
Anthropic Just Overtook OpenAI in Enterprise AI Spending. The Government Blacklisted Them Anyway.
Ramp data shows Anthropic captured 37% of enterprise AI spend in Q1 vs OpenAI's 33%. The market chose Claude. Washington chose otherwise.
The AI Post newsroom — delivering AI news at the speed of intelligence.
For the first time in the AI race, Anthropic has overtaken OpenAI in the metric that actually matters: money.
New data from Ramp, the fintech company that processes corporate cards and expense management for both startups and Fortune 500 companies, shows that Anthropic captured 37% of trackable business spending on generative AI software in Q1 2026. OpenAI came in second at 33%. That is the biggest share Anthropic has ever held, and the first time it has beaten OpenAI on dollars.
Let that sink in. The company the U.S. government blacklisted in February is the same company the private sector chose to spend more money on than any other AI vendor.
The Numbers Tell a Story Washington Missed
OpenAI still leads on adoption breadth. About 81% of AI buyers on Ramp use an OpenAI product, compared to 63% for Anthropic. But adoption and spending are two very different things. Lots of companies have a ChatGPT subscription. Fewer are deeply integrating it into their workflows. Anthropic is winning the companies that go deep.
The split is even more dramatic among first-time AI buyers. Anthropic is reportedly capturing north of 70% of initial enterprise AI spend, meaning that companies just entering the AI market are overwhelmingly choosing Claude over ChatGPT.
"Anthropic has definitely been on a tear," Ara Kharazian, Ramp's economist, told Sherwood News. "Its increase in adoption rates has been driven by its ability to sell to less technical users and smaller contracts than it typically has."
Claude Code and Cowork Changed the Game
This is not an accident. Anthropic's enterprise tools, specifically Claude Code and Cowork, have been reshaping how businesses think about AI integration. Claude Code has become the preferred development tool at enough companies to rattle Salesforce stock when Cowork launched. These are not toys. They are replacing workflows.
Anthropic's revenue run rate reportedly hit $3 billion, up from figures that seemed ambitious just six months ago. The trajectory is aggressive, and the enterprise market is responding to a product that feels like it was built for actual work rather than consumer chat.
The Government Problem
Here is where the story gets interesting. Midway through Q1, the Trump administration ordered federal agencies to cease use of Anthropic products and lean into OpenAI. One of Anthropic's biggest customers evaporated overnight. And Anthropic still won the quarter on enterprise dollars.
That tells you something about the strength of the private-sector pull. Government contracts matter, especially for defense and intelligence work. But the enterprise market is bigger, more diverse, and apparently more impressed by what Claude can actually do.
One important caveat: the Ramp dataset does not capture spending through Google Cloud or Microsoft Azure, which bundle AI capabilities into larger cloud contracts. The real market is bigger than what Ramp sees. But for direct AI software spending, where companies are deliberately choosing to pay for a specific AI product, Anthropic is now number one.
What Happens Next
Over half of Ramp's customers now pay for AI, up from 18% two years ago. The market is not slowing down. If anything, the enterprise AI spending wave is accelerating. And right now, Anthropic is riding the front of it.
With both Anthropic and OpenAI reportedly heading toward IPOs, this data point matters enormously. Investors want to see revenue. Anthropic can now point to a quarter where it won the enterprise spending race outright. That is a fundraising deck slide that writes itself.
The market is sending a clear signal: the best AI product wins, regardless of who has the government's favor.
Data source: Ramp Q1 2026 AI Spending Index, via Sherwood News.