Anthropic Hits $65 Billion Annualized Revenue as Claude Fuels Enterprise Rush
Anthropic disclosed to investors that its annualized revenue run rate surpassed the $65 billion mark by the end of July 2026 — a jump from $47 billion in May and from just $9 billion at the end of 2025. The disclosure, reported by Bloomberg and confirmed by CNBC, positions the creator of the Claude language model on a growth trajectory that few in the industry anticipated two years ago, and feeds ambitions for an IPO that investors already project at $2 trillion.
From $10 million to $65 billion: the trajectory
Anthropic's rise is, in itself, a rare case study in how quickly AI infrastructure can become a revenue engine. In 2022, the company was generating about $10 million annually. By January 2026, it was at $9 billion in annualized revenue. Between then and the end of July, annualized revenue grew nearly sevenfold — an increase of $56 billion in just nine months.
The curve was not linear. By February 2026, the figure had already been reported at $19 billion. In May, it reached $47 billion. Now, in July, it exceeds $65 billion. The acceleration coincides with the mass adoption of Claude Code — the AI-assisted programming tool, which according to the company now serves over 300,000 business customers, more than 1,000 of whom spend over $1 million annually.
The math is straightforward, but the phenomenon is extraordinary. Claude Code reached $500 million in annualized revenue by September 2025, jumped to $2.5 billion in February 2026, and now accounts for an increasingly large share of the total — with no indication of deceleration.
The enterprise bet
The engine behind this growth is Anthropic's strategy of focusing almost exclusively on the corporate market. While competitors seek mass consumer models, Anthropic built its offering around long-term contracts, data-sensitive handling security, and deep integrations with development tools.
Uber, for instance, rolled out Claude Code in December 2025 and had already spent its entire 2026 AI budget by April — an expense that CEO Marty Kausas noted reached $4,000 in just three days of use by a single engineer. Microsoft, Samsung, and T-Mobile are among the clients investing most heavily in Claude for internal automation.
Anthropic's value proposition also rests on its security positioning: while OpenAI and other competitors face debates about model containment and alignment, Anthropic sells the opposite — a model designed to be traceable, auditable, and restricted in potentially dangerous capabilities. For companies handling regulated data, that narrative carries commercial weight.
The road to IPO
The $65 billion revenue announcement comes amid preparations for an initial public offering that investors already project at $2 trillion — which would make Anthropic not only the most valuable IPO in history but the most highly valued company on Earth. The confidential filing was submitted on June 1, 2026, with a Nasdaq listing expected in October.
The current post-Series H valuation of $965 billion has already been more than doubled by market expectations. Investors including Founders Fund and Baillie Gifford believe the company could surpass OpenAI in annualized revenue by the first quarter of 2027 — a projection that depends on growth rates staying above 30% per quarter.
The pricing model is also a strategic part of the growth story. Claude Sonnet 4.5 on the API charges $3 per million input tokens and $15 per million output tokens — a price that, for companies replacing development teams, seems highly competitive. Claude Code pricing, meanwhile, is significantly higher, raising questions about the sustainability of intensive use. Reports indicate companies have already faced unexpectedly high bills — a signal both of adoption and of cost risk.
What to watch
The question is not whether Anthropic will grow, but whether the market is prepared to sustain a company at that scale of revenue without demanding tradeoffs on safety, transparency, or governance. When a company generates $65 billion a year, its decisions about AI alignment cease to be academic questions — they become powerful market forces. And the next round of this experiment will be the October IPO.
Sources: TechCrunch, CNBC, Benzinga
✓ Independent sources cross-checked and verified before publishing
