The round
The Artificial Intelligence Underwriting Company, known as AIUC, has raised a $40 million Series A led by Ribbit Capital, with First Harmonic participating, the company said on Tuesday. The round takes its total funding to $55 million, following a $15 million seed round.
AIUC was founded by Rune Kvist, an early Anthropic employee, and Rajiv Dattani, the former chief operating officer of the AI evaluation organisation METR, TechCrunch reported. The seed round came from Nat Friedman’s NFDG fund, Emergence, Terrain and Anthropic co-founder Ben Mann, according to TechCrunch.
What it sells
AIUC’s product is a standard for AI agents, AIUC-1, plus the audits and insurance built on it. TechCrunch compared it to SOC 2, the certification software companies use to show customers their security controls. AIUC says the standard was built with more than 250 security and risk leaders and tests “5,000 risk-and-attack combinations unique to each type of business”.
In practice, TechCrunch reported, an agent is run through about 5,000 tests covering jailbreaks, hallucinations and data leaks. AI agents conduct the tests and analyse the data, humans verify the final audit, and the result is a report of roughly 100 pages setting out where the agent performs safely and where there are concerns.
AIUC lists Cursor, ElevenLabs, Harvey, KPMG, Lovable, UiPath and Fin among the companies it works with. On its website, it says ElevenLabs secured the first AI agent insurance backed by AIUC-1 certification, and that KPMG was the first Big Four firm to achieve the certification.
From agents to frontier models
The company said the new capital “will help extend our audits, standards and insurance from agents to frontier models to create a strong ecosystem for frontier oversight”.
The timing puts AIUC next to a live policy debate. On the same day, OpenAI said it supports the part of the House FRONTIER Act that would require independent safety verification at frontier labs, TechCrunch reported. AIUC has not said it intends to act as a verifier under that bill, and the bill has not passed.
Why tie audits to insurance
The founders’ argument, as TechCrunch reported it, is that large organisations are held back less by what AI models can do than by their inability to guarantee what an agent will and will not do. An audit sets a benchmark; an insurance policy puts money behind it.
What to watch
AIUC has not named a frontier lab that has submitted a model to its audits, and it has not disclosed how much insurance cover it has written. Both would show whether the move from agents to frontier models is more than a plan.
