SAN FRANCISCO, United States — In one of the largest corporate investments in the history of the technology sector, Google has committed to pumping up to $40 billion into the artificial intelligence startup Anthropic. The deal, confirmed by Anthropic on Friday, April 24, 2026, significantly expands the alliance between the two companies as they compete for dominance in the rapidly evolving generative AI market.
The massive capital infusion underscores the astronomical costs of developing next-generation large language models (LLMs) and the specialized computing infrastructure required to run them.
According to an Anthropic representative, the $40 billion commitment is structured as a multi-stage agreement:
- Initial Infusion: Google has provided an immediate $10 billion investment.
- Milestone-Based Funding: The remaining $30 billion will be unlocked over several years, contingent upon Anthropic meeting specific performance goals and development milestones.
- Infrastructure Synergy: As part of the deal, Anthropic will continue to leverage custom Google chips (TPUs) and Google Cloud services to train and deploy its highly advanced AI models.
The investment comes at a pivotal moment for the startup. Earlier in April, Anthropic revealed that its annualized revenues had tripled quarter-on-quarter to over $30 billion. This surge in growth reportedly allowed Anthropic to surpass OpenAI in revenue for the first time, signaling a shift in the hierarchy of the “Big Three” AI developers (OpenAI, Anthropic, and Google/DeepMind).
Google’s move follows a similar escalation by Amazon, which recently boosted its collaboration with Anthropic:
- Amazon’s Play: Amazon announced a new $5 billion investment, with an additional $20 billion potentially on the table.
- AWS Commitment: Anthropic has committed to spending over $100 billion on Amazon Web Services (AWS) technology over the next decade.
- Dual-Cloud Strategy: By maintaining deep ties with both Google and Amazon, Anthropic ensures redundant access to the massive computing power necessary to maintain its lead.
The financial windfall arrives amidst internal and external security challenges for Anthropic. The company recently announced its newest model, Mythos, which is considered so powerful that it was withheld from public release due to potential cybersecurity risks.
- Unauthorized Access: Anthropic disclosed this week that it is investigating unauthorized access to the Mythos model.
- Selective Release: The company has currently restricted access to Mythos to 40 major tech firms to help them patch vulnerabilities before the model—which Anthropic fears could be “a boon for hackers”—is more widely deployed.
The deal also follows a visit by Anthropic CEO Dario Amodei to the White House. While the meeting was described as friendly, it occurred against a backdrop of tension regarding Anthropic’s refusal to grant the U.S. military unconditional access to its models. The company has long emphasized “AI safety” and “Constitutional AI” as its core differentiators, often placing it at odds with rapid-deployment government requirements.
With $40 billion from Google and potentially $25 billion from Amazon, Anthropic is now one of the most well-funded private companies in history. The capital is expected to go directly toward the development of even more massive neural networks and the procurement of the next generation of custom silicon, ensuring that the AI arms race will continue to accelerate through the late 2020s.
