Analysis
Artificial Intelligence Underwriting Company (AIUC) raised $40 million in a Series A led by Ribbit Capital, with First Harmonic also participating, TechCrunch reported. The round follows a $15 million seed AIUC raised in July 2025 led by Nat Friedman's NFDG, with Emergence, Terrain and angels including Anthropic co-founder Ben Mann.
Founded in 2025 by Rune Kvist, an early product and go-to-market hire at Anthropic and a board member at the Center for AI Safety, and Rajiv Dattani, a former McKinsey insurance partner who served as chief operating officer at METR -- the nonprofit that runs pre-release capability evaluations for frontier labs -- AIUC sells enterprises an independent, auditable standard for how safe their deployed AI agents actually are. Thiel Fellow Brandon Wang, who previously founded a consumer underwriting business, is the third co-founder and chief technology officer.
“Thiel Fellow Brandon Wang, who previously founded a consumer underwriting business, is the third co-founder and chief technology officer.”
The pitch sits squarely inside this week's industry-wide argument over AI pacing. Dario Amodei's essay calling for a slower frontier and Microsoft's newly published model code of conduct both assume enterprises need external assurance that a deployed agent will behave; AIUC is trying to sell that assurance as a product rather than wait for regulation to mandate it. Early customers reportedly include Cognition, ElevenLabs and Intercom -- three companies that ship customer-facing AI agents at meaningful scale.
The competitive field is thinner than the funding suggests. Lloyd's underwriters have started writing AI-specific liability policies, and Google has partnered with Beazley, Chubb and Munich Re to offer AI coverage bundled with its cloud AI products -- both are insurance-industry incumbents entering from the underwriting side rather than startups building the technical evaluation layer AIUC is selling. That leaves AIUC closer to a standards body with an insurance wrapper than a pure insurer, a positioning that is unproven at any scale.
The risk is adoption speed against a moving regulatory target: if Washington or the EU settles on a mandatory evaluation framework before AIUC's voluntary standard reaches critical mass, enterprises may simply wait for the compliance-mandated version instead of paying for a private one. Founder pedigree from Anthropic and METR buys credibility with technical buyers; it does not yet buy a defensible moat against Google's insurance partnerships or a future government standard that makes this optional layer mandatory.
