Google Has Quietly Built A 14% Stake In Anthropic

Google might be focusing on publicizing its own AI models and products, but it’s quietly built a substantial stake in an AI upstart.

Google has built a 14 percent stake in Anthropic, New York Times reports. Anthropic is currently the second highest valued AI startup with a valuation of $61.5 billion, and builds the Claude series of models. As per the filings, Google can own only up to 15 percent of Anthropic. Also, Google holds no voting rights, no board seats and no board observer rights at the startup.

The filings also revealed that Google is set to invest an additional $750 million in Anthropic in September through convertible debt. In total, Google has invested more than $3 billion in Anthropic.

Interestingly, Google isn’t the only tech giant that’s invested in Anthropic. In March last year, Amazon had announced a $4 billion investment in the company. The deal had come with increased collaboration between the two companies — Anthropic had also announced that AWS would become its primary cloud provider, and also said that it would use AWS Trainium and Inferentia chips to train and deploy its future foundation models.

OpenAI might have gone mainstream, but Anthropic is one of the most promising AI startups around. Its Claude series of AI models is widely regarded to be the best for coding, and it powers AI IDEs like Cursor. Anthropic’s potential has also been recognized by investors, and it’s now worth $61.5 billion.

Interestingly, Google owns 14 percent of Anthropic, which would help hedge its own AI bets. Google itself has released several impressive AI products and models, including its Gemini models, its video model Veo2, and products like Notebook LM. Google, though, is also an investor in the second-most valuable AI startup, which provides it a buffer against OpenAI, which is currently the most valuable AI startup with a valuation of over $150 billion. Microsoft, too, has taken a similar path — it has a robust AI division of its own headed by former DeepMind executive Mustafa Suleyman, but also owns half of OpenAI. With AI being a complete unknown, and progress being hard to predict, it appears that the top tech companies have looked to spread their bets around as they navigate the AI revolution.

Posted in AI