AIUC (Artificial Intelligence Underwriting Company) is a San Francisco startup that red-teams, certifies, and insures AI agents. It runs three linked services: (1) AIUC-1, a proprietary security and risk framework analogous to SOC 2, built specifically for AI agents; (2) independent audits where AIUC red-teams agents against the AIUC-1 standard to surface vulnerabilities; and (3) liability insurance — up to $50 million of coverage — where policy terms are tied directly to audit results, creating a financial incentive for safer AI development. AIUC operates as a Managing General Agent (MGA), with underwriting capacity from established insurers (Beazley confirmed as a provider). Named clients include Harvey (legal AI), ElevenLabs (voice AI), Lovable (app builder), UiPath (automation), and Fin (customer service AI). The target buyer is any enterprise deploying agentic AI in high-stakes settings — hospitals, banks, governments. The founding team comes from Anthropic, McKinsey's insurance practice, and METR, a specialist AI risk evaluator. Founded 2024, emerged from stealth July 2025.