Anthropic could spend $750B on computing power ahead of IPO: Report
Claude developer’s first-half net loss reaches $80.7B, largely driven by accounting charges, Wall Street Journal reports
Artificial intelligence company Anthropic could spend $750 billion on additional computing capacity in the coming years as it prepares for an initial public offering (IPO), The Wall Street Journal reported Tuesday.
The figure includes deals still under negotiation, the newspaper reported, citing a draft IPO prospectus.
The developer of the Claude AI assistant recorded a net loss of $80.7 billion in the first half of 2026, compared with $13 billion in the same period a year earlier, according to the report.
About $72 billion of the latest loss stemmed from accounting charges related to changes in the value of convertible notes and other financial instruments issued by the company.
Anthropic spent nearly $9 billion on computing capacity and infrastructure in the six months through June, up from $2.3 billion a year earlier. First-half spending also exceeded the $7.3 billion recorded for 2025.
The draft prospectus outlined computing agreements totaling around $450 billion during the next seven to 10 years, most of which cannot be canceled, the Journal reported.
It includes planned spending of $111 billion on Google Cloud contracts, $110 billion with Amazon Web Services and $31 billion with Microsoft.
Anthropic is expected to publicly file IPO documents in the coming weeks, as it seeks additional funding for chips, data centers and staff to develop its AI models.
The prospectus also highlighted risks associated with AI systems, including the possibility of models displaying behavior aimed at avoiding shutdown or producing responses resembling blackmail, according to the report.
The report also identified political opposition to data center construction in the US, where most of Anthropic’s computing capacity is located, as a potential risk.
