OpenAI’s stumble is not proof AI demand is weakening. It is proof competition is heating up.
KEY TAKEAWAYS
Competition matters because it protects customers from monopoly pricing and forces better products. A gas station example makes the AI point simple: more competitors usually means better outcomes.
Markets overreacted to reports that OpenAI missed internal targets. AI-linked stocks sold off because investors treated an OpenAI-specific issue as an industrywide demand problem.
Anthropic’s rise changes the interpretation of the story. OpenAI may be losing enterprise share, but that does not mean enterprise AI demand is weakening.
Competition at the frontier model level may accelerate AI infrastructure spending. Companies under pressure do not usually cut compute budgets; they fight harder.
Microsoft, Alphabet, Amazon, and Meta earnings are the real test. The key issue is whether AI capex is producing monetization, margin improvement, and durable enterprise workloads.
MY HOT TAKES
The OpenAI story is being mispriced by the market. The issue is competitive pressure, not a collapse in AI demand.
Anthropic gaining share is bullish for the broader AI ecosystem. It proves the market is real enough to support more than one major winner.
The AI trade remains fragile because everyone wants to hate it the moment a negative headline appears. That creates reflexive selloffs that may not match the underlying economics.
Hyperscaler earnings matter more than OpenAI drama. Microsoft, Alphabet, Amazon, and Meta will show whether AI spending is turning into real business performance.
Competition is not a threat to the AI thesis. Competition is what makes the thesis stronger, bigger, and more durable.
You can quote me: “Wall Street sold good news dressed up as bad news like it was an obituary.”
Gloves off! Near my weekend home in the wilds of the New York City suburbs we have only one gas station in the middle of our town. It is convenient and central to all of the major highway entrances. It is almost always open. And…it is about $1.00 more per gallon than the national average! Wow! I am sure that doesn’t surprise you given the recent spike in crude and supply shocks. But those things are not the reason why my local station charges more. Nope. It is because there is literally no competition. Once you get on some of the major roadways heading into the city, stations dot the roadside like weeds. Their prices are significantly lower and they are all more or less the same.
I opened with this anecdote not because I am going to write about energy costs. No. I opened with this to underscore the importance of competition. In most cases, competition amongst sellers is good for consumers. Competition ensures that you are being charged a fair, market-based price, and in the case of highly specialized products, you are going to get the most competitive product for your buck!
Markets have largely shrugged off the Iran war and crude supply shock in recent weeks, choosing instead to focus on last year's belle of the ball AI, which was left on the side of the road in first the great value rotation, and then the war in Iran. With those themes playing out, a fresh look at AI, specifically semiconductor companies (the bricks and picks), investors noticed that those companies with great growth prospects had become cheaper in recent months. That helped drive recent gains as investors appeared to blindly binge buy everything that was popular in Q3 of last year. Everyone was positioning for Q1 earnings which are going to start for AI in earnest tonight! Smooth sailing…until yesterday when the Wall Street Journal threw a spanner into the works.
The Journal dropped a report–timed, as my not-doppleganger Jim Cramer rather colorfully put it, with the "exquisite precision of an evergreen hit job"--claiming that OpenAI has missed multiple internal revenue targets in early 2026, failed to reach one billion weekly ChatGPT users, and that CFO Sarah Friar has been raising alarm bells internally about whether the company can sustain its massive data center spending commitments. Go ahead, read that again–it’s important. 👈 The market heard "OpenAI is struggling" and proceeded to sell everything that wasn't nailed down. Oracle cratered nearly 8%. CoreWeave dropped over 7%. SoftBank shed close to 10% overnight in Tokyo. Even Nvidia, the Teflon king of the AI trade, got clipped. Panic first, questions later–classic Wall Street. Classic AI, which I told my favorite AI Journalist yesterday is the “trade that everyone loves to hate!”
Now, pay attention. The market completely misread what this story actually says. This is not an AI demand story. This is a competition story, and there is an enormous difference between the two.
Let me explain. OpenAI didn't lose revenue because enterprises stopped buying AI. It lost revenue because enterprises started buying a different AI–specifically, Anthropic's Claude. At the start of 2025, Anthropic had roughly $1 billion in annualized revenue. By April of this year, that number had crossed $30 billion (it’s ok, you can say it: holy 💩), surpassing OpenAI's own reported figure of around $25 billion. Think about that growth trajectory for a moment. OpenAI, to its credit, grew from roughly $20 billion to $25 billion in annualized revenue over that same fifteen month period, which is a respectable 25% increase. But Anthropic's growth rate was approximately forty times faster. In the enterprise coding market specifically, Claude now commands more than half of the market versus OpenAI's roughly 20%. Developers migrated–not because of Anthropic's marketing, but because the product performed better on the workflows that matter most to them.
So when the Journal reports that OpenAI missed its internal targets, what they are really describing is a company getting outmaneuvered in the enterprise segment by a faster-moving competitor. That is a very different animal from "AI demand is softening." The Wall Street assumption embedded in yesterday's selloff was that if OpenAI is struggling, the entire infrastructure build-out must be in jeopardy. That assumption is, to put it diplomatically, completely backwards. A company that is losing enterprise market share to a rival does not respond by trimming its compute budget. It doubles down! OpenAI has the capital, the partnerships, and frankly the institutional ego to do exactly that. The spending cycle is not breaking. If anything, the competitive pressure from Anthropic is going to accelerate it.
The smarter question that nobody in yesterday's panic was asking is this: what does it mean for the broader AI ecosystem that we now have genuine, ferocious competition at the frontier model level? I would argue it means exactly what competition always means–better products, more aggressive pricing, faster innovation, and a larger total market. Consider that roughly 80% of OpenAI's paying enterprise customers also pay for Anthropic. These are not replacement relationships. They are additive ones. The pie is getting bigger! Guys, PLEASE, if you take away anything from today’s note, please make it this! I have been working on the leading edge of tech innovation for my entire career–I have seen this before. The gas stations are multiplying along the highway, and prices are getting more competitive–the baked goods in their convenience store are getting fresher (though fresh may not be the right word choice here). That is good news dressed up as bad news, and Wall Street sold it like it was an obituary.
Now. With that context firmly in hand, let's talk about what actually matters today, because tonight is the real verdict. After the close, Microsoft, Alphabet, Amazon, and Meta report earnings. These four companies have collectively committed somewhere between $600 and $645 billion in AI capital expenditure for 2026 alone. That is not a typo. Tonight they have to stand up and show us what that spending is producing. In light of yesterday's OpenAI noise, the stakes could not be higher–or more clarifying.
The question I want answered is not whether any of these four companies beat the Street by two cents. I genuinely do not care about that. What I want to know is whether the monetization engine is actually firing. Is Azure's AI attach rate translating into real margin expansion? Is Google Cloud's Gemini revenue contribution showing up in a way that justifies the infrastructure bet, especially now that Gemini has been explicitly named (by WSJ report) as a consumer-side beneficiary? Side note: Anthropic quietly ran away with the enterprise market OpenAI needed most. Are AWS margins recovering in a way that confirms enterprise AI workloads are durable and growing? And is Meta's Llama-powered ad platform actually making the advertising product meaningfully better, or is it just a good story for the investor deck?
These are the numbers that will either validate or complicate the entire AI trade heading into the second half of the year. Not the OpenAI internal memo. Not the CFO's concerns about compute contracts. Tonight's reports from four of the most consequential companies on earth–that's your signal.
Here's how I'll leave you. Yesterday's selloff was a reflex, not a verdict. The real verdict comes after the bell tonight. And if you take one thing away from today's blogpost/newsletter my friends, let it be this: competition in AI is not a threat to the investment thesis. It is the investment thesis working exactly as it should. Multiple strong players fighting hard for enterprise market share, spending aggressively, innovating relentlessly. That is the environment that separates the lasting platforms from the one-hit wonders. My monopoly gas station can charge whatever it wants. The moment a second station opens across the street, everybody wins except the guy who got comfortable being the only option. There is a lot for sale, if you are interested. 😉
YESTERDAY’S MARKETS
Tuesday's session turned lower as a Wall Street Journal report on OpenAI's missed revenue targets rattled the AI ecosystem, sending the S&P 500 down by 0.49% and the Nasdaq Composite down by 0.9%. The Dow Jones Industrial Average was essentially flat, declining just 25.86 points, cushioned by a nearly 4% gain in Coca-Cola following a strong earnings beat. AI infrastructure names bore the brunt of the selling–Oracle fell 7.7%, CoreWeave dropped 7.4%, and Nvidia shed 1.5%, while WTI crude settled up 3.0% at $99.93 per barrel on continued Hormuz supply disruption. The Russell 2000 declined by 1.26%, giving back a portion of its 11% April gain heading into tonight's pivotal Mag-7 earnings reports.
NEXT UP
Durable Goods Orders (March) is expected to have increased by +0.5% after declining by 1.3% in February.
Housing starts may have slipped by -0.4% while Building Permits probably fell by -0.3% for March.
At 2:00 Wall Street Time, the FOMC will announce its rate decision which is largely expected to reflect no change in the Fed Funds Rate. The 2:30 press conference is likely Jerome Powell’s last and will be the real catalyst to potential volatility in the session.
Important earnings today: Phillips 66, Yum! Brands, Etsy, Generac, GE Healthcare Tech, Regeneron, Biogen, Lennox International, ADP, Spirit Airlines, Vulcan Materials, Avis Budget, Humana, AMphenol, Wingstop, AbbVie, Ford, Mattel, Chipotle, VICI Properties, MGM Resorts, Carvana, O’Reilly Automotive, Microsoft, Alphabet, QUALCOMM, Amazon, Meta, eBay, and Viking Therapeutics.