As AI becomes central to operations across sectors, a nascent but growing insurance market is emerging to address its unique risks. Here’s a snapshot of key players and how their offerings differ in terms of clarity, capacity, and evolving coverage:
Munich Re – aiSure™ / aiSelf™
Munich Re was the first mover in AI-performance guarantees. Their flagship aiSure™ product provides cover against model errors, miscalibration, drift and hallucinations—protecting against financial and legal damages arising from AI failure. The related aiSelf™ extends this to self-developed AI solutions with tailored underwriting for internal adopters. Coverage includes business interruption, contractual liabilities, and reputational impact, anchored by the insurer’s technical expertise and rigorous model review.
Armilla AI (in partnership with Lloyd’s / Chaucer)
Armilla offers affirmative AI liability insurance underwritten by Lloyd’s syndicates such as Chaucer. Their policy explicitly addresses hallucinations, model degradation and malfunctions. Cover applies when AI performance drops below expected thresholds, triggering financial loss compensation. Underwriters review models for suitability before issuing cover and typically offer limits up to USD 5 million, though actual AI-specific sublimits may start at a modest USD 25,000.
Emerging Providers
- Relm Insurance launched multiple AI-specific policies in early 2025 aimed at innovators and AI-native firms.
- Testudo is developing a dynamic underwriting platform to track AI-related litigation and regulatory shifts for real-time policy issuance.
- Vouch and CoverYourAI are among start-ups offering coverage for AI errors, discrimination, IP issues, and business interruption tied to AI failure.
Sector Collaborations & Add-Ons
- Google Cloud partnered with insurers Beazley, Chubb and Munich Re to embed AI-tailored cyber coverage directly into cloud service offerings—addressing BI disruption, trade secret loss and property damage linked to AI tool failures.
Exclusions and Industry Response
As these affirmative products grow, insurers are also introducing exclusions. For example, Berkley now uses a broad “Absolute AI Exclusion” in D&O and E&O lines, effectively divorcing AI risk from legacy policies. That trend highlights the importance of securing AI-specific solutions rather than relying on retrofits.Hunton+1
What This Means for Businesses and Insurers
For businesses deploying AI
- Standard tech or cyber policies often impose low AI sublimits or ambiguity on coverage.
- Positive protection and transparency now come only from affirmative AI policies like those from Armilla or Munich Re.
- It is vital to assess and push for coverage tied to measurable performance thresholds—not vague exclusionary language.
For carriers and underwriters
- AI risk requires new underwriting models, clear performance metrics, and scalable capacity.
- Partners like Munich Re and Armilla are demonstrating how structured, tech-aligned underwriting can be built at scale.
- The market is expanding, and sustained growth depends on evolving products, credible limits, and trusted model governance.
As AI adoption accelerates, it creates a window for insurers and innovators to shape the coverage landscape thoughtfully. Insurers who offer clear, performance-linked, and sufficiently capitalised policies will not only manage risk but will capture leadership in the AI protection market.