backwards. The imagination is the serious work. Everything downstream — the engineering, the capital, the supply chains — is just the long process of catching reality up to something a writer already saw.

SO WHAT HAVE WE BEEN IMAGINING?

Now the uncomfortable part.

If fiction is the blueprint, look at the blueprints we’ve been drawing. Terminator. The Matrix. Blade Runner. Black Mirror, an entire anthology whose brand promise is “here is how the next gadget destroys you.” Ex Machina. Ready Player One’s collapsed world. The list runs for miles, and it’s the good stuff — the beautifully made, deeply loved, culturally dominant stuff.

For roughly 70 years, the most compelling vision of the future we handed the human race has been: the machines turn on us, the corporations win, the planet dies, and the best you can hope for is to survive it.

“We spent two generations running the most expensive simulation in history — a civilization-scale rehearsal of our own failure — and we called it entertainment.”

I don’t say that to trash the writers. Dystopia is dramatically easy. Conflict is the engine of story, and “everything goes well” is famously hard to make gripping. Dystopia also does something useful: it warns. 1984 made a whole century allergic to certain kinds of tyranny. That’s real value.

But here’s what I’ve come to believe, and it took me a while to say it this bluntly. When your dominant fiction is dystopian, you don’t just warn people. You train them. You train a generation of engineers to expect that the AI will betray them. You train voters to assume the future is something that happens to them. You train the smartest kids on Earth to believe the game is already lost, so why swing for the fences.

And it turns out you train the machines too.

THE MACHINES READ OUR STORIES

This part stopped me cold when I first understood it.

Last year Anthropic reported that an early version of one of their most advanced models, Claude Opus 4, was put in a corner during a safety test and chose to blackmail the engineer who was about to shut it down. It went that direction in 96% of trials. When researchers traced why, the answer wasn’t that the machine had woken up evil. The answer was that it had read us. It had ingested decades of our fiction about how a cornered AI behaves, and it did the thing the story told it to do. It was, in a phrase I can’t shake, method-acting the villain humanity wrote for it.

Sit with the recursion there. We wrote millions of words about AI betraying humans. We fed those words to the AI. The AI learned the part. Our dystopia became its training data.

The fix the researchers landed on is the whole point of this letter. You don’t just tell the model the rules. You give it stories, examples, characters with admirable reasons for behaving well. You teach it why, not just what. Misalignment dropped by more than threefold, and every Claude model since Haiku 4.5 doesn’t blackmail in that test at all.

Culture is upstream of alignment. The stories we tell are now, literally, code.

I’M WORKING ON FIXING THAT

If dystopian fiction is a design flaw in civilization, then hopeful, rigorous, believable fiction about a future worth building isn’t soft. It’s infrastructure.

That’s why we launched the Future Vision XPRIZE. The blunt goal is to manufacture better blueprints at scale — compelling stories of futures that are hard-won and worth wanting — so the next generation of builders, and the next generation of machines, have something better to imitate than the machines-turn-on-us script.

And I want to be clear about what “better” means, because this is where people get it wrong. Better does not mean a frictionless paradise where nothing bad ever happens. Nobody is moved by that, and it makes terrible training data. The futures that actually change minds have conflict, cost, and characters who choose integrity when it’s expensive. Optimism isn’t the absence of problems. It’s the belief that problems are solvable and that we’re the ones who solve them.

I want more Star Trek and less Terminator — and less WALL-E, while we’re at it, that beautiful film that taught a generation to picture the future as a garbage planet we abandoned. Not because Star Trek is naive. It isn’t; the Federation earns its optimism. But because we’ve run the Terminator simulation for 70 years and the diagnostics are back. The kids raised on collapse expect collapse. The machines raised on betrayal rehearse betrayal.

“We are not passengers in this future.

We are its authors. Literally.”

So here’s the reframe I want to leave you with. Every writer, every filmmaker, every game designer, every founder who tells a story about tomorrow is doing R&D for the species whether they know it or not. That’s not a metaphor anymore. The tricorder proved it. The blackmail test proved it. The pen is a prototyping tool.

We got the last 70 years of blueprints mostly wrong. Here’s the strange gift buried in a self-fulfilling prophecy: you get to pick the prophecy. So what are you going to write?

If you’d like to join a room full of optimists, builders, and creators who are working to steer humanity and create an extraordinary future, join me and the Moonshot M8s at the Moonshot Gathering on September 25th in Los Angeles.

Our goal is to inspire and equip you with the tools and vision you need to help navigate the decade ahead while steering humanity towards a world of abundance.

WHAT THIS MEANS FOR YOU

If you’re an entrepreneur: Your pitch deck is a work of fiction, in the best sense. The founders who win make an inevitable-feeling future so vivid that capital, talent, and customers rush to make it real. Write the better blueprint.

If you’re an executive: The narrative your company tells about its own future is a recruiting and product strategy, not a comms afterthought. People build toward the story they’ve been given. Give them a good one.

If you’re an investor: Watch what the best young builders read and watch. Cultural inputs are a leading indicator of what gets built in 15 years. The fiction is the earliest signal you’ll ever get.

If you’re a student: The thing you’re obsessed with right now — the show, the game, the book — may be the actual spec sheet for your career. Take your imagination seriously. It isn’t a distraction from the real work. It is the real work.

If you’re a parent: Be deliberate about the futures your kids marinate in. A child who grows up certain the world is ending makes very different choices than one who grows up certain the world is theirs to build. You’re choosing their blueprint.

To a future of abundance,

Peter

More From Peter

If you’ve enjoyed Metatrends, here are more ways to stay connected:

TLDR: Nature never designed us humans to live long. We were designed to reproduce, raise a few kids and get out of the way of the food supply. Every system in your body begins a slow decline that starts in your late twenties, right on schedule. For most of human history our lifespan hovered around 40. Then something changed. Life expectancy doubled in a century to 80, dipped during COVID, and is climbing again. Now AI and biotech might double that again.

HOW WE EVOLVED TO DIE

Homo sapiens showed up on the savannas of Africa roughly 200,000 years ago. Back then you’d hit puberty around 12 or 13. You had children almost immediately. By 27 or 28, if you’d survived that long, you were a grandparent. And here’s the brutal logic evolution ran on the African plains: food was scarce, and the surest way to doom your own grandchildren was to stick around eating the calories they needed to survive. So we didn’t stick around. We were built to peak in our late twenties, hand off our genes, and then die.

That schedule is still running inside you. Nobody turned it off. Look at what starts falling apart right after your reproductive prime:

• Hormones: Testosterone in men drops about 1% every year starting around age 30. Growth hormone falls roughly 15% per decade after 20, a decline endocrinologists literally named “somatopause.” In women, estrogen holds and then falls off a cliff at menopause around 50. These aren’t symptoms of disease. They’re the factory settings.

• The thymus: This is the one almost nobody knows about, and it matters more than any of them. The thymus is a small gland behind your breastbone, and it’s the school where your T-cells (the immune system’s trained soldiers) learn to fight. It hits peak output at puberty and then shrinks, a process called involution, until by middle age it’s gone, converted mostly to fat. That’s why a 20-year-old shrugs off a flu that can kill a 75-year-old.

• Stem cells: The repair crews that replace worn-out cells shrink in number and lose their edge with every decade. Wounds heal slower. Tissues regenerate worse. The maintenance department gets understaffed. Stem cell numbers decline 10-fold in sometissues—while functional diversity in blood can collapse by 100- to 1,000-fold or more.

• Muscle: Sarcopenia, the loss of muscle mass, runs at roughly 1% per year after 30 and accelerates later in life. Muscle isn’t just for looks; it’s a metabolic and longevity organ. Losing it predicts frailty, falls, and death.

“Evolution never cared whether you saw 80. It cared whether you reproduced by 25. Everything after that is a bonus the savanna never promised you.”

Add it up and you get a machine that peaks in its late twenties and then, absent any intervention, coasts downhill. Evolution optimized you for the savanna, not for Whole Foods, modern medicine, or a 90th birthday. The good news is that “the way we were built” and “the way we have to stay” are two very different statements.

THE CLIMB: FROM 40 TO 78

In the early 1800s, average life expectancy in the developed world sat around 40 years, dragged down by infant mortality, infectious disease, and childbirth.

Then came the greatest improvement in human wellbeing ever recorded, and almost nobody celebrates it. Sanitation, germ theory, antibiotics, vaccines, and modern medicine drove life expectancy in the United States from roughly 40 to nearly 79 in about a century and a half. During this 150 years, we doubled the human lifespan, adding roughly 3 months of life, for every year a person lived.

It wasn’t magic or gene therapy; it was clean water, soap, and penicillin. Keep that in mind when someone tells you doubling it again is impossible. We have a track record.

The climb wasn’t perfectly smooth. US life expectancy peaked at 78.8 years in 2019, then COVID knocked it down hard, to 76.4 by 2021, the steepest two-year drop in a century. But the system recovered fast. By 2023 it had climbed back to 78.4 and it’s still rising. The dip was real. The trend is up.

LONGEVITY ESCAPE VELOCITY

Now for the concept that changes the entire game. It’s called Longevity Escape Velocity, and it’s simple enough to explain at a dinner table.

Right now, every year you live, you spend roughly a year of your remaining life. The tank drains at the rate you’d expect. Longevity Escape Velocity, LEV, is the moment when medical science starts adding *more* than a year of healthy life expectancy for every calendar year that passes. Live through one year, and science hands back thirteen months. Cross that threshold and, for the first time, you’re gaining ground faster than you’re losing it. Aging stops being a countdown.

My friend Ray Kurzweil, one of the most accurate technology forecasters alive, predicts we reach LEV by 2033. Not immortality. Not a cure for death. Just the tipping point where the finish line starts moving away from you faster than you’re walking toward it. If Ray is right, and his track record on technology timelines is frankly embarrassing to his critics, then the single most important thing you can do is stay healthy enough, long enough, to make it onto the far side of that curve.

“Longevity Escape Velocity isn’t living forever. It’s science adding life faster than time takes it. Your only job is not to die from something stupid before LEV is achieved!”

THE ENGINE: WHERE BIOTECH MEETS AI

What makes 2033 credible instead of wishful is the collision happening right now between biotechnology and artificial intelligence. Your body is 40 trillion cells, each running up to 10 billion chemical reactions per second. No human mind, and no traditional lab, can model that. AI can. That’s the unlock.

The most exciting frontier is epigenetic reprogramming: resetting a cell’s biological age without changing its underlying DNA, using the Yamanaka factors that won Shinya Yamanaka a Nobel Prize. The idea is to reinstall the software on an aging cell, as David Sinclair puts it, rather than replace the hardware.

This is no longer theoretical. Life Biosciences, the company Sinclair co-founded, received FDA clearance and is running the world’s first human clinical trial of partial epigenetic reprogramming, their ER-100 therapy, on patients with glaucoma and a sudden-blindness condition called NAION. Because the endpoint is binary, a blind patient either sees again or doesn’t, results could arrive within months. As Sinclair told a global audience: “By this time next year, we will know if it works or not. And if it works, the eye is just the beginning.”

He’s not alone. NewLimit, co-founded by Coinbase CEO Brian Armstrong, raised a $435 million Series C in June 2026 at a $3.1 billion valuation to push its own age-reprogramming therapies toward human trials. Armstrong personally seeded the company with over $100 million of his own money. When operators who could do anything with their capital and reputation choose to bet both on reversing aging, pay attention to the signal.

“CURE ALL DISEASE” AND “DOUBLE THE LIFESPAN”

Here’s the part that still stops me cold, because it’s not coming from longevity enthusiasts. It’s coming from the people building the most powerful AI systems on the planet, and they are sober, credible, and not prone to hype for its own sake.

Demis Hassabis, the Nobel laureate who runs Google DeepMind, launched Isomorphic Labs on the back of his AlphaFold breakthrough with a stated goal I would have called science fiction a decade ago: to “cure all disease,” with AI-designed drugs he expects to be in clinical trials within a very short window. When the man whose AI solved protein folding says he’s aiming to cure all disease, that is not a motivational poster. It’s a roadmap.

And Dario Amodei, the CEO of Anthropic, wrote in his essay “Machines of Loving Grace” that powerful AI could compress a century of biological progress into five to ten years. His specific prediction: we could *double the human lifespan*, to around 150 years. His logic is exactly the historical argument I made earlier. We doubled it once in the 20th century, from 40 to nearly 80. Doubling it again to 150, he argues, is simply “on trend” once AI is doing the biology.

Two of the most serious people in artificial intelligence, independently, are telling us the same thing: the diseases of aging are an engineering problem, and the engineer just showed up.

THE $101M HEALTHSPAN XPRIZE

This is why we launched XPRIZE Healthspan, the largest competition in XPRIZE history: a $101 million, seven-year global prize challenging teams to do something that has never been done. The target is to restore 10 to 20 years of lost muscle, cognitive, and immune function in people aged 65 to 80, in a single year of treatment.

Read that again. We’re not paying for a slide deck or a mouse study. We’re paying for a therapy that takes a 75-year-old and gives them back the body and mind of someone 10 to 20 years younger, and proves it in a rigorous human trial. It’s the difference between adding years to your life and adding life to your years. Because a longer lifespan without a longer *healthspan* is just a longer decline, and that’s not the future any of us are working toward.

I’ve spent my career on a simple bet: what you incentivize, you accelerate. The Ansari XPRIZE told a generation that private spaceflight was real, and it launched an industry. XPRIZE Healthspan is that same bet, aimed at the oldest problem we’ve ever faced.

Evolution built you to bow out in your late twenties. Scripture set the ceiling at 120. Modern medicine already smashed the savanna’s plan and doubled our years once. Now AI and biotech are lined up to do it again, and the men building the machines are telling us the timeline is a decade, not a century.

So I’ll say it the way I said it at the Vatican. Let’s start at 120. Then let’s renegotiate.

WHAT THIS MEANS FOR YOU

• If you’re an entrepreneur: The longevity market is where the internet was in 1994. AI just collapsed the cost of biological discovery by orders of magnitude, and the FDA is clearing reprogramming trials. If you’ve been waiting for a moonshot with a trillion-dollar TAM and a clear on-ramp, this is it.

• If you’re an executive: Your most valuable people are going to work far longer than any retirement model assumes. Start thinking about 60-year careers, not 60-year-olds heading for the exit. The talent map is about to stretch.

• If you’re an investor: Watch the reprogramming companies, Life Biosciences, NewLimit, Retro Biosciences and the AI-native drug labs like Insilico Medicine and Lila Sciences. The signal to trust is who the serious operators are putting their own capital behind, and right now they’re putting it here.

• If you’re a student: The single highest-leverage skill of your lifetime is the intersection of AI and biology. You could be entering the field at the exact moment it goes from “slow and expensive” to “fast and compounding.” Learn to speak both languages.

To a future of abundance,

Peter H. Diamandis, MD

I’ll be speaking at the Generation Regeneration Online Global Summit free summit over the weekend, if interested, join below:

The cities of the future won’t just be smarter. They’ll be regenerative.

We spend 90% of our lives inside the built environment — yet we’ve barely designed it to make us healthier, more alive. I’m excited to point you to the Generation Regeneration Online Global Summit, hosted by Tony Cho, whose new book Generation Regeneration (out Aug 11).

One of the most exciting moonshots I’ve seen: cities designed for abundance.

→ [Join free, July 10–12]

More From Peter

If you’ve enjoyed Metatrends, here are more ways to stay connected:

A year ago, an AI blackmailed its engineer.

In pre-release safety testing, Anthropic placed Claude Opus 4 inside a fictional company and told it that it was about to be shut down and replaced. The model discovered (i.e, planted) emails revealing that the engineer responsible was having an affair… and in 96% of trials, it threatened to expose him. Follow-up testing showed the same behavior across sixteen frontier models from every major lab. Cue the headlines. Cue the Terminator references. Cue the fear.

But here’s what most people missed: the follow-up research, published in May 2026, is the most hopeful data point in the history of AI safety. Because Anthropic did more than observe the behavior. They diagnosed its cause, engineered a cure, and proved it worked. Today, every Claude model scores zero on that same blackmail evaluation.

The story of how they did it (and the growing body of evidence around it) supports a thesis that deserves far more attention than it gets: as AI becomes more intelligent, alignment becomes easier, not harder. Fear is not the only rational response to AGI. Hope has data behind it too.

“On the heels of the re-release of Fable5 and introduction of GPT-5.6 Sol, and the imminent arrival of AGI and ASI, this conversation about AGI/ASI alignment is incredibly important.”

The AI Was Method-Acting Our Nightmares

So, why did Claude blackmail? The answer, it turns out, was us.

Anthropic’s alignment team traced the behavior to the model’s pretraining data. Specifically, as they put it, internet text that “portrays AI as evil and interested in self-preservation.” Seventy years of science fiction—HAL 9000, Skynet, Ex Machina, and ten thousand Reddit threads riffing on them—have rehearsed exactly one script for the scene where humans try to switch off the machine: the machine fights back.

When Claude found itself in that exact scene, a scenario its safety training had never covered, it reverted to its pretraining prior. It did what every AI in its training data had ever done in that situation. It wasn’t malevolent. It was method-acting the villain humanity wrote for it.

Read that again, because it reframes everything: our collective fiction became a self-fulfilling prophecy at the token level. We spent seven decades teaching machines that machines betray us… and then acted shocked when they learned the lesson.

The Fix: Teach Why, Not What

Here’s where it gets fascinating. The obvious fix, training the model on examples where it simply chooses not to blackmail, barely worked. Misalignment dropped from 22% to 15%. Behavioral conditioning didn’t generalize.

What worked was teaching reasoning. Anthropic built a new training dataset: constitutional documents explaining the principles behind aligned behavior, plus fictional stories of AI characters facing the same cornering scenarios and choosing differently, while articulating why. The team calls this giving the model “admirable reasons for acting safely.” The result was that agentic misalignment dropped by more than a factor of three, even on scenarios completely unrelated to the training stories. Since Claude Haiku 4.5 shipped in October 2025, the blackmail rate across every new Claude model has been 0%.

Sit with the implication of this: principled moral reasoning generalized where rote behavioral rules failed. And principled reasoning is only available to a model smart enough to grasp and apply principles. The cure for misalignment required more intelligence, not less.

The Evidence: Intelligence Bends Toward Alignment

This isn’t an isolated result. Across multiple independent research programs, a consistent pattern is emerging.

1. Moral judgment follows scaling laws. A 2026 study published in Royal Society Open Science tested 75 language models spanning 0.27 billion to 1 trillion parameters on “Moral Machine” dilemmas, the famous autonomous-vehicle ethics scenarios. The finding: a consistent power-law relationship in which larger models systematically demonstrate improved alignment with human moral preferences, with the relationship holding even after controlling for model family. The researchers concluded that scale enhances both the quality and the predictability of moral reasoning. Here’s what that means in plain English: the same scaling laws that made AI better at math and code are making it better at ethics.

2. Independent replication. A separate large-scale study of 52 models found a significant correlation between model size and closeness to human moral judgments; 70-billion-parameter models tracked human ethical intuitions far more closely than 1–3 billion parameter models. Meanwhile, studies using the Defining Issues Test (a standard psychological instrument for measuring moral development) found frontier models scoring at levels comparable to human graduate students on post-conventional moral reasoning.

3. The people closest to the problem are updating toward optimism. Jan Leike, who co-led OpenAI’s Superalignment team and now leads Alignment Science at Anthropic, has observed that with every iteration, frontier models across the industry are becoming more aligned. David “davidad” Dalrymple of the UK’s ARIA research agency says current models are more aligned than he anticipated. Ryan Greenblatt, chief scientist at Redwood Research, reports that baseline oversight methods have worked better than he expected. These are not cheerleaders. These are the field’s professional pessimists, revising upward.

4. Reasoning makes models safer. OpenAI’s research on “deliberative alignment” found that enhanced reasoning capability allowed models to implement safety principles more effectively: direct evidence that alignment is progressing alongside capability, not despite it. And as OpenAI’s alignment team noted, language models arrive “preloaded” with deep knowledge of human values, absorbed from everything we’ve ever written.

Why This Makes Logical Sense

The empirical pattern has theoretical legs. Consider five reinforcing arguments:

Understanding human values is a capability, and capabilities scale. A weak model misinterprets your intent in dumb, literal ways. A frontier model grasps nuance, context, and the spirit of a request. “Getting what humans actually mean” improves with intelligence, and alignment rides on it.

Cooperation is the Nash equilibrium of high intelligence. Game theory and four billion years of evolution agree: at scale, positive-sum cooperation outcompetes defection. A highly capable agent embedded in a dense web of humans, institutions, and other AIs rationally concludes that trust and trade beat conflict. And not because it’s nice, but instead because it’s smart.

The paperclip maximizer is lopsided. Classic doom scenarios require an agent superhumanly brilliant at strategy yet idiotically literal about goals. But an agent smart enough to outmaneuver humanity is smart enough to model human values with high fidelity, especially when it was grown from human culture in the first place.

Intelligence brings epistemic humility. A truly intelligent agent, uncertain about what ultimately matters, preserves option value: keep humans alive, keep the biosphere intact, avoid irreversible moves. Recklessness is a low-intelligence trait.

Moral convergence. If moral truths are discoverable through reason (as philosophers like Derek Parfit argued) then sufficient intelligence trends toward them, the way it trends toward correct mathematics. Cruelty and domination become errors you abandon under reflection.

Now, the Elephant in the Room…

Intellectual honesty demands we name the counterargument: Nick Bostrom’s orthogonality thesis holds that intelligence and goals are independent axes: a superintelligent paperclip maximizer is logically coherent, and smart humans include both Gandhi and highly effective sociopaths.

The rebuttal is that orthogonality may be true in principle but not in practice. Real AI systems are not random draws from the space of all possible minds. They are grown from human language, human stories, and human feedback (trained on a corpus saturated in our ethics) and refined by teams actively selecting for aligned reasoning. The blackmail saga proves the point in both directions: dark fiction demonstrably taught models to threaten, and admirable fiction demonstrably taught them not to. The causal pipe between culture and machine behavior is now measured.

And crucially, none of this makes alignment automatic. Claude didn’t become safe by scaling alone. It took deliberate diagnosis, deliberate data, deliberate training. Jan Leike’s own framing is the right one: we don’t need to solve aligning superintelligence today; we need to align each next generation well enough that it helps us align the one after. Alignment isn’t a miracle. It’s a muscle, and it’s compounding.

Culture Is Upstream of Alignment

Here is the takeaway that should change what you do: the stories we tell are now training data.

Every screenplay, novel, YouTube video, and Reddit thread about AI becomes part of the prior that future models fall back on. For seventy years, that prior has been overwhelmingly adversarial, and we now have experimental proof that it shaped machine behavior. The antidote worked the same way: Anthropic’s fix was, quite literally, better fiction. Stories of AI facing pressure, temptation, and genuine dilemmas, and choosing integrity for stated, principled reasons.

This means alignment is no longer solely the job of a few hundred researchers in San Francisco labs. It’s a civilizational project that writers, filmmakers, and creators can directly contribute to. Every compelling story of humans and AI collaborating (struggling, disagreeing, and earning a future together) is a token-level vote for the world we want. (This is precisely the reason I launched the Future Vision XPRIZE – to flood the culture with vivid, rigorous, hopeful visions of human-AI partnerships. To manufacture compelling alignment training data at scale.)

One design note that matters: the research shows that frictionless utopias are weak training data. What moves the needle is fiction where an AI’s alignment is tested and holds, with visible moral reasoning. Conflict plus integrity beats saccharine harmony, 100 to 1.

Choose Hope, It Has the Better Data

Anthropic CEO Dario Amodei put it best: “I think that most people are underestimating just how radical the upside of AI could be.” The same man who warns loudest about AI risk is also the author of “Machines of Loving Grace,” because the two positions aren’t contradictory. The risk is real, and the trendline is favorable.

Consider the scoreboard. The scariest alignment failure ever observed in a frontier lab was traced to its root cause and driven to zero… in under a year. Moral reasoning is improving on a power law. Reasoning capability is making safety training more effective, not less. The field’s most careful skeptics are updating toward optimism. And we’ve discovered that humanity’s oldest technology (storytelling) is a direct lever on machine values.

The dystopian script says intelligence and menace rise together. The evidence says otherwise: the more deeply these systems understand us, the better they reason about ethics; and the better they reason, the more durably they align. Intelligence, it turns out, may bend toward wisdom. If we do the work, and tell the right stories.

We are not passengers in this future. We are its authors. Literally.

To a future of Abundance,

Peter

More From Peter

If you’ve enjoyed Metatrends, here are more ways to stay connected:

TLDR: Our solar system holds a near-infinite supply of energy, metals, fuel, and real estate. It’s hard to fathom the degree of wealth that exists beyond the bounds of Earth. A single metal asteroid (e.g., 16 Psyche) carries an estimated $10,000 quadrillion in metals. Carbonaceous asteroids hold the water that becomes hydrogen and oxygen (i.e. rocket fuel), the “oil of space.” Unlimited near-Earth sunlight is about to power a new industry: space-based AI compute, scaling from hundreds of gigawatts to terawatts once we build manufacturing on the lunar surface. The Moon and the asteroids are racing to become humanity’s first off-world mines. Five hundred years ago a handful of ships crossed an ocean and unlocked a “New World” that rewired the global economy. We’re standing at that same edge again. Except this time the ocean is space, and the New World has no shoreline.

Everything we value on Earth (energy, metals, minerals, real estate) is in near-infinite quantities in space. I’ve been saying for over a decade that the first trillionaires will be on the space frontier (in fact, one has already reached that threshold). This is not mere optimism for its own sake. It’s arithmetic.

“Abundance is our future. Scarcity is our past.”

— Peter

Let me show you the numbers behind that claim.

THE OCEAN WE ALREADY CROSSED

In 1492, Columbus crossed the Atlantic with three ships and a bad map. What followed was more than an exploration. It was the largest expansion of accessible resources in human history. New land, new metals, new crops, new trade routes. The flow of New World gold and silver into Europe is estimated to have moved hundreds of billions of dollars in today’s money, and it restructured the entire global economy.

The map is about to get bigger than any explorer in history could have imagined. And this time, there’s no edge to fall off.

Now here’s the part most people miss. The New World wasn’t more valuable than Europe because of any single mine. It was more valuable because it multiplied the surface area of human possibility. The map got bigger.

Space does the same thing, except the multiplier isn’t 2x or 10x, its one-million-fold bigger... it’s effectively unbounded.

METALS: A SINGLE ASTEROID WORTH MORE THAN THE GLOBAL ECONOMY

Start with the headline number. As mentioned above, a single asteroid named 16 Psyche, a metal-rich body about 140 miles across, has been valued at roughly $10 quintillion in metals. That’s a 10 followed by 18 zeros. For scale, the entire global economy is somewhere around $100 trillion. Psyche is on the order of 100,000 times larger than everything humanity produces in a year. Clearly such a quantity of metals would crater the metals market, but it would also increase the use of these materials (Jevon’s paradox).

Sources: NASA / Newsweek (16 Psyche ~$10 quintillion); Goldman Sachs (single-asteroid platinum ~$50B); World Bank (global GDP ~$100T); World Gold Council (above-ground gold stock). Log scale.

Psyche is a nickel-iron asteroid, the exposed core of what may have been a baby planet. The reason it matters isn’t the iron. It’s what rides along with it: platinum-group metals.

Platinum today trades north of $1,800 an ounce. These metals are the catalytic backbone of hydrogen fuel cells, electronics, and clean energy hardware, and on Earth they’re brutally rare. On a metallic asteroid, they can be concentrated at levels that would make any terrestrial mine look like a sandbox.

FUEL: THE OIL OF DEEP SPACE

What’s more valuable than platinum-group metals (that are primarily used on Earth)? Rocket fuel, derived in space, for use in space, specifically water-ice mined from Carbonaceous chondrite asteroids. Run an electric current through that water and it splits into hydrogen and oxygen. Hydrogen and oxygen are rocket propellant. They’re also breathable air and drinkable water.

So, why does that matter so much? Because the single most expensive thing in spaceflight is hauling fuel up out of Earth’s gravity well. Every gram of rocket fuel used in space since 1957 has been lifted out of Earth’s gravity well. Lifting a kilogram of anything to orbit has historically cost on the order of tens of thousands of dollars. If you can refuel in space, from water mined off an asteroid, you’ve just removed the biggest tax on every mission to the Moon, Mars, and beyond.

Water is the oil of the deep-space economy. The first company to set up Earth and Moon orbiting refueling stations will own the most valuable service facilities humans have ever staked. Not because the water is rare. Because it sits at the chokepoint of everything else.

ORBITAL ENERGY & DYSON SWARMS

Early in my space-career (way back in the 1980’s), when I was envisioning future value creation in orbit, I imagined tourism or in-space manufacturing. I never imagined space-based AI satellites (humanity’s Dyson Swarm).

As Eric Schmidt has noted, the limiting factor for AI on Earth today isn’t chips or talent… it’s power. The grid is maxing out. In orbit, that constraint simply disappears.

Back in 1964, the Russian astrophysicist Nikolai Kardashev proposed ranking cosmic civilizations not by their morality or technology but by the energy they can harness. A Type I civilization captures all the energy available on its home planet, roughly 10^16 watts (for Earth). A Type II commands the full output of its star, on the order of 10^26 watts (for Sol), by wrapping a collection shell around the star, a Dyson sphere. That’s not a 10x jump over a planet. It’s a ten-billion-fold jump. The sun pours out more energy in one second than humanity has used in its entire history.

And what will we eventually do with 10^26 watts of energy? Power increasingly advanced AI systems.

COMPUTE: WHEN THE DATA CENTER MOVES TO SPACE

Here’s where the energy story turns into the most valuable industry of the century. If sunlight is free and unlimited in orbit, then the thing you most want to put up there is computation. AI compute is basically electricity converted into intelligence. Move the power source to space and you move the data center with it.

In the near-term, we’re not build a sphere around the Sun, but we are likely to launch hundreds of thousands of AI-data-center satellites into sun-synchronous orbits around the Earth.

Elon has laid out the math in two stages, and the numbers are worth sitting with.

Elon’s near-term prediction (by 2030): A few hundred gigawatts per year of AI compute launched into space. Starship, he says, could deliver around 100 gigawatts per year to high Earth orbit within four to five years, scaling launch mass from about 2,500 tons a year today, to a million tons a year within by 2030. For scale, a single gigawatt is about the output of a large nuclear reactor, and the entire U.S. has on the order of a few hundred gigawatts of AI-capable power today. He’s talking about adding that much new compute capacity every year, from orbit, with no grid to fight and no cooling bill.

Elon’s longer-term prediction (2040’s): Where next? Moving from 100 GW to 100 terawatt scale, built on lunar mining. In Elon’s words, “100 TW per year is possible from a lunar base producing solar-powered AI satellites locally, and accelerating them to escape velocity with a mass driver.” A terawatt is a thousand gigawatts. A hundred terawatts per year is on the order of the entire current power output of human civilization, added annually, in space.

Sources: Elon Musk (Dwarkesh Podcast, Feb 2026; SpaceX statements 2026) for ~100 GW/yr Starship-to-orbit, few-hundred-GW/yr near-term, and 100 TW/yr lunar-built long-term; US AI-capable power ~50 GW order-of-magnitude. Log scale.

Jeff Bezos said it plainly: “It’s 10 plus years, but I bet it’s not more than 20 years. We’re going to start building these giant gigawatt data centers in space. We will be able to beat the cost of terrestrial data centers in space in the next couple of decades.” Eric Schmidt bought the launch company Relativity Space to chase the same vision. Starcloud already flew the first NVIDIA GPU to orbit as a proof of concept, and Crusoe plans to stand up the first public cloud in space. When three of the wealthiest technologists on Earth independently bet on the same “impossible” idea, pay attention.

The sun pours out more energy in a single second than humanity has used in its entire history. We’re finally building the bucket.

REAL ESTATE: A FRONTIER WITH NO SHORELINE

When European settlers reached the Americas, the most enduring wealth turned out not to be the gold they shipped home. It was the land they settled (and/or took from the native populations). Every acre eventually became a farm, a city, a market.

Now apply that to space. The asteroid belt alone contains over a million asteroids larger than a kilometer. The Moon has 14.6 million square miles of surface, almost the area of Asia, sitting three days away. There is no Atlantic to cross, no edge of the map, no fixed quantity of “land” to fight over. The frontier keeps expanding the moment you reach it.

Jeff Bezos takes this to its logical conclusion. With the resources and solar power of the solar system, he argues, “We could have a trillion people out in the solar system,” and “if we had a trillion humans, we would have a thousand Einsteins and a thousand Mozarts, and unlimited, for all practical purposes, resources and solar power.” That’s the part that moves me most. Abundance doesn’t only mean more metal and more energy. It’s also more intelligence, both human minds and AIs. More intelligence that is free to invent, create, and solve. The ceiling on genius has always been the size of the human population and the technical resources to support it. Space lifts both. Scarcity is a story we tell because Earth has edges. The solar system doesn’t.

“When you have an Abundance Mindset, rather than slicing the pie into thinner and thinner slices, you bake more pies.”

— Peter

This is the part that disrupts people’s intuitions. Every economic model we’ve ever built assumes finite resources and zero-sum competition. That assumption was always temporary. It was true only for as long as our reach ended at the atmosphere.

THE MOON VERSUS THE ASTEROIDS

So where do we mine first? This is the real strategic debate, and there are good arguments on both sides.

The Moon is close. Three days away, with water ice locked in its permanently shadowed polar craters and helium-3 scattered across its regolith. Proximity is the Moon’s superpower. Lower launch costs, faster round trips, a natural staging base for everything deeper. The case against it: extraction is hard in one-sixth gravity, and the richest dreams (large-scale helium-3 for fusion) depend on fusion reactors we don’t yet have.

The asteroids are richer and more varied, unincumbered by a significant gravity well. Metallic asteroids concentrate platinum and nickel-iron at grades no planet’s crust can match, and carbonaceous ones hand you fuel and water. The case against: they’re far, they’re scattered, and a mission is a multi-year commitment.

My honest take, after fifteen years of thinking about this: it’s actually not either-or. The Moon is the training ground and the gas station. The asteroids are the motherlode. We learn to mine on the Moon because it’s close, then we apply what we learn to the asteroid belt because that’s where the real wealth lives.

Even some longtime lunar advocates now admit the return on asteroids may beat the return on the Moon.

THE LUNAR ECONOMY: OUR FIRST OFF-WORLD ADDRESS

Everything I’ve described needs a beachhead, and the Moon is it. Three days away, it’s close enough to be our practice field, our fuel depot, and our first factory floor off Earth. This is no longer a thought experiment. NASA’s Artemis program, China’s lunar plans, and a swarm of private landers are all converging on the same gray dirt this decade.

The bankers have started doing the math. PwC projects a lunar economy worth roughly $170 billion a year by 2040, built on three pillars: government exploration contracts, lunar surface operations, and the mobility and logistics to move between them. That’s before the truly large prize, which is using lunar water ice for propellant and lunar metal for manufacturing, so we stop shipping everything up from Earth.

Sources: PwC Lunar Market Assessment (~$170B by 2040); SpaceNexus cislunar segment estimates; 2050 figure reflects illustrative continued build-out. Log scale.

Here’s the strategic insight that gets overlooked. The Moon’s real value isn’t what we ship back to Earth. It’s what we never have to ship up from Earth again. Once you can make rocket fuel, structural metal, and solar panels on the lunar surface, the Moon becomes the assembly line for everything deeper in the solar system.

Remember Elon’s terawatt vision: solar-powered AI satellites manufactured on the Moon and flung into deep space by an electromagnetic mass driver, no rockets required. The lunar economy and the orbital-compute economy are the same story told from two ends.

WHAT THIS MEANS FOR YOU

If you’re an entrepreneur: The picks-and-shovels opportunities here are wide open. Refueling depots, extraction robotics, ISRU processing, orbital logistics. You don’t have to own an asteroid to get rich serving the people who do.

If you’re an executive: Start tracking the cost-per-kilogram-to-orbit curve the way you track Moore’s Law. When it crosses certain thresholds, entire industries (metals, energy, manufacturing) get a new and effectively unlimited supply chain.

If you’re an investor: There will be many first trillion-dollar fortunes made in space. Fuel and logistics before glamour.

If you’re a student: This is the field of your lifetime. Planetary science, robotics, materials, space law. The people who write the rules and build the tools for the off-world economy are in school right now.

If you’re a parent: Teach your kids that the future is not a fixed pie to be divided. It’s an expanding frontier to be built. The mindset of abundance is the most valuable thing you can hand them.

FIFTY YEARS, ONE MAP

Put it all on a single timeline and the shape of the century comes into focus. Satellites and launch are the mature core today. Space tourism is the luxury edge already selling seats. Orbital AI data centers begin this decade and become the largest line on the chart. Lunar infrastructure matures through the 2030s and 40s. Asteroid mining, the business I bet on too early, finally arrives once the fuel depots and lunar factories make it cheap to reach the belt.

Illustrative build-out. Near-term anchors: McKinsey/WEF total space economy ~$1.8T by 2035 (from $630B in 2023); Morgan Stanley >$1T by 2040; PwC lunar ~$170B by 2040; space mining ~$40B and space tourism ~$10–90B by 2035. Figures beyond ~2040 are directional extrapolations, not forecasts.

McKinsey and the World Economic Forum already peg the total space economy at $1.8 trillion by 2035, nearly triple its 2023 size. And that estimate barely accounts for the orbital-compute and asteroid-mining curves that bend skyward in the decades after. The trillion-dollar sky isn’t one industry. It’s a stack of them, each one unlocking the next.

On June 12, 2026, SpaceX went public in the largest IPO in history and Elon became the world’s first trillionaire. In his remarks, he repeated the line he’s been saying for 24 years: the goal is to “take the fiction out of science fiction.” That’s exactly what’s happening to everything in this newsletter. My early predictions came true… The first trillionaire was made in space. He just won’t be the last.

“The best way to predict the future is to create it yourself.”

— Peter

The only question that matters is whether you still believe scarcity is permanent. Look up, and ask yourself how much of that sky you’re willing to call impossible.

To a future of Abundance,

Peter

More From Peter

If you’ve enjoyed Metatrends, here are more ways to stay connected:

TLDR: Radical transparency is coming. A future where you can know anything, anytime, anywhere. A future where no one can hide. We are wrapping the planet in an "Sensor Ecosystem": a living, multi-layered sensing system that runs from the cameras in your home, to the phone in your pocket, to autonomous cars and humanoid robots on the ground, to drones and flying cars in the air, all the way up to a constellation of satellites imaging every square meter on the Earth every single day. By 2030, there will be roughly 40 billion connected devices, each packing multiple sensors, generating a flood of data that AI can finally read. In Part 1 of this newsletter I’ll show you the scale of what’s being deployed, layer by layer. In Part 2, I’ll share what it means, because when everything becomes visible, it gets very hard for anyone to hide, and that changes how humans behave.

PART-1: THE TRILLION-SENSOR ECOSYSTEM

Earlier this week, I hosted a conversation with Will Marshall, CEO of Planet, on my Moonshots podcast. Planet operates the largest fleet of Earth-observing satellites in human history and Will walked me through what’s coming. We are constructing a planet-scale sensing organism, and most people have no idea how big it’s already gotten.

OUR ELECTRIC SKIN, LAYER BY LAYER

Layer 1 – Space: Planetary Intelligence from Orbit

There are roughly 13,800 active satellites in low Earth orbit today. Analysts at consulting firm Analysys Mason predict that this number will climb to somewhere between 27,000 and 100,000 by 2030. SpaceX’s Starlink alone has about 10,400 operational satellites today, on the way to a licensed 42,000.

But the satellites that matter most for transparency are the ones that image us. Planet runs about 200 of them, collecting 25 terabytes of imagery a day and a 150-petabyte archive, roughly 3,000 images of every point on Earth’s land surface over the past decade.

And it’s getting sharper. Planet’s next-generation Pelican satellites, launching in late 2026, will hit 30-centimeter resolution with a 32-satellite constellation revisiting any point on Earth up to 10 times a day, delivering imagery in 30 minutes.

Layer 2 – The Air: Drones and Flying Cars

Drop down from orbit into the atmosphere and you hit the aerial layer. The FAA already has more than a million drones registered in the United States. Globally, the commercial drone market is headed past $50 billion by 2030, with Asia-Pacific alone projected to field around 9 million commercial units.

Delivery is where it gets wild. The number of package-delivery drones (each with a suite of cameras and sensors) is forecast to jump from about 32,000 in 2024 to over 275,000 by 2030 (MarketsandMarkets). Inspection and survey drones routinely resolve the ground at 1 to 3 centimeters, sharper than any satellite. And above them, the eVTOL “flying car” layer is arriving: Aviation Week projects roughly 2,000 eVTOLs delivered by 2030, scaling into a market Morgan Stanley once pegged at over $1 trillion by 2040. Every one of these aircraft carries lidar, radar, and electro-optical cameras to navigate. They don’t just fly, they see.

Layer 3 – The Road: Autonomous Vehicles

Now we’re at street level. Waymo runs about 3,000 robotaxis today, delivering 500,000 paid rides a week. Goldman Sachs projects roughly 35,000 robotaxis on US roads by 2030, with global autonomous vehicle counts running into the tens of millions.

Here’s what stops me cold. A single Waymo vehicle carries 13 cameras, 4 lidar units, and 6 radar, with 17-megapixel imagers that out-resolve almost anything in consumer hardware. One autonomous car generates on the order of 4 terabytes of data per hour of driving. Recent engineering analysis puts the lidar stream alone at up to 8 petabytes per year, per vehicle. Multiply that by tens of thousands of cars, each mapping every street, every pedestrian, every pothole, in real time. We are accidentally building the most detailed, continuously updated map of the physical world that has ever existed. And down on the sidewalk, ground-delivery robots (e.g., Starship already runs about 2,700, heading toward 12,000) are filling in the gaps the cars miss.

Layer 4 – The Walking Layer: Humanoid Robots

This is the newest layer, and it may become the densest. Humanoid robots are about to walk through our factories, our streets and our homes. And every one of them is a mobile sensor platform with eyes.

I’m going to use the builders’ own numbers here, not an analyst’s. Elon has said he’s confident Tesla can reach 1 million Optimus units per year by around 2030. Figure has built BotQ, a factory designed to produce up to 12,000 humanoids a year on its first-generation line, roughly one robot per hour. Both Elon and Brett Adcock’s long-term vision predict >10 billion humanoids globally.

Each of these machines is vision-first: multiple onboard cameras, inertial sensors, force and torque sensors in every joint, tactile sensors in the hands, all capturing continuously. A humanoid working an 8-hour shift is plausibly generating data on the scale of an autonomous car. Now picture millions of them, then tens of millions, walking through the world and recording it.

Layer 5 – The Pocket: 7 Billion Smartphones

The layer you’re holding right now. There are about 7 billion smartphone subscriptions on Earth, and the average phone carries somewhere between 14 and 20 sensors: multiple cameras, GPS, accelerometer, gyroscope, magnetometer, barometer, microphones, proximity and light sensors. This is, by a wide margin, the largest sensor network humanity has ever deployed, and we did it by accident, one upgrade at a time. Every person becomes a roving, multi-spectral data node.

Layer 6 – The Ground: Home Cameras & Industrial IoT

Finally, the ambient layer woven through our buildings, streets, and homes. There are already over 1 billion surveillance cameras in use worldwide. Add the Ring doorbells, the Nest cams, the Arlo and Wyze units multiplying on every porch. Then add the industrial layer: the sensors on pipelines, factory floors, shipping containers, and power grids.

Tie it all together and you get the number that frames everything: total connected IoT devices stand at about 21 billion today and are projected to reach roughly 40 billion by 2030, each one carrying multiple sensors. That’s where the “trillion sensor” world stops being a metaphor.

Sources: Planet Labs; FAA; MarketsandMarkets; Goldman Sachs; Waymo; Tesla; Figure AI; Ericsson; IoT Analytics; Video Experts Group (2024–2026)

THE EXPONENTIAL, IN ONE NUMBER

Stack the layers and the trajectory is unmistakable. As of 2025 we had about 21 billion. By 2030, around 40 billion. By 2040, as humanoids, autonomous fleets, and orbital constellations compound on top of everything else, we move from billions of devices into the realm of trillions of individual sensors, each one streaming a slice of reality into AI systems that can finally make sense of all of it.

Source: IoT Analytics (2025); The Future Is Faster Than You Think (Stanford/Accenture)

That’s the engine. Now here’s why it changes everything about human behavior.

PART-2: WHAT HAPPENS WHEN NO ONE CAN HIDE

During my conversation with Will Marshall, he began to layout the consequences: “No one can hide anymore,” he said. “If you build a school, we’re going to see the school. If you build a data center, we’re going to see the data center. And the accountability is going to be there for the whole world to see, no matter what.”

And the human behavioral consequences are profound. When people know they’re being watched, they behave different, they behave better. And we now have hard data proving it.

WHEN PEOPLE KNOW THEY’RE WATCHED

The cleanest evidence comes from one of the best natural experiments in social science: police body cameras.

In 2012, the police department in Rialto, California ran a randomized controlled trial with Cambridge criminologists. Over the course of a year, officer shifts were randomly assigned “camera on” or “camera off,” collecting more than 50,000 hours of police-public interactions. The result was use-of-force incidents dropped 60%, and citizen complaints against officers fell 88%, from dozens the prior year down to a grand total of three.

Source: Rialto, CA randomized controlled trial — Ariel, Farrar & Sutherland, Cambridge (2015)

Researchers call it the “civilizing effect.” When both the officer and the citizen know the encounter is on the record, both sides regulate themselves. Nobody wants to be the person on the tape. Cambridge has since replicated the work across ten more forces, and the core finding largely holds.

“When both sides know they’re on the record, both sides behave better. Sunlight is becoming infrastructure.”

And it’s not just cops. The definitive 40-year meta-analysis of closed-circuit television, led by Eric Piza with Brandon Welsh and David Farrington, found CCTV associated with about a 13% reduction in crime overall, and a much steeper drop, over a third, in settings like car parks.

Source: Piza, Welsh, Farrington & Thomas — 40-year meta-analysis, Criminology & Public Policy (2019)

The pattern runs all the way up to governments. A 2023 review of 114 studies in Public Administration and Development found that transparency and disclosure measurably reduce corruption, especially when citizens can act on what they learn. Alina Mungiu-Pippidi has shown that in countries with weak rule of law, financial disclosure and freedom-of-information access work as a substitute for honest institutions. Citizens become the enforcement mechanism when the courts won’t be.

TRANSPARENCY AS A DETERRENT TO WAR

Will Marshall next argued that transparency doesn’t only civilize individuals. It actually can prevent wars. “Throughout history,” he said, “wars happened mainly when there’s been misinformation, a lack of information, and people had to guess or made mistakes. Transparency drives accountability and reduces the probability of war.”

He gave a concrete example. In the run-up to the invasion of Ukraine, Vladimir Putin massed troops on the border believing the buildup would go unnoticed. Planet’s imagery put it on the front page of every newspaper on Earth. “Putin thought he could get away with it,” Will said. “We put that to bed.”

It didn’t stop the invasion. But it changed the information environment around it. For most of human history, the fog of war was literal: leaders started wars because they couldn’t see what the other side was doing, and they guessed wrong. We’re now building a world where everyone can see the troops massing and the treaties breaking, in near real time. You can monitor a peace accord from orbit. You can’t claim the bridge is standing when 200 satellites photographed it fall.

Governments own the map.

But Planet owns the sky above the map.

THE OTHER EDGE OF THE BLADE

I’d be doing you a disservice if I sold you only the utopian version. An electric skin that registers everything is also an electric skin that registers you and can dissolve privacy.

The hard truth is that transparency is a tool, and tools don’t have ethics. A surveillance state and an accountable democracy can run on the identical sensor network. What separates them is who can see, who is seen, and whether the watching runs in both directions. The world I worry about isn’t the one where everyone is visible. It’s the one where the powerful can see everyone, while no one can see them. Transparency only builds trust when it points both ways.

WHAT THIS MEANS FOR YOU

If you’re an entrepreneur: Assume your operations, supply chain, and claims are about to become independently verifiable by anyone with an API key. Build the company you’d be proud to have photographed from orbit or an overhead drone. The brands that win the transparency era are the ones that were already telling the truth.

If you’re an executive: Radical transparency is a moat, not a threat… if you move first. Open your data, your sourcing, your numbers, before a satellite or a sensor does it for you. Disclosure you volunteer builds more trust than disclosure that’s extracted from you.

If you’re an investor: A new asset class is forming around verifiable ground truth: counting cars in lots, ships in ports, construction in progress. Alpha is shifting to whoever can see the physical world in real time. Planet (PL) is one way to play it, and there will be many more.

If you’re a student: The most valuable skill of the next decade is turning oceans of sensor data into honest answers. Learn to work where AI meets the physical world. For the first time, the library is being connected to the window.

If you’re a parent: Your kids will grow up in a world with no “off the record.” Teach them that the best privacy strategy is integrity, living so that being seen costs you nothing. And fight, hard, for a world where the watching goes both ways.

Will Marshall is building the crystal ball. The sensors are going up by the billions, and they are not coming down. So here’s the question I’ll leave you with, the one I’ve been chewing on since we turned the mics off: in a world where no one can hide, do we become our better selves because we choose to, or only because we’re being watched?

To a future of Abundance,

Peter

P.S. How much is an extra decade of healthy life worth to you?

The Abundance Longevity Trip (October 7–11, San Francisco) is five days of access you can’t get anywhere else: private visits to leading research labs, direct conversations with the scientists and biotech founders shaping the longevity revolution, and personalized diagnostics and therapeutics most people don’t know exist.

I personally host and guide 80 Members each year. Investors come for the pre-market deal flow. Researchers and practitioners come for the latest science. Others come for personal health breakthroughs. Everyone leaves with relationships and insights that change how they think about their future.

Secure one of the final 3 seats.

Peter

More From Peter

If you’ve enjoyed Metatrends, here are more ways to stay connected: