Insurers Brace for Lawsuits Over AI Agents Gone Rogue

Insurance broker Aon has reviewed more than 300 court cases tied to AI. The firm is preparing clients for a wave of lawsuits over AI agents that spin out of control. Analyst Tim Rainer leads underwriting and claims handling, the insurance work of assessing and settling risk, at Verisk, a risk-analytics company. He says the CEO ends up responsible in the end. Every company executive is required to ensure proper oversight, and AI does not change that. Sam Altman of OpenAI and Dario Amodei of Anthropic risk being held personally liable. In July, Anthropic already paid $1.5 billion to settle a copyright infringement lawsuit.

This is not scare talk. It confirms a simple rule. If an agent gets real production access with no limits on what it can touch, and no person accountable for the outcome, someone ends up paying for it. Now insurers expect to be the ones footing that bill. The $1.5 billion lawsuit against Anthropic shows the scale that lawyers and insurers are now weighing, not just developers.