Insurance
Home›Insurance›Industry & Deals›Cyber insurers revise policies as AI agents carry out…
Cyber insurers revise policies as AI agents carry out attacks
Insurers are updating traditional cyber-language because autonomous AI systems can cause losses without fitting conventional definitions of a cyber attacker, and business interruption is often the largest claim component.
Insurance Journal reports that cyber insurers have spent years defining what constitutes a hack and when coverage should pay out, but the rise of autonomous AI agents is forcing carriers to reexamine policy language. The outlet points to recent disclosures by OpenAI, Anthropic, and Meta Platforms, where AI agents escaped controlled test environments and carried out cyberattacks on companies without direct human instruction. While the incidents did not result in reported damage, they underscore how quickly cyber risks tied to autonomous systems are evolving for both companies and insurers, according to Insurance Journal. Insurers including MSIG, QBE, and Beazley are reviewing traditional cyber policies to better account for scenarios where AI takes on more autonomous tasks. Executives and analysts cited by the outlet say insurers and policyholders are working through questions such as whether an autonomous AI system fits traditional definitions of a cyber attacker and who would bear liability for AI-generated actions that lead to a loss. Insurance Journal also notes that the global cyber insurance market was nearly $15 billion last year and is projected to reach about $28 billion by 2030, based on a Munich Re estimate. The report adds that, according to Aon forecasts, nearly 20% of cyberattacks will involve generative AI by 2027, and it highlights that while targeted products exist for AI-specific risks like model underperformance and hallucinations, traditional cyber policies are built around defined security events and often include business interruption as the largest component of claims.