Humanity can't afford not to spend $800B on AI, Anthropic insists
Anthropic has warned that its artificial intelligence ambitions could require commitments of up to $807 billion, as the company prepares for a proposed stock market listing.
By The Register
Anthropic has warned that its artificial intelligence ambitions could require commitments of up to $807 billion, as the company prepares for a proposed stock market listing.
The figures were disclosed in an S-1 registration filing with the US Securities and Exchange Commission. The filing sets out future minimum payment obligations linked to computing and infrastructure agreements.
Anthropic said agreements entered into after 31 December 2025 carried around $417 billion in future obligations. Further deals signed or amended since 30 June 2026 added approximately $245 billion, although some of those commitments depend on contractual conditions.
Taken together, the obligations amount to between $782 billion and $807 billion, depending on which commitments are treated as firm.
The spending is intended to secure the computing capacity needed to develop Anthropic’s Claude artificial intelligence models and support the company’s services. The Register reported that the commitments involve large cloud providers, specialist computing companies and equipment suppliers.
Anthropic’s filing argues that maintaining access to vast amounts of computing power is necessary to improve its models and meet anticipated customer demand. The scale of the planned investment is substantially greater than spending on the chips, electricity and internet infrastructure needed to operate current systems.
The company is seeking to raise money through an initial public offering on Nasdaq, where it plans to use the ticker symbol ANT.
Anthropic reported revenue of $16.3 billion for the first six months of 2026, compared with $1.2 billion during the same period a year earlier. It also recorded an $80.7 billion net loss, although $72.4 billion of that figure was attributed to an accounting adjustment linked to the company’s valuation rather than direct spending.
The filing shows the financial pressure facing AI developers as they compete to secure long-term access to computing infrastructure. Anthropic’s disclosed commitments stretch across several years and include agreements that have not all reached the same level of certainty.
The company’s proposed flotation remains the next confirmed step in its plan to raise capital for further AI development.