Marsh's Doyle warns of heightened cyber exposure from AI, questions broad exclusions

By Ashish Tiwari Oct 15 (The Insurer) — Marsh on Thursday said insurers should take a measured approach to managing risks arising from AI, warning that broad-based exclusions could create “real confusion and uncertainty” for policyholders as the technology gets deployed across businesses. The comments come after Marsh Risk and its reinsurance broking unit earlier this month wrote to (re)insurers expressing concerns about the growing use of broad AI exclusions, arguing that blanket restrictions risk creating unintended coverage gaps. “When it comes to AI, we believe it will be good for our country and the global economy.

However, we need to deploy the technology responsibly to avoid potentially severe, systemic or irreversible risks,” CEO John Doyle said during Marsh's third-quarter earnings call on Thursday. “We're particularly concerned about the potential for heightened cyber exposure from AI.” The New York-listed company’s shares were up 5.5% at $188.18 at 12.25 p.m. ET (1625 GMT) after it reported Q3 adjusted earnings per share of $2.03, up 10% year on year and ahead of analyst expectations. Revenue rose 8% to $6.9 billion, while underlying revenue increased 5%.

When asked about the debate around AI exclusions, Doyle said AI was already embedded in most companies' workflows in lots of different ways, and in their processes and products. “A broad-based exclusion could create real confusion and uncertainty,” he said, adding that Marsh was trying to have a “thoughtful discussion" about how AI could change exposures and how those risks might aggregate across the market. Nick Studer, CEO of Marsh Risk, said many AI-related risks were extensions of existing perils and should not automatically be excluded from coverage. “It will be impossible to unbundle it from the causal pathways to whatever outcome may be," Studer said.

“That's really where we were focused on not excluding things which we think would lead to uncertainty, to contention, to gaps in cover.” Studer added that Marsh viewed some AI risks as requiring existing insurance products to evolve, while it remained unclear whether an entirely new category of AI-specific risks would ultimately emerge. Marsh's stance has found support in the broking space, with Aon and WTW also voicing concerns about the use of sweeping AI exclusions and the potential impact on client coverage. Separately, Marsh said commercial insurance pricing continued to soften during Q3, with rates falling 5% globally according to its Global Insurance Market Index.

Property rates declined 10% and cyber rates fell 4%, while global casualty rates rose 1% and US excess casualty rates increased 13%, reflecting inflationary pressures and loss-cost trends. Doyle said the company expected soft reinsurance market conditions to continue through the January 1 renewal season, supported by abundant traditional and alternative capital. The broker also increased the savings target for its three-year Thrive transformation programme, launched last year, to $600 million from $400 million, citing additional opportunities from technology, automation and business services initiatives.