LessWrong Author James_Miller: Fear of AI-Inflicted Suffering May Itself Constitute a Material Risk Requiring Prospectus Disclosure
James_Miller argues that under Item 105 of Regulation S-K and the probability-times-magnitude standard from Basic v. Levinson, even if the risk of AI inflicting prolonged suffering never materializes, public fear of it could damage an AI company's value by triggering political and religious opposition—making it a material risk factor that the prospectus must disclose.
Reuters reported that Anthropic's draft prospectus warns of "catastrophic or existential risks to humanity." LessWrong author James_Miller argues that the prospectus also omits a category of risk capable of destroying the company's value: AI subjecting humans to prolonged suffering.
His legal reasoning begins with Item 105 of Regulation S-K, which requires a prospectus to discuss "the material factors that make an investment risky." The U.S. Supreme Court in Basic v. Levinson endorsed weighing an event's probability against its magnitude for the company: even an event with very low probability may constitute material information if its impact is sufficiently large. James_Miller applies this to argue that even if the probability of AI inflicting suffering is extremely low, its potential magnitude—"a billion years of torment"—is enough to cross the materiality threshold.
The key step is that fear itself can cause damage without the catastrophe ever occurring. James_Miller writes, "Public fear of catastrophic suffering could destroy Anthropic's value even if the catastrophe never occurs." He cites the Supreme Court's holding in Matrixx Initiatives v. Siracusano: reports linking a drug to loss of smell could be material even without statistically significant evidence of causation, because regulators sometimes act before causation is proved. By analogy, credible warnings about AI suffering could affect regulation, demand, and financing before the underlying danger is conclusively established.
To illustrate how fear translates into political pressure, James_Miller cites the Pew Research Center's 2023-24 survey finding that 55% of American adults believe in hell. He argues that the prospect of AI creating something resembling hell could mobilize religious opposition and demands for restrictions on AI development and deployment—restrictions that could delay products, increase costs, or prevent profitable uses of the technology.
Real-world signals already exist. On October 14, 2026, New York City Comptroller Mark Levine urged Anthropic to pause its public offering until it addressed safety, governance, and development concerns. Levine serves as investment adviser, custodian, and trustee of the city's pension funds, which hold more than $325 billion, with approximately $7 million in exposure to Anthropic through private equity holdings. James_Miller simultaneously acknowledges that Levine's letter concerns AI safety broadly and does not specifically address suffering risks; nor does the Pew survey establish how investors or religious groups would respond specifically to suffering risks.
James_Miller also invokes the SEC's 2010 climate-disclosure guidance, which recognizes that public perceptions of a company's emissions data could cause financial harm through reputational damage and directs companies to consider that potential harm for risk-factor disclosure. Additionally, Item 303 requires management to discuss "known uncertainties reasonably likely to materially affect revenues, income, or liquidity"—if management knows that safety concerns are threatening financing or creating a sufficiently concrete prospect of costly restrictions, this provision could require discussion alongside the risk factors.
On the securities-law liability front, James_Miller notes that under Section 11 of the Securities Act, purchasers whose shares are traceable to an effective registration statement may sue over an omission of material facts required to be disclosed. The issuer can be liable without intent to deceive. This means that even before any catastrophic suffering occurs, omitting suffering-risk disclosure itself could trigger lawsuits.
James_Miller lists multiple hypothetical scenarios in which AI could inflict suffering—religious punishment, rival AIs threatening humans with torture to compel cooperation, the "frozen smiles" thought experiment from Nick Bostrom's Superintelligence, factory farming for humans, suffering reproduced in simulations, and others—but states that these scenarios serve to illustrate why fear has political mobilization potential, not to prove probability. He explicitly acknowledges the boundary: "The strength and context of the evidence matter; merely imagining a frightening outcome does not establish materiality."
His final conclusion is conditional: if that threat is material when Anthropic offers its shares, it belongs in the prospectus. The author does not assert that the SEC will definitely mandate such disclosure; rather, he argues that the existing legal framework can logically accommodate this requirement.