As noted by Dark Reading, insurance companies are increasingly navigating the complexities of artificial intelligence risks as enterprises seek coverage for potential threats from rogue AI agents and automated system failures.
The insurance industry is currently divided regarding AI-related coverage. While some specialty firms offer purpose-built policies or affirmative endorsements to address risks like rogue AI agents and model hallucinations, many large carriers are implementing strict exclusions. Major insurers, including Chubb, Travelers and AIG, have adopted Verisk/ISO endorsement language to limit liability for damages arising from generative AI. This shift leaves many organizations with inadequate protection, as standard cyber or errors and omissions policies often fail to cover these emerging threats. Experts advise that enterprises must conduct thorough policy reviews to identify potential gaps. To qualify for specialized coverage, companies are expected to demonstrate robust security controls, including strict guardrails, human oversight and comprehensive supply chain management. Underwriters are increasingly focused on how businesses monitor AI applications, treating them similarly to critical service accounts. As AI becomes ubiquitous, executives must address shadow AI usage and ensure that infrastructure, data pipelines and third-party contracts are adequately secured against both malicious attacks and non-malicious system failures.
Source: Dark Reading
