Google (NASDAQ:GOOGL) (NASDAQ:GOOG) will invest $10 billion in AI startup Anthropic, with a potential additional $30 billion tied to performance milestones, according to a report published on Friday.
According to a Bloomberg report, Anthropic said Google’s initial $10 billion cash investment values the startup at $350 billion, in line with its February funding round, excluding the latest capital raised. The search engine giant could invest a further $30 billion if certain targets are met, the report added.
Anthropic and Google did not immediately respond to Benzinga’s requests for comment.
Compute Demand Ramps Up
Before this, Google’s investment in Anthropic was more than $3 billion, and it owned a 14% stake in the company, according to a New York Times report.
Strong demand has prompted Anthropic to sign several major deals recently to acquire more computing capacity.
The investment comes weeks after Anthropic announced a deal with Google and chipmaker Broadcom (NASDAQ:AVGO) to secure 5 gigawatts of computing capacity, set to begin coming online next year, with scope to expand further over time.
Through its cloud business, Google offers customers access to Anthropic’s Claude models, putting it in competition with Amazon’s (NASDAQ:AMZN) AWS and Microsoft‘s (NASDAQ:MSFT) Azure, while also promoting its tensor processing units (TPUs) as an alternative to Nvidia (NASDAQ:NVDA) chips. At the same time, Google remains a rival to Anthropic through its Gemini AI models and services.
Trump Woes And Mythos
Anthropic’s tensions with the Donald Trump administration escalated earlier this year after the company resisted Pentagon demands to loosen guardrails around military uses of its models, triggering a government freeze on some federal dealings and a broader legal and political standoff.
But relations have recently shown signs of thawing, with the administration backing away from a full confrontation after quiet talks with Anthropic leadership, according to POLITICO.
The dynamic has shifted further with Anthropic’s launch of Mythos, a powerful cybersecurity-focused model. The company has kept it from public release over concerns it could be misused by cybercriminals. However, the National Security Agency NSA is reportedly using this restricted “preview” model for cyber defense, even though the Pentagon has flagged the company as a supply chain risk.
Image via Shutterstock
Recent Comments