Report warns AI lawsuits are exposing gaps in business insurance

In short: A new report says lawsuits tied to AI are growing, and many of the risks may fall into “silent” coverage inside standard insurance policies.

What's going on

A report by the Artificial Intelligence Underwriting Company, co-authored with researchers from Anthropic and OpenAI, says insurers may be exposed to AI risks in ways they did not plan for. It estimates that more than 90% of insurers’ exposure to AI “agents” is sitting in “silent” cover inside regular policies.

“Silent” cover means the policy never mentions AI, but the wording might still make the insurer responsible. It is like finding out your home insurance also covers a new kind of damage you never discussed, and nobody priced it in.

The report says the risk is changing as companies move from simple chatbots that only talk, to AI “agents” that can take actions on their own. That shift could lead to a wider set of legal claims, including professional negligence, cyber attacks, and in extreme cases, wrongful death.

Recent cases show the range of problems. Google is defending a lawsuit seeking at least $110 million tied to its AI Overviews feature, which allegedly defamed a US solar company. Air Canada was ordered to honor a discount that its customer service chatbot made up. In another case, UK engineering group Arup lost about $25 million after fraudsters used AI-made video and audio, known as deepfakes (realistic fakes), to trick an employee into sending money.

What to watch

The report warns that a major AI-related incident could trigger fights over whether insurers must pay, or lead insurers to pull back coverage. The authors compare this to what happened after September 11, when terrorism insurance became hard to get until governments stepped in. Insurers may respond by writing clearer rules, excluding some AI risks, or selling new policies designed specifically for more autonomous AI systems.

Source: Financial Times