A new RAND Corporation report documents that some insurers, including W.R. Berkley, are explicitly excluding AI-related harms from D&O, E&O, and fiduciary liability policies, and that standardized forms from Verisk/ISO now include optional exclusions for generative AI harms. The report finds that existing insurance lines do not neatly cover emerging AI-related incidents such as deepfakes, privacy violations, IP disputes, discriminatory decisions, and agentic failures. For a financial institution, the structural implication is that enterprise AI adoption faces a new, quantifiable risk: if the AI industry becomes partly uninsurable, corporate customers may delay or scale back AI projects to meet fiduciary obligations, slowing recognized enterprise revenue and casting doubt on ROI-driven acceleration claims.
301 other tracked developments touching the same claims.