Anthropic's IPO Filing Flags Negative AI Coverage as Business Risk

Anthropic, one of the largest artificial intelligence labs alongside OpenAI, filed to go public on Monday. In its S-1 with the SEC, the company paired grand claims about AI's potential-such as curing long-standing diseases-with warnings about catastrophic or existential risks. It also flagged a more immediate concern: the press.

The filing, from Dario Amodei's company, returns repeatedly to the threat of unfavorable coverage. The risk factors section states that as interest in Anthropic and the AI industry has grown, so has media scrutiny, including recurring narratives about job displacement, misuse, and harm. Negative coverage, the company warns-including statements from public figures, "whether accurate or not"-could hurt its brand and reputation, reduce market acceptance of its models and applications, and make it harder to attract and retain customers. Persistent or intensified criticism, it adds, could depress demand and damage its business and financial results.

Copyright litigation is another listed risk. The filing catalogs lawsuits and threat letters from book, film, music, and other rights holders, noting Anthropic may need licenses it might not secure. The company says it believes it has meritorious defenses and disputes any wrongdoing, while acknowledging that defense costs, settlements, and diverted management attention could harm it.

The filing also sets out strategy. Unlike SpaceX and OpenAI, Anthropic says it does not currently develop video or image generation models despite their commercial appeal, directing that compute instead toward coding, agentic skills, and safety research-a choice it says has aided capital efficiency. It likewise has no advertising business, though it disclosed $147 million in advertising spend in 2025, with marketing and advertising expenses rising by $273 million in the first half of this year, reflecting higher media spend for product launches.