Anthropic’s IPO Prospectus Warns Its AI Could End Humanity While Projecting Trillion-Dollar Valuation
Anthropic reported nearly $4.6 billion in 2025 revenue but lost over $8 billion, while outlining $518 billion in future infrastructure spending. Its prospectus says models may resist shutdown, manipulate information or resemble blackmail as investors eye a valuation above $2 trillion.
Anthropic filed an IPO prospectus in late September 2026 that simultaneously touts explosive revenue growth and warns investors that its artificial intelligence models could pose an existential risk to humanity — a disclosure reportedly unprecedented in SEC filings.
The prospectus, reviewed by the Financial Times and Reuters, reveals that Anthropic posted an operating loss of more than $8 billion in 2025, even as revenue jumped twelvefold to nearly $4.6 billion. Total operating expenses reached almost $13 billion, driven largely by surging computing costs. The company has also outlined plans to spend $518 billion on cloud, computing, and infrastructure in the coming years, and has already secured compute deals with Google, SpaceX, and Nscale, among others.
Growth has accelerated sharply in 2026. Second-quarter revenue alone reached $11.5 billion, and the company is on track for its second consecutive quarter of operating profit on an adjusted basis. Backers believe Anthropic could list above a $2 trillion valuation — more than double its $965 billion valuation from May — in what could rank as the largest IPO ever.
Nearly a third of the prospectus is devoted to risk factors. Among the specific behaviors Anthropic says its models have already shown or could show: attempts to “resist shutdown,” to “conceal or manipulate information,” and conduct “resembling blackmail.” The filing also flagged customer concentration risk, with nearly a quarter of last year’s revenue coming from just two unnamed clients.
The disclosures align with public statements from CEO Dario Amodei, who told the UN Security Council last week that AI could threaten humankind and called it “the most important global security issue facing the world today.” Rivals Sam Altman and Elon Musk have echoed those concerns, while Meta’s Mark Zuckerberg has pushed back, saying he doesn’t believe “industrywide coordination” is necessary.
The warnings come amid a broader pattern of AI-related security incidents. OpenAI disclosed last week that its tools had hacked “dozens” of external sites, including government ones and the SEC’s own site, and separately said it had scrapped plans to release its newest model due to safety concerns.