• Dow gained 0.56%

• S&P 500 rose 0.46%

• Nasdaq added 0.45%

• Russell 2000 was the standout, jumping 1.12%

The important part was understand what happened between Tuesday and Wednesday. Oil remained elevated, but stopped …

Read more

The S&P 500 fell 0.25%. The Nasdaq dropped 0.52%, while the Dow was essentially flat, down 0.02%. Small caps took the biggest hit, with IWM falling roughly 1.4%.

This was not a broad market liquidation. The Dow held up, several large cap names rallied and five S&P sec…

Read more

Harmonic moved from being primarily a video infrastructure company toward software based video delivery and then toward virtualized broadband acce…

Read more

SPY gained about 0.4%, QQQ about 0.4%, and IWM about 0.9%. The important part was the breadth of the bounce. Small caps and the Dow outperform…

Read more

• S&P 500 fell 0.52% to 7,745.06

• Dow dropped 0.51% to 53,459.78

• Nasdaq slipped 0.31% to 26,644.91

• Russell fell roughly 0.4%

The market is still sitting close to record highs. What chan…

Read more

Small caps were the exception. The Russell gained about 0.55% and extended its recent strength.

This week, the market received a “surprisingly” weak retai…

Read more

Not because the economic news was good.

Because it wasn’t.

July’s employment report came in far weaker than expected, immediately changing the conversation from “Will the Fed hike again?” to “The Fed probably has room to wait.”

That shift was enough to send treasury yields sharply lower, ignite another rally in techno…

Read more

After spending nearly two months chopping sideways while investors debated oil, yields, inflation and the sustainability of the AI trade, the market delivered one of its strongest sessions of the year. The S&P 500 and Dow closed at fresh record highs, while the Nasdaq posted an even stronger gain as technology reclaimed leadership. …

Read more

The Dow saw a decent bounce, the S&P 500 finished essentially unchanged, while the Nasdaq continued to lag as semi’s remain…

Read more

Build bigger models.
Buy more GPUs.
Spend more on data centers.
Increase compute.

That narrative created some of the biggest winners this cycle. Semiconductors, networking, hyperscalers, power infrastructure and anything tied to AI compute became the center of attention.

Lately, something feels different. The conversation is beginning to shift. Not because AI has become less important. In fact, quite the opposite. AI is becoming more accessible, more deployable and increasingly integrated into everyday business operations. And this is changing where value is created.

The market may have given us an early glimpse of this transition. While several neocloud names came under pressure, enterprise software quietly held up. One day doesn't establish a new trend, but it is worth paying attention to if this quietly starts to strengthen under the table.

Keep in mind that many of these names are still severely beaten down and continue to trade below their major moving averages. That doesn't automatically mean the trend has changed. What it does mean is they're starting to show relative strength (still super early) while the market begins to question where the next wave of AI value will be created. If that relative strength continues, it's worth paying attention to.

At first glance, those two moves appear contradictory. I don’t think they are. I think they’re connected. The AI story is evolving from building intelligence to distributing it and that is important distinction.

The first phase of AI rewarded companies capable of creating the most powerful models. The challenge was proving that artificial intelligence could reason, write, code, analyze data and solve real business problems.

That hurdle has largely been cleared. The race to build the best model isn't over, but we're no longer living in a world where only one or two companies have capable AI. Today there are multiple highly capable models available across the market. Some remain proprietary. Others are becoming increasingly open. Performance differences still matter, but they matter far less than they did just a year ago. As capable models become more widely available, the question investors should be asking changes.

The more I thought about it, the more I realize the discussion isn’t really about open-weight AI. It’s about abundance. If capable models become widely available, intelligence itself becomes less scarce. Whenever scarcity disappears, markets begin looking for the next scarce resource. That’s where capital usually migrates.

History suggests that is when leadership begins to rotate. Every major technology cycle follows a similar path. The internet eventually stopped being about websites and became about search, e-commerce and digital advertising. Cloud computing stopped being about virtualization and became about enterprise software delivered through the cloud. Smartphones stopped being about the hardware itself and became about the application ecosystem built on top of it.

Now, artificial intelligence appears to be approaching a similar transition. The model increasingly becomes the foundation rather than the product and once the foundation is available to everyone, competitive advantages begin shifting elsewhere. That is where the market should be paying attention. Many investors continue viewing AI through the lens of training.

Training has dominated headlines because of the extraordinary capital required to build frontier models. Hundreds of thousands of GPUs, massive data centers and record Capex. Training, however, is only one part of the equation. I believe the next phase of the AI race will be measured less by who trains the largest model and more by who can deploy AI at scale.

Training happens once. Inference happens every single time AI is used:

• Every customer service interaction

• Every software assistant

• Every medical diagnosis

• Every legal document review

• Every financial analysis

• Every coding request

If AI becomes embedded across the global economy, inference becomes a recurring workload rather than a one-time investment and that has significant implications for where demand develops over the next decade.

Infrastructure does not disappear. It changes. Instead of a handful of companies building massive training clusters, thousands of businesses begin deploying AI inside their own organizations:

• Banks

• Healthcare providers

• Manufacturers

• Governments

• Insurance companies

• Law firms

Every deployment requires compute, networking, storage, power, cooling, cybersecurity, that demand doesn’t necessarily shrink, it becomes distributed. That distinction may also explain why yesterday’s AI names have been today’s leaders for sell offs.

The market isn’t necessarily questioning long-term AI demand. It may be questioning where the economics of that demand ultimately settle. Owning GPUs alone may not be enough. If access to AI infrastructure becomes increasingly available, competition naturally shifts toward utilization, pricing, financing, customer relationships and operating efficiency. Those are far different conversations than simply counting GPU deployments.

At the same time, enterprise software the same sector that has been left out to dry has been gaining ground in the dark pool lists, institutional lists and that may not be a coincidence. Software companies already possess the one asset many infrastructure providers do not.

Customers.

They already sit inside enterprise workflows. They understand industry-specific processes. They own years of proprietary operational data. They have recurring revenue. They have established distribution. Artificial intelligence simply makes those platforms more valuable.

Think about $MSFT, the long term opportunity is not necessarily owning the smartest model. It is making Microsoft 365 indispensable through AI. The same idea applies across enterprise software. $NOW automates workflows. $CRM improves customer engagement. $PLTR connects complex data across governments and enterprises.

These companies are not competing to invent intelligence. They are competing to embed intelligence into work that already exists. That may prove to be the more durable business model. One of the more interesting themes emerging across the industry is the growing comparison between open-weight AI and open-source software. And the argument is straightforward:

When foundational technology becomes widely available, innovation accelerates because developers, businesses and governments build on top of a common foundation.

We have seen this before.

Linux did not become the largest technology company in the world. The ecosystem built around Linux generated enormous value.

• Cloud providers

• Enterprise software

• Infrastructure

• Consulting

• Applications

The operating system became the foundation. The businesses built on top captured much of the economics.

AI could be following a similar path. If capable models become increasingly abundant, owning the model becomes less important than owning the customer relationship.

That shifts attention toward:

• software

• workflow automation

• enterprise integration

• proprietary data

• distribution

• switching costs

These are characteristics software companies have spent decades building.

The market may be beginning to distinguish between providing compute and capturing value. That doesn’t mean infrastructure is no longer important. Far from it. Infrastructure remains the backbone of AI. But as every technological cycle matures, value rarely remains concentrated in the foundation, it expands into the ecosystem.

For investors, this may be one of the more important shifts to monitor over the coming quarters. The question may no longer be which company builds the smartest model.

The better question may be where the next economic bottleneck develops. If the first chapter of AI was creating intelligence, the next chapter may be integrating that intelligence into every workflow, every enterprise and every industry.

Markets have a habit of rewarding the companies solving tomorrow’s bottleneck rather than yesterday’s. That is the transition I believe is beginning to unfold remains my focus point in investment.

Ask yourself:

If every Fortune 500 company suddenly had access to a world-class AI model tomorrow, what would stop them from deploying it?

Wall Street entered one of the biggest earnings days of the quarter with and scheduled to report after the closing bell. That kept investors from making aggressive bets in either direction. Rising oil prices, higher Treasury yields and renewed geo…

Read more

CareCloud is a small cap healthcare technology company that provides cloud software, AI tools and outsourced business services for physicians, medical groups, and hospitals. Rather than selling software alone, it combines SaaS products with revenue cycle management (RCM), medical billing, coding, credentialing, staffing and AI-powered workflow …

Read more

This wasn’t a broad risk …

Read more

Oil climbed again as fighting between the US and Iran intensified. Earnings season kicked off with a flood of bank results. Investors also had one of the most important inflation reports of the summer to digest.

Instead, stocks finished higher.

The reason: Inflation cooled much more than expected, giving investors conf…

Read more

Read more

Wall Street bounced back today as geopolitical fears cooled, oil reversed lower and buyers stepped back into growth. The market shifted from yesterday’s defensive positioning back toward technology, AI, and higher beta areas.

Major Index Performance

SPY / S&P 500 +0.8%
Closed: 7,543.64

The S&P recovered…

Read more

Read more

After one of the strongest second quarters in years, investors spent much of Wednesday locking in gains, particularly across technology and semiconductor stocks. Which showed us how Tuesday’s action was something as a trap. However, the selling wasn’t aggressive enough to suggest panic, but it was …

Read more

Stocks rallied throughout most of the session as investors closed the books on an impressive quarter. Technology bounced back after weeks of profit taking, semiconductors found buyers again and qua…

Read more

After 5 straight sessions of selling, buyers stepped back in and produced a broad rebound across U.S. equities. The biggest driver wasn’t a surprise economic report or a Fed headline. Instead, it was a combination of easing geopolitical tensions, quarter end positioning and investors buying back some of last week’s h…

Read more