Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor!

Generative AI Presents: A case for evolution rather than revolution

By Gregory Eskins | Head, Global Cyber Insurance Center & Jaymin Kim | Managing Director, Emerging Technologies

As generative AI use evolves, business leaders are asking how this technology will reshape their risk profile and whether their current insurance program is equipped to respond. For many within our industry, the instinct is to develop a new insurance product that exclusively covers the use of generative AI.

We believe it may be more effective to first consider whether generative AI has created a new kind of risk, one never before contemplated by existing insurance solutions. Our perspective is that the market may already have many of the necessary tools to address generative AI-related exposures — and that those exposures are, more often than not, extensions of risks that businesses already understand and business insurance products may already cover.

AI isn’t new — and neither is its risk

The pace of generative AI’s evolution can make it feel like we’re in uncharted territory. However, businesses have been using various forms of AI for decades — from predictive modeling to computer vision — and some insurance policies have encompassed the exposures these technologies create.

Generative AI to date has introduced new capabilities, but not a new class of risk. It can amplify known exposures — including data privacy, intellectual property, technology errors, and fraud — and deliver them through new channels, sometimes at increased pace and scale. Real-world claims are already bearing that out: deepfake-enabled wire fraud financial loss, which can be covered under crime insurance; a chatbot-related misinformation leading to E&O claims; IP lawsuits from unlicensed training data triggering media liability coverage. These cases are evidence that the market already has the tools to respond.

What, then, underpins the call for AI-specific endorsements or new standalone policies? In part, the instinct to compartmentalize new technology into a new insurance product may feel innovative; however, that has the potential to lead to more problems than it solves. Creating new products that overlap with existing coverage can add cost, increase uncertainty, and risks leaving clients stuck asking: Which policy applies? What qualifies as an AI-related event? Is silence the same as exclusion?

How do we manage evolving risks without breaking what already works?

Given that generative AI has presented nuances within existing lines of insurance, any amplification of corresponding risks should be underwritten and priced within existing lines where possible. Thoughtful, additional underwriting questions can help inform underwriting to perils associated with generative AI risks in the absence of sufficient loss data. No doubt that some tweaks may be needed to policy language over time, but we should start by evolving existing products, as opposed to creating entirely new ones.

It’s incumbent upon our industry to take the challenge head on — to resist the urge to reactively start adding AI exclusions on existing products and hastily introduce niche solutions for each major technological advance. Instead, let us focus on the real challenge: managing emerging risks without undermining the structures that may already protect against them.

An elegant solution starts with respecting the spirit and intent woven into well-designed coverage. AI is not a peril; nor is it an impact. It is simply a tool that may be associated with a causal link of familiar types of loss and impacts. Silence on AI and other technologies — often misunderstood as a gap — is frequently a feature. It can allow policies to remain adaptable as technology evolves, without limiting protection or forcing policyholders to enumerate every tool they might use.

Of course, new product innovations are useful and encouraged where there are no established solutions, i.e. no specific policy exists, for example, performance guarantees. There is always room for new product innovations that do not conflict with existing ones.

Our industry has asked if there is a need to create new insurance policies in reaction to almost every new wave of technology, including cloud, autonomous vehicles, and non-fungible tokens. Generative AI will certainly not be the last novel technology that elicits the question, or reaction.

Strong policies are built to reflect the spirit of the contract — to cover impacts / outcomes, not tools or technologies that may have been associated with such impacts / outcomes. This is one reason why replacing thoughtful silence with reactive specificity can backfire: It may create ambiguity, slow claims, and can shift the burden to policyholders, asking them to map every technological innovation to an insurance product that may already exist.

Getting back to the basics of smart risk protection

Businesses don’t want to buy a new policy every time a new technology hits the market. And they shouldn’t have to. The market already has a robust framework of solutions that may respond to perils associated with AI risk. What’s needed now is not a revolution, rather an intentional refinement of strong controls, thoughtful underwriting, and clear engagement between businesses, brokers, and insurers. When those fundamentals are in place, coverage can be more effective rather than more complicated.

  

The information contained in this article, including but not limited to any recommendations, analysis or advice contained herein (the “Marsh Analysis”) provides only a general overview of subjects covered, is not intended to be taken as advice regarding any individual situation, and should not be relied upon as such. The information contained herein is based on sources we believe reliable, but we make no warranty, representation, or guarantee regarding the information herein or the suitability of these suggestions or information for any particular purpose. Marsh shall have no obligation to update the Marsh Analysis and shall have no liability to you or any other party arising out of this publication, or any matter or information contained herein. Marsh makes no representation or warranty concerning the application of policy wording or the financial condition or solvency of insurers or reinsurers. Marsh makes no assurances regarding the availability, cost, or terms of insurance coverage. Insureds should consult their insurance and legal advisors regarding specific coverage issues. All insurance coverage is subject to the terms, conditions, and exclusions of the applicable individual policies. Marsh cannot provide any assurance that insurance can be obtained for any particular client or for any particular risk. It cannot be assumed that every acceptable risk transfer procedure is contained herein or that unusual or abnormal circumstances may not warrant or require further or additional risk transfer policies and/or procedures. The use of any of the information or suggestions described herein does not amend, modify, or supplement any insurance policy. Consult the actual policy or your agent for details about your coverage.