Anthropic has told prospective investors that artificial intelligence could present "catastrophic or existential risks to humanity", as the company published details of its planned stock market listing in the United States.
The maker of the Claude chatbot disclosed that it lost more than $80.7bn (about £60.5bn) in the first half of the year, while bringing in $16.3bn in revenue over the same period. The number of shares to be offered and the price range have not yet been set.
Anthropic's valuation at listing is widely expected to surpass that of SpaceX, whose debut this summer was the largest stock listing on record, at an initial valuation of about $1.8 trillion.
In its documents, the company said policymakers, regulators and researchers, as well as people inside its own organisation, had raised concerns that AI systems pose a significant risk to humanity. It said advanced systems "could, if not developed, deployed, and operated with appropriate safeguards, oversight, and security measures, pose catastrophic or existential risks to humanity, particularly if such systems were misused, replicated, weaponised, or allowed to operate with a high degree of autonomy".
Anthropic was founded five years ago by chief executive Dario Amodei, his sister Daniela Amodei and five other co-founders. It announced its intention to go public in June. Its chief rival, OpenAI, soon said it also planned to list, and a person familiar with that company's plans says the listing is now expected early next year.
Since both companies settled on those plans, a series of cyber-security and other technological breaches has involved their AI agents, which are bots built to work with a degree of autonomy. Current and former employees of both firms have also drawn public attention with fresh warnings that the technology is being built without the means to protect all of humanity from possible future harms caused by AI tools.
In its new filings, Anthropic repeated earlier pledges to develop AI in ways that benefit people, while also cautioning investors about its dangers. It acknowledged that its losses outstripped its revenue in the first half of the year, and said it would often make forward investments "that may seem irresponsibly large at first glance". The company added: "In general, we will aim to be bold but sane, and to win trust with a track record of successful bets."
Anthropic plans to list on the Nasdaq under the symbol ANT. It will be the first true AI company to go public, which makes the offering a test for the future of the technology. Tech firms have poured more than $1tn into costly AI infrastructure and tools, such as data centres and the models underpinning all AI, over the last year alone.