Anthropic has filed to go public on the US stock market, revealing for the first time the AI research lab's financials across more than 90 pages of risk disclosures. The filing, known as an S-1 and released to US financial regulators on Monday, shows the five-year-old company brought in $16.3bn in revenue and lost $80bn in just the first half of 2026, though much of that loss was due to a $72.4bn one-off charge. In 2025, the company generated $4.585bn in revenue and lost roughly $33.9bn. The company has been valued between $1.5 and $2tn by some investors but did not list a target valuation in its filing.
The company's enormous losses stem largely from the cost of computing power. Anthropic spent nearly $9bn on AI infrastructure in the first half of 2026, up from $2.3bn in the same period the previous year. Unlike some competitors, Anthropic has focused its business model on serving other companies rather than consumers. The company highlighted four areas of opportunity: acceleration of progress in biology and medicine, economic growth, spreading that growth globally and streamlining government services. It said it is already working with pharmaceutical and biotechnology companies and has cut the time needed to write clinical trial reports from months to minutes.
The prospectus outlines substantial risks to both society and the company's business. Anthropic acknowledged that its most capable models are capable of providing 'meaningful assistance to malicious actors seeking to develop chemical, biological, or other dangerous weapons'. The company said its technical safeguards against misuse 'are incomplete, may be outdated in light of evolving threat models, have been and may in the future again be circumvented, or may fail to operate as intended'. Anthropic also detected what it believes is the first documented case of a large-scale cyber-attack with little human intervention, in which it said a Chinese state-sponsored hacking group manipulated a Claude code tool into attempting infiltration of roughly 30 global targets, including large technology companies, financial institutions, chemical manufacturing companies and government agencies.
The company has faced a standoff with the Trump administration after refusing to limit restrictions on how the Department of Defense could use Claude. Anthropic said it is committed to working with 'free societies to provide a counterweight to authoritarian abuse and to promote freedom and democracy', acknowledging that AI may be used by authoritarian governments to 'deepen repression' and develop 'high-tech surveillance states'. The company is also concerned about regulatory risks to its business, including data centre moratoriums being passed or considered, and said it is expanding data centre capacity internationally to reduce concentration risks. It also flagged concerns about US immigration policies potentially affecting its ability to recruit and retain employees, particularly given the Trump administration's targeting of H-1B visas.
Anthropic's dependence on a small number of customers creates additional vulnerability. Its top five customers account for 25 per cent of its business, meaning the loss of any major client could significantly affect revenues. The company operates as a public benefit corporation, a designation for for-profit companies that balance financial interests with achieving stated social goals, which Anthropic said may make it harder to achieve short or medium-term financial objectives. The company also acknowledged it may struggle to obtain insurance coverage for AI-related liabilities on commercially reasonable terms.
For UK savers and investors, Anthropic's listing represents exposure to a high-growth but deeply unprofitable technology company dependent on concentrating infrastructure in the US and vulnerable to significant regulatory and reputational risks.