AI and Insurance: A gradual but challenging shift
Moody’s analysts expect artificial intelligence to change insurance operations over time, with clear efficiency gains in routine retail lines but tougher questions for casualty underwriters. Automated underwriting, claims automation and predictive analytics can lower cost and speed decisions in personal lines. At the same time, liability tied to AI-driven decisions raises complex allocation questions that the market has not yet settled.
Data centres: the new risk frontier
Investment in hyperscale data centres has surged, creating a concentrated pool of high-value exposure. That concentration creates three insurance challenges: valuation uncertainty for bespoke facilities, limited historical loss data for modeling, and geographic clustering that amplifies catastrophe-type exposures. Physical risks that impair availability – such as cooling failure or a targeted cyberattack on operational control systems – can produce outsized business interruption claims.
AI liability: where will legal responsibility land?
One of the largest open questions is whether legal responsibility for AI errors will sit with developers, platform operators or end users. Regulation, litigation and contractual terms will shape outcomes, but for now casualty markets face ambiguity. Potential outcomes include expanded product liability claims, professional liability suits against deployers, and novel third-party claims for harms caused by autonomous decision-making. The result is difficulty in defining covered causes and in pricing aggregate exposure.
Strategic outlook for insurers
Insurers should treat these trends as strategic risk-management priorities rather than near-term losses alone. Practical steps include:
- Refining exposure models to capture concentration risk from clustered data centres and correlated business interruption scenarios.
- Updating underwriting and policy language to address AI-specific causes of loss and to clarify responsibilities across the value chain.
- Investing in cyber-physical loss modelling and risk engineering focused on operational controls such as cooling and OT networks.
- Pursuing reinsurance structures and limits that reflect new accumulation patterns and tail risks.
- Monitoring regulatory developments and litigation trends to adjust reserving and capital planning.
Moody’s framing underscores a simple reality: technological change will raise opportunities and contested liabilities. Insurers that combine sharper modelling, targeted underwriting expertise and proactive contract design will be best placed to price and absorb these evolving risks.
