Chubb CEO Greenberg: Industry Has to ‘Put Arms Around’ AI Liability

The insurance industry has dealt with emerging risks in the past, and Chubb CEO Evan G. Greenberg said he thinks the same approach can be applied to the rapid escalation of apparent liabilities presented by business’ use of artificial intelligence.

Speaking during a conference call to discuss third quarter earnings, Greenberg was asked about insurance coverage for AI liability.

“As an industry, we have to put our arms around this,” he said, adding that the industry “had to do this with emerging liabilities over the last 50 years, and there’s a pretty good template to follow.”

Greenberg described AI liability as a “growing and emerging exposure” which is “not contemplated in casualty underwriting to date.”

He said he can imagine “a combination of coverage clarifications, ultimately – exclusions and separate AI-related liability coverages.”

“This will all emerge along with underwriting and pricing to discern between risks,” he added. “AI-related liability exposures will vary between insureds and type of insureds…to assist businesses generally and add to greater certainty overall in the liability environment as it emerges.”

Greenberg went on to contemplate the materialization of some certification standards – best practices in the use and management of AI and agentics to also assist in insurance underwriting.

At Chubb, Greenberg said the insurer has governance and management to protect the data it holds, as well as its agentics and large language models. Best practices consider “how we control and limit what they can do, where they can do it, how they can do it, how we instruct for it – and then protect our own perimeter [with] multiple layers of detection and defense against third-party or external use of agentics and AI that could imperil Chubb’s data or systems.

“We invest significantly in this. We pay very close attention to it,” the CEO said.