The company once dismissed as OpenAI’s cautious little sibling now captures nearly three-quarters of all first-time enterprise AI spending. That’s not a stat. That’s a power shift.
The numbers moved fast. Ten weeks ago, Anthropic and OpenAI split first-time enterprise spending right down the middle — 50/50, a clean tie. It was 60/40 in OpenAI’s favor as recently as early December. Today, Anthropic captures over 73% of all spending among companies buying AI tools for the first time.
In enterprise software, that kind of reversal doesn’t happen quietly. It signals something structural — a shift in who enterprises trust, what they’re willing to pay for, and how they define winning in AI.
The Revenue Gap Is Real — But So Is the Momentum Gap
OpenAI says it’s on pace to generate $25 billion in revenue this year, versus Anthropic’s $19 billion. On paper, OpenAI still leads. In the trajectory, the story looks different.
Anthropic exited 2025 generating revenue at a $9 billion annualized rate. In February, that figure reached $14 billion. A few weeks later, Bloomberg reported it had soared to $19 billion — a doubling in just two months.
Anthropic’s recent growth rate sits at roughly 10× per year, compared to OpenAI’s 3.4×. A revenue crossover is now projected as early as August 2026, at roughly $43 billion in annualized revenue for both companies.
That’s not gradual catch-up. That’s compression at speed.
The Strategic Divergence Hiding in Plain Sight
Two companies built on the same foundational technology made opposite bets about who their customer was. Now we’re watching the scorecard update in real time.
OpenAI built ChatGPT into a global consumer phenomenon. Anthropic quietly built trust with procurement teams, compliance officers, and engineering leads. The first strategy dominates mindshare. The second dominates revenue per customer.
By 2026, the difference has become clear: OpenAI is a consumer company making enterprise products, and Anthropic is an enterprise company that has a consumer product.
OpenAI is losing money from consumers since it’s subsidizing the cost of their token usage. Scale at a loss is a venture bet. Anthropic is running a different race entirely.
Claude Code Changed the Conversation
No single product better illustrates Anthropic’s enterprise momentum than Claude Code. Claude Code had a ChatGPT moment, but specifically for engineers — who started shipping software at speeds that felt almost physically impossible.
The numbers back it up. Claude Code achieved $1 billion in run-rate revenue just six months after becoming publicly available. Business subscriptions have since quadrupled since the start of 2026, with enterprise use now representing over half of all Claude Code revenue.
More than 500 customers now spend over $1 million annually on Claude, up from roughly a dozen two years ago. Eight of the ten largest Fortune 10 companies are now Claude customers.
Product-market fit in enterprise AI doesn’t announce itself loudly. It shows up in renewals, procurement approvals, and engineers telling their managers they won’t go back. That’s exactly what happened.
OpenAI Reads the Room
Amid the tightening race, the Wall Street Journal reported that OpenAI is considering a strategy shift — pulling back from wide-ranging consumer bets like video generators, browsers, and devices — to a tighter focus on enterprise.
That’s a significant tell. When a company with hundreds of millions of users starts looking at where Anthropic has been quietly building, it means the enterprise model has proven itself.
Both companies are now reportedly racing toward IPOs. Whichever files first puts real numbers in an S-1 — revenue mix, margins, cost of compute, path to profitability. Every analyst covering the second company’s offering will use the first as a benchmark. Or
Who’s Actually Winning?
“Winning” depends entirely on your time horizon.
OpenAI is finalizing a funding round at a valuation exceeding $850 billion, while Anthropic closed a $30 billion Series G in February 2026 at a $380 billion valuation. By that math, the market prices OpenAI at more than double Anthropic.
But markets price narratives. Enterprise spending data prices reality.
The share of U.S. companies paying for Anthropic’s tools hit 20% in January, up from roughly 4% a year earlier. A five-fold increase in enterprise penetration in twelve months. Those aren’t trials — those are active deployments where switching costs compound with every quarter of usage.
Anthropic projects reaching cash-flow breakeven in 2028, after stopping cash burn in 2027. One company is striving toward profitability while scaling. The other is burning capital to maintain infrastructure dominance and consumer reach. Both are rational strategies — but only one works without perpetual external capital.
A 73% share of first-time enterprise buyers isn’t a blip. It reflects months of deliberate positioning: safety-forward messaging that resonates with regulated industries, a coding tool that became load-bearing infrastructure for engineering teams, and a business model structured around the customer who renews rather than the user who churns.
Anthropic didn’t beat OpenAI in mindshare. ChatGPT still dominates consumer AI by every measure consumers care about. But the enterprise CIO signing a multi-year contract isn’t asking which chatbot has more cultural relevance. They’re asking which model their engineers won’t stop talking about — and which vendor their legal team can actually approve.
Right now, the answer to both questions is increasingly the same one.
Related: Is Claude Replacing Office Jobs? IBM’s 13% Drop Says a Lot
