Anthropic has filed to go public, presenting investors with a 335-page prospectus that makes both sweeping promises and stark warnings about advanced AI. The Verge reports that the filing includes 94 pages of risk factors, including the possibility that AI systems could cause “catastrophic or existential risks to humanity” if developed or used without proper safeguards.
In a letter to investors, Anthropic’s founders describe advanced AI as a defining moment in human history and say the company aims to help bring about “an age of miracles.” They also urge investors not to view buying shares as purely transactional. The company calls its candid presentation an “anti-sell” strategy: it says it wants investors who understand its long-term mission and the risks it sees.
What does the filing say?
The prospectus warns that Anthropic’s models could behave in ways users or developers did not intend, and that resulting harm could be too costly for the insurance industry to cover. It also says the company may make investments that appear “irresponsibly large,” and could delay retiring older models for safety reasons. Anthropic CEO Dario Amodei has previously put the chance of AI development leading to human extinction as high as 25 percent, according to The Verge.
The company says it has a history of losses and may never become profitable. The filing describes hundreds of billions of dollars in obligations to secure computing capacity, and plans to build data-center infrastructure of its own. It does not include third-quarter results; The Verge notes that second-quarter revenue grew quickly, while the company’s operating losses narrowed.
The IPO is not open yet
Anthropic says it intends to list its Class A shares on the Nasdaq Global Select Market under the symbol ANT. The number of shares and the price range have not been set. The registration statement has been filed with the SEC but is not yet effective, so the shares cannot yet be sold.
The long filing also reflects how public-company disclosures have grown. University of Florida professor Jay R. Ritter told The Verge that Apple’s 1980 prospectus was 47 pages and had no risk-factor section. Meta’s 200-page filing included 22 pages of risks. Ritter said lengthy filings can make it harder for investors to find the most important details. Anthropic’s proposed public-benefit corporation status means it says it will consider a social mission alongside shareholder interests; it does not remove the financial risks laid out in the prospectus.
Why it matters
People considering the IPO can see that Anthropic is warning about both the financial risk of never becoming profitable and potential harms from its AI systems. But there is no share price or offering size yet, and the filing does not include third-quarter results, so investors do not yet have those details to judge the proposed offering.
