Anthropic crossed a $30 billion annualized run-rate in April 2026, passing OpenAI's $24 billion for the first time since the AI boom started. The reported figure represents a 30x increase from January 2025, when Anthropic's ARR was $1 billion. The most recent leg — from $14 billion to $30 billion — happened in roughly eight weeks.
How Anthropic Got There
The growth has been built almost entirely on enterprise. Anthropic never had a meaningful consumer phase. When the company announced its Series G in February 2026, more than 500 business customers were each spending over $1 million per year. Two months later that count exceeded 1,000. Eight of the Fortune 10 are now customers.
OpenAI's mix is shifting in the same direction — enterprise has crossed 40% of revenue and is trending toward parity with consumer by year-end — but it started from a $20-per-month ChatGPT base. Anthropic started where OpenAI is going.
The other unexpected detail: training spend. SaaStr's analysis notes OpenAI is projecting roughly $121 billion in 2028 compute spend; Anthropic's training costs are projected to peak at about $30 billion. That's a 4x differential. Anthropic's per-dollar-of-revenue training cost looks structurally lower at this scale.
The Disputed Number
OpenAI publicly disputes Anthropic's figure, arguing the methodologies aren't comparable. Anthropic books revenue routed through cloud partners (Google Cloud, AWS, Bedrock); OpenAI does not count cloud-partner revenue the same way. OpenAI's position is that Anthropic's number is roughly $8 billion higher than apples-to-apples accounting would show.
Even with the dispute, the gap is smaller than it has ever been. Twelve months ago, OpenAI's ARR was several multiples of Anthropic's. Today, even on OpenAI's accounting, the two companies are at near-parity, and on Anthropic's accounting, Anthropic is ahead.
Why It Matters
The narrative for two years has been that OpenAI was the consumer brand and Anthropic was the developer brand — that ChatGPT's billion users would translate to enterprise dominance through sheer mindshare. That narrative is now under direct stress. Coding has become the killer application that drives enterprise spend, and Claude has been the model enterprises buy for it. Cursor, GitHub Copilot's new agentic features, the Claude Code subscription tier, and the Anthropic API itself have collectively concentrated the dollars on Anthropic's side of the ledger.
What's left to watch: whether OpenAI's own enterprise growth keeps pace through the back half of 2026, and whether Anthropic's training-cost advantage holds as both labs scale toward the next generation of frontier models. The revenue race is no longer a foregone conclusion.
Source: SaaStr · The AI Corner
Related on Uristocrat: Anthropic's $900B-valuation funding round.
Comments