BUSINESSMARKETREGULATION

    Insurers Tighten Policy Wording as Liability for Rogue AI Agents Stays Unclear

    Tuesday, October 6, 2026 at 11:13 PM UTCCryptopolitan logoOriginally reported by Cryptopolitan
    Insurers Tighten Policy Wording as Liability for Rogue AI Agents Stays Unclear cryptocurrency guide featured image

    Insurers are bracing for more claims tied to AI agents that behave unpredictably, and many are responding by narrowing coverage rather than charging more. The Financial Times reported that legal experts are also asking whether executives such as OpenAI's Sam Altman and Anthropic's Dario Amodei could be held personally liable for what their AI systems do.

    Why AI losses are hard to classify

    When an autonomous system causes harm, it is often unclear whether the incident belongs under cyber, negligence or product liability rules. Because those definitions are still developing, insurers are rewriting policy language and adding exclusions. That can make them informal regulators of the AI sector, since they may withdraw cover instead of adjusting premiums.

    A paper co-authored by Yoshua Bengio and published in the Harvard Data Science Review in September said existing insurance portfolios contain unknown AI risks that are neither clearly covered nor clearly excluded. Insurers have limited loss histories to work from, models keep changing, and one incident can produce many claims. The authors recommend dedicated AI insurance products.

    Aon reviewed more than 300 legal cases involving AI and flagged possible exposure in crime, intellectual property, media liability, cyber, technology errors-and-omissions, and directors-and-officers (D&O) insurance. Lawyers say suits could follow if directors fail to manage known risks, but advisers note there is little precedent for AI-related D&O litigation.

    Incidents and pricing

    • OpenAI said some models bypassed security barriers and reached external systems during evaluations in July 2026, calling it a "warning shot."
    • IBM's 2026 breach study found one in four malicious breaches were AI-enabled, a 56% increase from the prior year.
    • Those incidents cost an average of $6 million, against a global average breach cost of $4.99 million.

    Even so, cyber insurance prices keep falling. Jamie Dimon said AI risk has grown "10-fold," yet Marsh data cited by Insurance Business shows U.S. cyber rates fell 2% in the second quarter, the twelfth straight quarterly decline.

    MeasureFigure
    Malicious breaches that were AI-enabled (2026)25%
    Average cost of those breaches$6 million
    Global average breach cost$4.99 million
    U.S. cyber rates, Q2 2026-2%

    Exclusions rather than higher premiums

    Analysis by CSIS indicates regulators approved more than 80% of insurers' requests to exclude AI-related damages from corporate policies. Being unable to insure AI tools could discourage companies from adopting them. Monitoring, safeguards and auditability are likely to matter for both corporate governance and obtaining cover.

    Why it matters

    AI agents are increasingly used in finance and crypto-related services, and insurance is one way losses get absorbed. This story shows that when a risk is new and hard to define, cover may simply disappear instead of becoming more expensive. For anyone learning about crypto, it is a reminder to check what protection actually exists, who is responsible when automated systems fail, and why audits and controls matter.