Anthropic, a prominent artificial intelligence startup, is reportedly preparing for an initial public offering (IPO) that could value the company at approximately $2 trillion. According to reports from Reuters and the Financial Times, the company’s internal prospectus includes formal disclosures regarding the potential for its AI models to cause harm. These documents explicitly warn investors that advanced AI development could lead to "catastrophic or existential risks to humanity."

While the company has previously advocated for a more measured pace in AI development, the upcoming IPO filing underscores the tension between rapid commercial growth and safety concerns. The Verge reports that the filing also details mounting financial losses and leadership proposals designed to consolidate power within the company. The projected $2 trillion valuation would represent a significant increase from the $965 billion valuation Anthropic held just four months ago, potentially positioning it to become the largest IPO in history.

There is a slight variation in how these outlets characterize the company's stance. The Guardian emphasizes the "existential" nature of the risks, framing the disclosure within a broader industry context of AI safety concerns. In contrast, The Verge focuses more heavily on the financial mechanics of the IPO, such as the company's spending plans and the specific leadership structures intended to maintain control during the transition to a public company.