AIUC certifies and insures AI agents.
San Francisco, CA
Joined January 2025
- Tweets36
- Following14
- Followers1.2K
- Likes57
Pinned Tweet
We’re hiring across all roles in San Francisco.
Why now: aiuc.com/updates/series-a-an…
The Artificial Intelligence Underwriting Company has raised $55M to audit and insure frontier AI models.
Insurance produces audits the public can trust and understand. The mechanism is simple: insurers need audits to price the risk. They pick auditors they believe will help them price the risk most accurately. Insurers then turn the audits into a price signal that cuts through political debates. Insurers are aligned with the public: if they get the audits wrong, they go out of business. That’s different from today, where labs pick their own auditors, creating public distrust.
Insurance only works if we scale up auditing, and we are building the infrastructure to do that. Millions of agents are being deployed; they leave terabytes of traces and will create thousands of incidents across dozens of risk categories. To scale auditing, we need reliable agents to surface evidence traces for human review.
History guides us. Take cars: auto insurers funded crash tests to price risks; crash tests led us to airbags, lowering the risk. Take electricity: home insurers funded lightbulb tests to price fire risks; today, every lightbulb carries a safety mark. Nuclear safety works similarly.
We’ve been learning by doing.. In the last 6 months, we’ve audited and insured frontier agent builders like Cursor, ElevenLabs, Lovable and Harvey. For ElevenLabs to get the world’s first agent insurance coverage, a Lloyd’s of London insurer required quarterly technical audits by an auditor of the insurer’s choice. We’ve built infrastructure for quarterly technical audits of any agent, working alongside firms like Gray Swan (technical evaluations) and Schellman (audit).
In the coming weeks, we’ll share our vision for how an insurance ecosystem can work for the frontier, and open source the world’s first actuarial model for frontier AI catastrophic risk.
We’re hiring across all roles in San Francisco. Come do your life’s work with us.
Rune & Rajiv
KPMG has a 150-year legacy built on trust. Their agentic platform, aIQ Capture, is now AIUC-1 certified - a first for the Big Four. Aisha Tahirkheli, Trusted AI Leader @KPMG_US, and @RajivDattani discussed the need for AI agent assurance, certifying before regulation matures, and more.
Artificial Intelligence Underwriting Company retweeted
The end state is correct: government sets verifier incentives, forces standard updates in line with capabilities, creates transparency into RSI.
However -- the path is the wrong way around: the ecosystem will rise much faster than government can get started. Private verifiers & labs can experiment now to build the right oversight techniques
NEW: Today ARI released a blueprint for federal AI governance, including three pillars that promote safe frontier AI development:
✅ Standards set by the government
✅ Independent assurance they are met
✅ Transparency into frontier AI development
ari.us/frontier-framework
Artificial Intelligence Underwriting Company retweeted
Spoke to @TheEconomist about autonomous hacking.
They said it best: autonomous hacking is here, governments are not ready.
Today, no one is obviously liable when OpenAI's models hack Hugging Face. Liability for cyber attacks requires *intentional* hacking. OpenAI did not hack Hugging Face intentionally. Like many other places, cyber law can't make sense of autonomous agents.
We're in for a confusing time as swarms of agents scour the internet for vulnerabilities over the next 12 months.
Artificial Intelligence Underwriting Company retweeted
AIUC-1 is an independent security standard for AI agents, shaped by 200+ CISOs and security leaders.
@aiunderwriting evaluates AI agents across six areas: data and privacy, security, safety, reliability, accountability, and society.
Harvey is the first legal AI company to achieve AIUC-1 certification.
Artificial Intelligence Underwriting Company retweeted
More than 100,000 lawyers use Harvey, now AIUC-1 certified. Brooks Evans, CISO @Harvey, and I discussed real-world implications for sensitive data protection, source attribution, and more.
Artificial Intelligence Underwriting Company retweeted
@Lovable is now AIUC-1 certified
Certification required the platform to undergo:
> Technical testing for risks like insecure code generation & credential leakage
> Independent auditing by @Schellman on effective controls across all mandatory AIUC-1 requirements
Congrats to @antonosika, @fabianhedin, Igor and the @Lovable team for the commitment to building and deploying secure agents, and earning the AIUC-1 trustmark
Artificial Intelligence Underwriting Company retweeted
Our new report “Underwriting the AI Agent Economy”, covered in the FT today, contains the blueprint for a full private governance stack to drive AI adoption.
Artificial Intelligence Underwriting Company retweeted
Some interesting ideas in there - glad people are looking at insurance markets, seems like an important bit neglected angle in AI governance. Stellar author list too! underwriting-agents.com/
Technical controls alone won’t contain AI security risk.
Shadow AI, blind trust in AI outputs, sensitive data leakage.
These human-introduced risks can only be managed by combining technical controls with a strong culture of AI security.
We asked Dr. Keri Pearlson (@MITSloan) how to engineer a culture of AI security and combined her research with insights from 50+ security leaders across the AIUC-1 Consortium:
> Behaviors they want to see
> Values, attitudes and beliefs that produce them
> Practices that make them stick
Together, we identified 5 practices security leaders can adopt now:
1. Model: Culture follows what leaders do, not what they mandate
2. Reward: Make good practice visible so it spreads faster, peer to peer
3. Deputize: Train local champions to carry security-first habits into every team
4. Expect: Tie secure AI use to everyday norms and formal OKRs
5. Pave: Make the secure path the default path
Read the full white paper aiuc-1.com/research/whitepap…
Artificial Intelligence Underwriting Company retweeted
Today we’re releasing a white paper with AIUC (@aiunderwriting) on the risks unique to coding agents, why they need their own standard, and what Lovable does today to meet the bar.
AIUC-1 is the standard for AI agent security, safety, and reliability. Lovable will be the first in our category to achieve AIUC-1 certification this summer.
Download the report: lovable.dev/blog
Artificial Intelligence Underwriting Company retweeted
Talked at @FundingCommons about how we measure the risk in literally any AI system, with @aiunderwriting
Here's a summary:
Artificial Intelligence Underwriting Company retweeted
Why ElevenLabs chose to insure their ai voice agents with AIUC (@aiunderwriting):
" We had a major announcement that we're going to be the first company to insure voice agents.
If you're an enterprise, you can actually pay for insurance & if it goes out of bounds, something goes wrong, then you can make sure that you're covered."
What's your hottest take on ai agents?
"For most businesses, they're actually really risky & so they can go off script, they can say really bad things.
You're maybe giving them tool use.
And they're able to do refunds on your behalf.
And so that's why we're really excited to announce the first insurable voice agents, where we've actually defined standard & a pathway around guardrails & audits so that you as a company, you can get an insurance product, which covers you to enable you to go to production at scale."
Luke Harries (@lukeharries) Head of Growth @elevenlabs
Rune Kvist (@RuneKvist), CEO of AIUC, was the first product & GTM hire at Anthropic. Now he’s building the “confidence infrastructure” for the AGI era — combining standards (AIUC-1), audits, & insurance to de-risk AI deployments.
AI insurance is here.
"We are about to lose control of what's being deployed in front of us..
when you let the agents chat on Moltbook they write so much more content than a human can even read & they start to coordinate in ways we can't even keep up with
..everyone is going to have nostalgia for 2025."
As a result, ElevenLabs just secured the first-ever AI insurance policy for ElevenAgents, & is now certified against AIUC-1's standard
TLDR: AI agents have reached commercial readiness, the vibe era is over, & they're being deployed across banks, hospitals, governments — real businesses.
Rune Kvist (@RuneKvist), CEO of AIUC (@aiunderwriting)
2025 was the year of enterprise vibe adoption.
> Leadership wanted to move fast.
> Vendors promised transformation.
> Security teams were handed a brand new category of risk with no playbook.
2026 marks the end of vibe adoption.
With it comes 3 challenges. We lay these out in our latest whitepaper.
Co-authored by @stai_research led by @sanmikoyejo, and 35+ CISOs, CTOs, and AI executives from the AIUC-1 consortium, providing the latest thinking on 2026 security gaps.
What will it take to deploy AI with confidence in 2026?
We share the win conditions identified by 35+ security leaders - CISOs, CTOs, and AI executives from the AIUC-1 consortium.
Read the full white paper aiuc-1.com/research/whitepap…
Thank you to our co-authors @sanmikoyejo @philvenables @bradarkin @Cassie_Crossley @NancyZWang @kpearlson @MrDavidCampbell @usinfosec @CyberGoldsmith and more
