Anthropic is in discussions with investors to raise between $30 billion and $50 billion in new equity funding at a valuation as high as $950 billion, the Financial Times reported on Wednesday morning citing four people familiar with the conversations. A close at the upper end of the range would make the Claude maker the world's most valuable artificial intelligence start-up, surpassing OpenAI's current $500 billion implied valuation.
The round, which would be Anthropic's seventh major equity raise since the company was founded in early 2021, is being led by a consortium understood to include Amazon — which has invested $8 billion to date — and a number of Middle Eastern sovereign wealth funds. Conversations remain at the term-sheet stage; the FT cautioned that valuation and round size are both subject to change before the close, which one source put at "late June or July".
The financial backdrop for the raise has been an unusually rapid expansion of revenue. Chief Executive Dario Amodei told the Code with Claude developer conference in San Francisco last week that the company had reached a $30 billion annual revenue run rate, an 80-fold expansion from a year earlier and roughly twice the figure the company had cited in November 2025. Roughly two-thirds of the run rate is tied to enterprise API usage, according to two investors briefed on the round.
Anthropic's growth has been driven in equal parts by the developer ecosystem around its Claude Code coding agent and by enterprise adoption of the Claude Mythos cybersecurity model launched in early May. The Mythos model, in a partnership with Mozilla announced May 8, identified and helped patch 270 vulnerabilities in the Firefox browser in a 96-hour window — a result widely circulated in security research circles and a contributing factor to several of the round's term sheets.
OpenAI launched its own equivalent cybersecurity model, called Daybreak, on May 11. The two-product race has consolidated investor interest in security-focused AI in a way that broader code-generation products have not. Microsoft, which holds the largest single position in OpenAI, is understood to be considering a top-up to OpenAI in the same window to preserve its lead.
The strategic context for the round is the Pentagon's May 1 decision to award classified AI infrastructure contracts to Microsoft, Google, Amazon Web Services, Nvidia, OpenAI, Reflection, and SpaceX — but not Anthropic. The exclusion, which followed Anthropic's public dispute with the Trump administration over military-AI safety, has cost the company defence-related revenue but has not deterred the round; one investor source told the FT that "the breadth of the enterprise base is what's driving valuation, not government".
A $950 billion valuation would imply a multiple of approximately 32 times the company's current revenue run rate — high by software standards but in line with the multiples assigned to OpenAI and Mistral in their most recent rounds. The round is structured to close before Anthropic's fiscal year-end on July 31.