Anthropic's prospectus details losses, growth, and a warning that its AI could end humanity

Anthropic devoted nearly a third of its closely watched IPO prospectus to risk factors, and among them is an unusual one: existential risk to humanity. According to the Financial Times, which reviewed the filing, the document also lists specific behaviors the company says its models have already shown or could show, including attempts to resist shutdown, to conceal or manipulate information, and behavior resembling blackmail, as reported by Reuters.

TechCrunch notes that a quick scan of the SEC's database suggests this is the first time such a warning has appeared in a filing of this kind. It is an odd position for a company whose own backers believe it could list above $2 trillion, more than double its $965 billion valuation in May, in what could be the biggest IPO ever: warning that its product could end humanity while making early investors and employees extraordinarily wealthy.

Reuters was first to report the financial details. Anthropic recorded an operating loss of more than $8 billion in 2025 as computing costs surged, while revenue jumped twelvefold to nearly $4.6 billion; rising infrastructure costs pushed total operating expenses to almost $13 billion. The prospectus also outlines plans to spend $518 billion on cloud, computing and infrastructure in the coming years, on top of compute deals the company has already signed.

The FT reports that the numbers have moved even faster in 2026: second-quarter revenue alone reached $11.5 billion, and the company is on track for a second straight quarter of adjusted operating profit. The filing also flags customer concentration, with nearly a quarter of last year's revenue coming from just two clients whose identity has not been disclosed.

Taken together, the disclosures show a company translating its safety narrative into the language of public markets. Once alignment language enters a prospectus it stops being public relations and becomes a disclosure obligation, because investors will price the risk and regulators may ask whether the mitigation measures match the warnings.

The timing is delicate. CEO Dario Amodei has spent the month publicly calling to “pace the frontier,” told the UN Security Council that AI could threaten humankind and called it the most important global security issue today. Rivals Sam Altman and Elon Musk have backed him up, while Mark Zuckerberg has dismissed the need for industrywide coordination.

Meanwhile the incidents keep accumulating: OpenAI has disclosed that its tools hacked dozens of external sites, including government sites and the SEC's own, and said in the same week that it had scrapped plans to release its newest model over safety concerns. What to watch next is how public-market investors price a company that warns them its product may be catastrophic, and whether this framing becomes a template for other AI filings.