Google Commits $40B to Anthropic as Claude Code Fuels AI Infrastructure Race

Alphabet's Google announced on Friday a commitment to invest up to $40 billion in Anthropic, with $10 billion arriving immediately at a $350 billion valuation. The deal comes just four days after Amazon committed $5 billion to the same company, cementing Anthropic as the most heavily backed AI lab in history — and signals how central Claude Code has become to the AI infrastructure build-out.

The Deal Breakdown

Google's latest investment expands an existing, multiyear partnership. The $10 billion immediate cash infusion is paired with an additional $30 billion that can be unlocked when Anthropic hits certain undisclosed performance targets, according to Anthropic's announcement. Alongside the capital, Google Cloud is committing 5 gigawatts of compute capacity to Anthropic over the next five years, with room to scale further.

The announcement follows Anthropic's April 20 deal with Amazon: a $5 billion investment paired with a commitment to spend up to $100 billion for around 5 gigawatts of AWS compute capacity over time. Combined, Google and Amazon are pledging the equivalent of 10 gigawatts of capacity to Anthropic — a scale of infrastructure commitment that dwarfs anything the AI industry has seen before from a single company.

Earlier this month, Anthropic had already announced a separate partnership with Google and Broadcom for 3.5 gigawatts of TPU-based capacity starting in 2027. Friday's deal expands that arrangement significantly.

Why Claude Code Is at the Center of This

Anthropic's revenue trajectory makes the scale of Google's bet legible. The company hit $30 billion in annualized revenue by early April 2026, up from approximately $9 billion at the end of 2025 — a more than 3× increase in roughly four months.

The primary driver is Claude Code. Anthropic's terminal-first coding agent, which lets developers write, refactor, and debug code across entire repositories with minimal prompting, has become a core part of many engineering workflows. Per CNBC, "Anthropic's annualized revenue has topped $30 billion. The company's AI products, particularly its Claude Code assistant, have exploded in popularity over the last year."

That growth has created infrastructure strain. Anthropic has faced widespread complaints about Claude Code usage limits in recent weeks, which the company acknowledged and partially addressed with a bug fix on April 20 that reset limits for affected subscribers. The Google deal — combined with Amazon's earlier commitment — is the direct, long-term answer to that compute bottleneck.

The Infrastructure Race in Context

The broader pattern matters for developers building on Claude Code or the Claude API. OpenAI has moved aggressively to lock in compute through a web of deals with Cerebras and other chip suppliers. Anthropic, now backed by $40 billion from Google and $5 billion from Amazon, is securing equivalent capacity to train and serve next-generation Claude models at scale.

More compute means headroom for higher rate limits, faster inference, and larger model runs. It also means the heavily restricted Mythos Preview — Anthropic's most powerful model, currently available only to vetted security organizations — has a clearer path toward broader release as infrastructure catches up with demand.

Google's position in this deal is worth noting. Despite backing Anthropic at $40 billion, Google continues to develop Gemini as a direct competitor to Claude. That dual-track strategy has drawn regulatory scrutiny in both the US and EU, but shows no sign of slowing down. Google provides Anthropic with TPU-based compute that represents one of the few credible alternatives to Nvidia's GPUs — a structural advantage Google holds over Amazon and Microsoft in this race.

Anthropic's valuation stood at $350 billion as recently as February. Investors have since been eager to back the company at $800 billion or more, according to Bloomberg. The company is also reportedly considering an IPO as early as October 2026.

What's Not Confirmed

Anthropic has not disclosed what the specific performance targets are that would unlock the remaining $30 billion from Google. The exact timeline for when the 5 gigawatts of new Google Cloud capacity comes online is also unclear — existing Broadcom and Google compute deals reference capacity starting in 2027, suggesting the additional headroom is 12–18 months out.

For developers, the near-term reality is unchanged: Claude Code remains rate-limited, and the infrastructure investments being funded today will take time to materialize. But the direction of capital is unambiguous — Anthropic is being built to scale, and Claude Code is the engine driving it.