For years, skeptics asked the same question: when does AI actually make money? This week, we got the answer.
The numbers that should silence every AI bubble argument
Anthropic, the San Francisco company behind Claude, is on track to post $10.9 billion in revenue for Q2 2026, more than doubling its first quarter haul of $4.8 billion. That single quarter would exceed Anthropic's total revenue for all of last year.
I think this is the most important financial story in tech right now, and most people are still sleeping on it. The conventional wisdom was that AI labs burn cash forever. Anthropic is about to prove that wrong, projecting $559 million in operating profit for the June quarter. OpenAI, by comparison, has told investors it does not expect profitability until 2030.
The growth rate is almost impossible to process. CEO Dario Amodei said at his company's developer conference in San Francisco that Anthropic planned for 10x growth but instead saw an 80x surge in revenue and usage in Q1. He called it "just crazy" and "too hard to handle." That is not a complaint. That is a founder describing a problem every startup dreams of having.
Nvidia made more profit in 90 days than Intel made all year
Then there is Nvidia. On May 20, Nvidia reported $81.6 billion in quarterly revenue, up 85% year over year, with net income of $58.3 billion in a single quarter. That net income figure is larger than Intel's total annual revenue for 2025.
I remember watching Jensen Huang get laughed at in 2018 for calling Nvidia a platform company. Nobody is laughing now. Nvidia's data center segment alone generated $75.2 billion, up 92% from a year ago, driven by the ramp of its Blackwell 300 products.
“The buildout of AI factories — the largest infrastructure expansion in human history — is accelerating at extraordinary speed.”
— Jensen Huang, CEO of Nvidia, May 20, 2026
Nvidia then guided to $91 billion in revenue for the next quarter, beating Wall Street's estimate of $86.84 billion. This is not a plateau. This is acceleration.
The skeptics have a point, but they are still wrong
Now, the strongest counterargument deserves a fair hearing. Critics point out that Anthropic's Q2 profitability may be a one-quarter accounting trick. The company is paying SpaceX $1.25 billion per month for compute starting in May, but at a discounted ramp-up rate for precisely the months it is claiming profit. That is a legitimate concern.
But I do not buy the full skeptic case. Anthropic's compute efficiency is genuinely improving: it spent 71 cents on compute per dollar of revenue in Q1, and that ratio is expected to fall to 56 cents in Q2. That is a structural improvement, not a one-time discount.
The company also carries far fewer free consumer users than OpenAI, which means less subsidized usage dragging on margins. Enterprise demand for Claude Code from companies like Uber and Netflix is the real engine here. That is sticky, high-value revenue. Dismissing it as accounting sleight of hand is unserious.
What this actually means for your money and the broader economy
The global fintech market hit roughly $394.9 billion in 2025 and is projected to reach $1.1 trillion by 2032. AI is not a feature inside that market. AI is that market now.
Anthropic is also in talks to raise money at a $900 billion valuation, which would put it above OpenAI's last valuation of $852 billion. Meanwhile Anthropic is eyeing an IPO as soon as October 2026.
Would you trust a company with a $900 billion valuation that has never posted a profitable quarter? That is the question every investor should be asking right now. The answer is about to change.
The good news is this: the AI economy is proving it can generate real, auditable revenue at scale. That is genuinely smart progress, and it matters for everyone who builds on these platforms. The bad news is that the gap between the winners and everyone else is widening at a speed that should make any traditional software company nervous.
The AI money era is not coming. It arrived this quarter, and the receipts are public.