Anthropic lines up $1.5B AI venture with Blackstone, Goldman Sachs

Anthropic is close to finalizing a roughly $1.5 billion joint venture with Blackstone, Goldman Sachs, and several other Wall Street firms to distribute artificial intelligence tools to private-equity-backed companies.

A report by The Wall Street Journal said the platform will introduce AI applications across finance, operations, customer service, analytics, and enterprise software.

Anthropic, Blackstone, and Hellman & Friedman are leading the effort, with each expected to commit about $300 million to the venture. Goldman Sachs is set to join as a founding investor with an estimated $150 million contribution. 

The structure brings together major financial institutions and an AI developer in a single push to commercialize enterprise-grade tools. The report noted that a formal announcement could come as early as May 4.

Interest in Anthropic has accelerated in recent months as its enterprise-focused AI products gain traction. The company is reportedly considering a new funding round that could lift its valuation beyond $300 billion, with some projections pointing as high as $900 billion. Those expectations have drawn strong attention from private equity players seeking early exposure to AI infrastructure and software providers.

The planned venture also arrives as competition intensifies. Rival OpenAI has been exploring similar partnerships with private-equity firms to expand adoption of its tools across business operations. The parallel efforts show how leading AI developers are turning to financial sponsors to scale deployment and integrate automation into portfolio companies.

Both Anthropic and OpenAI are also seen as potential candidates for an initial public offering later this year, adding another layer of urgency for investors looking to secure positions ahead of any listing.

Separately, Anthropic has entered early-stage discussions with UK-based semiconductor startup Fractile. Talks are focused on securing access to specialized inference chips designed to run trained AI models more efficiently.

Such hardware is critical for lowering operating costs and improving processing speeds as demand for AI workloads grows.

The discussions underline how developers are working to lock in compute supply alongside expanding their software reach through partnerships like the proposed joint venture.