Anthropic, the artificial intelligence company known for its Claude chatbot, recorded a net loss of more than $80 billion in the first half of 2026, the Wall Street Journal reported on October 13. The figure comes as the market awaits the company's initial public offering.
The loss is a sharp increase from $13 billion in the same period a year earlier. The main driver is understood to be roughly $72 billion in accounting charges linked to changes in the value of convertible notes and other financial instruments Anthropic issued to raise capital.
Anthropic also said it could spend up to $750 billion in the coming years to buy additional computing capacity, including potential deals now under negotiation, according to the Wall Street Journal. The figures appear in a draft prospectus, the legal document a company files with financial regulators ahead of an IPO.
The financial picture lays bare the enormous costs of the AI race, as investors increasingly scrutinize whether the money pouring into chips and data centers will pay off.
The company also listed potential risk factors, including uncertainties over how to prevent its products from causing harm, for example through sophisticated cyberattacks or extortion carried out by uncontrolled models.
Anthropic did not respond to a request for comment. The report is based on non-public information and could not be immediately verified.
Investors reconsider valuation
Earlier, in September, Reuters reported that Anthropic lost $42 billion on a net basis in 2025, part of a series of exclusives drawing on the company's IPO prospectus.
Anthropic has benefited from the popularity of Claude Code, a programming tool developers use to write software and build applications. But according to four people familiar with the matter who spoke to Reuters, some current and prospective investors who attended Anthropic's Investor Day event in San Francisco last week are weighing a lower valuation for the company following the latest financial disclosures.
Attendees discussed a valuation of about $1.5 trillion, according to three sources — a significant drop from the roughly $2 trillion once expected for the November share sale.
According to the Wall Street Journal, Anthropic's revenue in the six months ended June 30 rose to $16.3 billion from $1.1 billion a year earlier, while spending on AI infrastructure nearly quadrupled to almost $9 billion.
Last week, Anthropic also published a report arguing that AI may be only months away from being able to self-replicate on ordinary household devices.
