The continuing growth and integration of Artificial Intelligence into business operations—from automated customer service agents to advanced decision-making systems—promises unprecedented efficiency and cost reduction. However, as these tools become ubiquitous, the insurance marketplace is signaling a critical pivot. The safety net businesses have traditionally relied upon is being redesigned in real-time as insurers aim to limit their exposure AI liability.
This shift requires executives to look beyond the hype of AI and critically assess the architecture of their risk transfer programs.
The Market Signal: Insurers Move to Cap Exposure
Three prominent insurance leaders—AIG, Great American, and WR Berkley—have recently submitted requests for regulatory approval to limit their liability for claims arising from artificial intelligence systems, including chatbots and other automated services.
This coordinated approach signals a broader industry trend where traditional insurers are adjusting underwriting practices to keep pace with rapidly evolving technology. The primary driver behind this move is the growing concern over the potential for multibillion-dollar claims tied to AI-related errors or harm.
While AI technologies offer operational benefits, they introduce new, complex risks that historical actuarial data cannot easily predict. By seeking to limit AI liability, these carriers are proactively addressing emerging exposures to ensure their policies remain viable in the face of potentially catastrophic AI-related losses.
Essentially, limiting liability allows them to manage the uncertainty inherent in AI systems while continuing to offer some form of coverage.
The Implication: A Massive Transfer of Risk
For policyholders, this development is not merely administrative; it fundamentally alters the risk landscape. If insurers succeed in imposing AI liability caps, the responsibility for losses will shift more heavily onto the businesses that deploy these AI systems.
This development may influence:
- The types of coverage available to your organization.
- The specific terms and exclusions written into your policies.
- The premiums charged for AI-exposed risks.
While regulatory approval is still pending, this proposal marks a significant step in the insurance industry’s adaptation to digital innovation.
It underscores a new reality: businesses using AI must reassess their risk management strategies to address potential coverage gaps.
Strategic Response: Re-Engineer Your Insurance Program Around AI
As insurers file for broad generative-AI exclusions or move to exclude AI-related Errors & Omissions (E&O) claims altogether, relying on “silent AI” coverage (where coverage is assumed because it isn’t explicitly excluded) is no longer a viable strategy.
To protect your business, you must actively design for insurability. Here is the Metropolitan Risk roadmap for navigating this hardening market:
1. Conduct an AI-Specific Coverage Gap Analysis
You cannot manage what you do not measure. Work with your broker and coverage counsel to review your current portfolio, specifically:
- General Liability (GL) / Products Liability
- Tech E&O / Professional Liability
- Cyber Liability
- Directors & Officers (D&O) / Employment Practices Liability (EPL): Critical where AI is used in HR hiring algorithms or board-level strategic decisions.
What to flag:
- Any “technology error” exclusions that could be interpreted broadly.
- Endorsements referencing generative AI, algorithms, automated decision-making, or “intelligent agents.”
- Sublimits that are dangerously low for AI-related regulatory fines or penalties.
2. Negotiate Targeted, Narrower Exclusions
Insurers will attempt to push broad language. You must push back on wording that excludes any claim “arising out of AI.”
- Aim for: Exclusions that are limited only to specific, high-severity scenarios rather than blanket denials.
- Clarify Coverage: Seek endorsements that affirmatively clarify what is covered, rather than simply removing coverage.
3. Document Controls to Win Underwriter Confidence
In a hard market, the “best-in-class” risks get the best terms. You must prove to underwriters that you are not a risk to them.
- Provide Evidence: Submit your AI usage policy, inventory of AI tools, and governance frameworks.
- Human-in-the-loop: Demonstrate that human oversight and data controls are active.
- The Payoff: Strong controls equate to a better chance of avoiding aggressive exclusions and securing higher limits.
The initiative by AIG and others highlights critical questions about accountability and risk allocation in the AI space. Clear communication with insurers and proactive risk management will be essential as AI technologies continue to expand across sectors.
Companies operating with AI must remain vigilant, remaining aware of evolving insurance practices. The era of “set it and forget it” insurance renewals is over for AI-enabled businesses.
If you are unsure if your policies fully cover your exposures or contain any AI-related coverage gaps, connect with a risk advisor for an analysis of your current exposures and exclusions.
