AI insurance carriers compared: who covers AI liability

Segment

Territory

Product posture

Sort by

Carrier	Product	Posture	Segments	Territory	Capacity	Coverage	Frameworks referenced
AIUCArtificial Intelligence Underwriting Company (AIUC)	AIUC-1, a standard for AI agentsLaunched from stealth July 2025	Standard	Enterprise	US	Not applicable. AIUC is an underwriting company and standards body, and publishes no capacity figure.	AIUC-1 sets 51 requirements and 130 controls across six pillars. A certificate can stand behind a policy, as with the first AIUC-1-backed AI agent policy for ElevenLabs in February 2026, placed through Lloyd's of London.	AIUC-1NIST AI RMFMITRE ATLAS
Exclusions. Not applicable. Distribution. Certification through an accredited auditor route; Schellman is the first accredited auditor. Launched from stealth in July 2025 with a USD 15 million seed round led by Nat Friedman at NFDG. AIUC-1 is a standard of 51 requirements and 130 controls across six pillars: data and privacy, security, safety, reliability, accountability and societal risks. ElevenLabs secured the first AIUC-1-backed AI agent insurance policy in February 2026, placed through Lloyd's of London. No carrier or reinsurer is named at source for that policy, and none is named here. AIUC is an underwriting company and standards body; it is listed because its standard is what that policy rests on. Verified 2026-08-17 at https://aiuc.com/
ArmillaArmilla Insurance Services, Coverholder at Lloyd's	Affirmative AI Liability Insurance; Standalone AI Liability Policy; Vanguard AI with ChaucerCoverholder at Lloyd's since 2024; launched with Chaucer 30 April 2025; Vanguard AI 10 February 2026	Specialty indemnity	Mid-marketEnterprise	USUK	Up to USD 25 million per organisation.	AI regulatory violations, including defence costs and insurable fines under the EU AI Act and the Colorado AI Act. Full policy terms are not published.	Not published
Exclusions. Not published; confirm with a broker. Distribution. Underwritten by certain underwriters at Lloyd's, including Chaucer. Reachable by European buyers through wholesale brokers. Armilla Insurance Services is a Coverholder at Lloyd's, the first dedicated exclusively to AI liability. Recognised partners named at armilla.ai: Chaucer, AXIS Capital, Convex, Swiss Re, Greenlight Re. Vanguard AI is a coordinated cyber, technology and AI liability structure launched with Chaucer on 10 February 2026. Not a European-domiciled product. The USD 25 million is the policy limit; trade press has reported it as a funding round, and there was no such round. Verified 2026-08-17 at https://www.armilla.ai/ai-insurance
CorgiCorgi Insurance Services, Inc., a licensed insurance producer (CA licence 6012791) acting as programme administrator, not the insurer	AI liability for technology companiesLaunch date not stated	Specialty indemnity	SMEMid-market	US	Not disclosed.	AI liability for technology companies, administered as a programme. Read the producer point first: it changes who carries the risk on your policy.	Not published
Exclusions. Not stated here. Distribution. Programme administrator and licensed producer; the insurer is a separate party. Corgi announces a USD 108 million fundraise on its own site. An ARR figure, an investor list, a regulatory approval date and a July 2025 launch date carried here earlier are not published by Corgi and have been removed, as has a detailed coverage description that could not be read at source. Verified 2026-08-17 at https://www.corgi.insure/ai
CounterpartCounterpart	Affirmative AI Coverage and a technology E&O insuring agreementAnnounced 24 November 2025	Policy extension	Mid-market	US	Not separately disclosed.	Affirmative AI Coverage plus a technology errors and omissions insuring agreement, across miscellaneous professional liability and allied health. Management liability and D&O are outside it.	Not published
Exclusions. Not published. Distribution. Through Counterpart's miscellaneous professional liability and allied health products. Verification caveat, weaker than the rest of this table: Counterpart's own site does not describe the AI coverage, so this rests on its announcement. No carrier panel and no revenue threshold is published, and none is stated here. Verified 2026-08-17 at https://knowledge.yourcounterpart.com/
HSBHartford Steam Boiler (Munich Re subsidiary)	AI Liability Insurance for small and mid-sized businessesIntroduced 18 March 2026	Policy extension	SME	US	Standard limits USD 25,000 or USD 50,000, with a USD 500 deductible; higher limits available.	Bodily injury, property damage, and personal and advertising injury arising from the insured's use of AI.	Not published
Exclusions. AI model errors, biased outputs and AI-facilitated data loss, which HSB's release names as outside the cover. Distribution. Added to the business policies of carriers that partner with HSB; not sold direct. Subject to insurance regulatory approval. An earlier version of this table said no limit was disclosed; HSB publishes the limits above. An SMB adoption statistic carried here earlier is not in the release and has been removed. Verified 2026-08-17 at https://www.munichre.com/hsb/en/press-and-publications/press-releases/2026/2026-03-18-introducing-ai-liability-insurance-for-small-businesses.html
Munich Re aiSureMunich Re, with Mosaic Insurance	aiSure AI performance insuranceMosaic partnership announced 26 February 2026	Performance guarantee	Mid-marketEnterprise	EUGlobal	Up to EUR/USD/CAD 15 million in initial capacity, through Mosaic Insurance.	Performance shortfalls of an AI system, settled on measurable performance data against an agreed specification. Written for AI developers and vendors worldwide.	Not published
Exclusions. Not published. Munich Re requires its own technical due diligence before cover is written. Distribution. Mosaic Insurance underwrites and markets it through its cyber specialists; Munich Re supplies the technical foundation. The stated audience is AI developers and vendors worldwide, which is narrower than every enterprise deploying an agent. Munich Re does not describe aiSure as an index-triggered product anywhere it publishes; it settles on measurable performance data and requires Munich Re technical due diligence. Verified 2026-08-17 at https://www.munichre.com/en/solutions/for-industry-clients/insure-ai.html and https://www.mosaicinsurance.com/resources/press-releases/
TestudoTestudo UK Ltd is an Appointed Representative of Pro MGA Solutions Ltd, FCA FRN 1017533	AI liability insurance for United States enterprisesLaunched 21 January 2026	Specialty indemnity	Enterprise	US	USD 9.25 million per insured, from 26 February 2026.	Detailed coverage terms are not stated here. Ask Testudo or a broker.	Not published
Exclusions. Not published. Distribution. Lloyd's coverholder with Apollo capacity, joined by Atrium and QBE on 26 February 2026. Launched 21 January 2026 as a Lloyd's coverholder with Apollo capacity. Capacity expansion announced 26 February 2026: Atrium and QBE joined Apollo, at USD 9.25 million per insured. Distributed in the United States, so for a European reader this is a market signal rather than a product to buy. Verified 2026-08-17 at https://www.testudo.co/insights/ and https://atrium-uw.com/
VouchVouch	AI insuranceLaunch date not stated	Policy extension	SMEMid-market	US	Not disclosed.	Affirmative AI cover naming hallucinations, algorithmic bias and certain intellectual property disputes.	Not published
Exclusions. Not published. Distribution. United States. No client count and no acquisition is published, and neither is stated here. Verified 2026-08-17 at https://www.vouch.us/coverages/ai-insurance

Inclusion is editorial. No carrier pays for placement, ranking or visibility. The same records, with source URLs and verification dates, are published at carriers.json, with a notVerified block naming what this desk removed and why. If you cite the matrix, cite the carrier source named on the record alongside it.

Regulatory note, verified 17 August 2026. The AI Omnibus entered into force on 27 July 2026. Annex III high-risk obligations now apply from 2 December 2027 and Annex I obligations from 2 August 2028; the transparency and governance rules applied from 2 August 2026. The nearest deadline for a European buyer is 9 December 2026, the transposition deadline for the revised Product Liability Directive, which the Omnibus did not touch. The deferral changed nothing in this matrix: specialist capacity kept expanding and submission requirements did not relax. Source: European Commission. For what it means for cover, read the Omnibus in force and AI agent insurance.

The generalists. Reported, and checked.

As specialist products are built, the mainstream market is reported to be moving toward exclusions and sublimits. A buyer cannot assume an existing general liability, professional indemnity or management liability policy responds to an AI loss. What each carrier has published, and what rests only on reporting, is set out below.

ISO generative AI exclusions. Two of the forms have been read from specimens that carry ISO's own copyright line. CG 40 48 01 26, Exclusion, Generative Artificial Intelligence (Coverage B only), removes personal and advertising injury arising out of generative AI from the commercial general liability coverage part. CG 35 08 01 26, Exclusion, Generative Artificial Intelligence, removes bodily injury and property damage arising out of generative AI from the products and completed operations coverage part. A specimen held by a law publisher is one step short of a filing record. CG 40 47, reported as the broad form across Coverage A and Coverage B, has not been read, and its title and edition are not stated here. Ask your broker to confirm the form number, edition and title on your own schedule before relying on any analysis of it, this page included. A search engine will return confident answers about these forms; they are assembled partly from pages this network published, and our own work reflected back is not verification.

Carrier	Lines	Position	Basis
QBE	Cyber, North America	Writes affirmatively. QBE North America publishes two AI-focused cyber coverages: an AI regulatory compliance endorsement, and LLMjacking protection for the service fees and retraining costs that follow credential theft against a cloud-hosted model. It belongs with the writers.	Published at qbe.com
Beazley	Cyber, professional liability	No AI sublimit wording is published at beazley.com. Percentage caps circulated in the London trade press are not confirmed by the carrier and are not stated here.	Checked at beazley.com
W.R. Berkley	General liability, professional lines	Trade press reports an absolute AI exclusion. Nothing was found at berkley.com, and the circulating form number and wording are unconfirmed, so neither is stated here.	Reported only
AIG	Management liability, D&O, professional lines	Reported to have filed AI exclusion language with US state regulators. Not confirmed at the carrier's own domain.	Reported only
Great American	Management liability	Reported to have sought clearance for AI exclusions in management liability lines. Not confirmed at the carrier's own domain.	Reported only
Chubb	Commercial lines, management liability	Reported position on systemic or correlated AI events. Not confirmed at the carrier's own domain.	Reported only

How the matrix is kept.

What gets in

A carrier or programme writing affirmative AI liability cover as of the review date: a standalone AI liability product, or explicit affirmative AI cover within an existing line, with a publicly documented product. Smaller limits are included where they mark a market move, as with HSB. Carriers writing AI risk only as a reinsurer, with no direct programme, are noted and left out of the table. The full inclusion criteria are published.

What counts as a source

Every fact is read at the named party's own domain: a product page, a press release on its site, its own documentation. Trade press, law-firm briefings and search summaries are used as leads and never as verification. Capacity figures are what the carrier publishes, and are not guaranteed terms. Where a point could not be read at source, it is marked as not published.

How it is updated

The matrix is reviewed quarterly and on any published carrier change: a launch, a capacity change, a withdrawal or a material regulatory development. The editors do not contact carriers for data before publication. Carriers and brokers with an update can send evidence through the form on the waitlist page; every submission is reviewed.

Four questions the matrix exists to answer.

Which carriers write AI agent liability insurance for European buyers?

As at 17 August 2026, four sources of dedicated AI agent cover reach a European buyer: Munich Re aiSure, underwritten and marketed by Mosaic Insurance, with up to EUR/USD/CAD 15 million of initial capacity, announced 26 February 2026; Armilla, a Coverholder at Lloyd's, writing a Standalone AI Liability Policy of up to USD 25 million per organisation, underwritten by certain underwriters at Lloyd's; Counterpart, on miscellaneous professional liability and allied health, announced 24 November 2025; and bespoke Lloyd's programmes placed through a wholesale broker.

HSB and Testudo write AI cover but are distributed in the United States, so for a European reader they are market signals rather than products to buy. No European-domiciled primary carrier sells an off-the-shelf AI agent liability policy.

What is the largest AI liability limit publicly available?

Armilla's, at up to USD 25 million per organisation, on a Standalone AI Liability Policy underwritten by certain underwriters at Lloyd's.

Munich Re aiSure publishes up to EUR/USD/CAD 15 million of initial capacity through Mosaic. Testudo publishes USD 9.25 million per insured following its 26 February 2026 capacity expansion, when Atrium and QBE joined Apollo. Larger totals can be built by layering carriers through a Lloyd's wholesale broker, but no syndicate publishes an AI aggregate, so any layered figure is arithmetic on published single-carrier limits rather than a stated capacity.

Is Munich Re aiSure an index-triggered product?

No. Munich Re does not describe it that way anywhere it publishes. It settles on measurable performance data against an agreed specification, and it requires Munich Re technical due diligence, neither of which an index trigger involves.

The distinction matters at placement. An index product pays on a published index without asking about the actual loss; this one needs the performance specification defined and agreed up front. This site used the wrong label across many pages until it was corrected in August 2026, and the corrections log records where.

How is the matrix verified, and how often is it updated?

Every entry is read on the named party's own domain, and each record carries its source URL and the date it was checked.

A trade article, a law-firm briefing and a search-engine summary are not treated as verification, because search summaries of this market already draw on pages published by this network. Figures that could not be verified at source are absent rather than hedged. Carrier facts were last verified on 17 August 2026, the matrix was read again against them on 1 October 2026, and it is reviewed quarterly and on any published carrier change.

Sources and references

Market map: The 2026 AI liability insurance market map, Agent Insured, 23 April 2026.
HSB: AI Liability Insurance for small businesses, 18 March 2026, munichre.com/hsb.
Armilla: Standalone AI Liability Policy, limits up to USD 25 million per organisation, armilla.ai/ai-insurance; launch with Chaucer, 30 April 2025, armilla.ai.
Testudo: launch, 21 January 2026, and capacity expansion, 26 February 2026, testudo.co; atrium-uw.com.
AIUC and AIUC-1: aiuc.com.
Counterpart: affirmative AI coverage announcement, 24 November 2025, knowledge.yourcounterpart.com.
Corgi: corgi.insure/ai, which publishes Corgi Insurance Services, Inc. as a licensed insurance producer and programme administrator rather than the insurer.
ISO forms: specimens of CG 40 48 01 26 and CG 35 08 01 26, assets.alm.com. CG 40 47 is reported in trade and law-firm analysis and has not been read.
W.R. Berkley, AIG, Great American, Chubb: trade reporting only, not confirmed at the carriers' own domains.
QBE North America AI-focused cyber coverages: qbe.com.
EIOPA: survey on generative artificial intelligence, 2 February 2026, 347 undertakings across 25 countries, eiopa.europa.eu. An EIOPA AI consultation with an April 2026 closing date appeared in an earlier version of this table; none could be found at eiopa.europa.eu and the reference was removed.
Munich Re aiSure: munichre.com; Mosaic Insurance partnership announcement, 26 February 2026, mosaicinsurance.com.
Vouch: vouch.us/coverages/ai-insurance.

Who writes AI liability cover. Carrier by carrier, with sources.