Underwriting Superintelligence: Backing Agents you can Sue — Rune Kvist, AIUC

Frontier AI adoption faces a critical bottleneck: the lack of trust and liability frameworks. Rune Kvist, founder of AIUC, addresses this by building "confidence infrastructure" through standardized technical testing and insurance. AIUC1, an agent certification framework, requires quarterly simulations to verify safety, reliability, and security, creating a verifiable audit trail for enterprises. By partnering with established insurers like Lloyd's of London, AIUC transforms abstract risk into quantifiable, underwritten policies, enabling banks and critical infrastructure to deploy autonomous agents with legal and financial recourse. This model shifts from speculative risk to a structured, evidence-based approach, mirroring historical precedents like UL testing for electrical goods. Future efforts will extend this certification model to frontier models and robotics, establishing a neutral, third-party mechanism to bridge the trust gap between AI labs, government regulators, and enterprise users.