Future Proof The Authority Stack
01

Enforcement Deadline Ladder

Key dates under Regulation (EU) 2024/1689 and the revised Product Liability Directive. The Digital Omnibus provisional agreement of 7 May 2026 proposes to defer Annex III high-risk obligations. Formal publication in the Official Journal was pending as of 2026-06-13. The original dates remain legally binding until formal adoption.

Omnibus status A provisional political agreement was reached 7 May 2026 to defer Annex III high-risk obligations from 2 August 2026 to 2 December 2027, and Annex I obligations from 2 August 2027 to 2 August 2028. As of June 2026 it has not been adopted or published in the Official Journal, so 2 August 2026 remains legally binding until adoption. Art. 5 prohibitions, Art. 4 literacy, Art. 50 transparency, Art. 53 and 55 GPAI obligations, and the revised Product Liability Directive are not affected by the Omnibus.
ID Date (binding) What applies Omnibus Revised date
DL-001 2024-08-01 Regulation (EU) 2024/1689 enters into force No N/A
DL-002 2025-02-02 Art. 5 prohibited practices enforceable. Art. 4 AI literacy obligation applies. No N/A
DL-003 2025-08-02 Art. 53 and 55 GPAI obligations apply. AI Office mandate and Art. 99 penalty regime operational. No N/A
DL-004 2026-08-02 Original date for Annex III high-risk obligations (Art. 6, 8 to 15, 26, 27, 43, 50, 72, 73). Binding now; deferral to 2027-12-02 proposed 7 May 2026 but not yet adopted. Provisional, OJ pending Yes 2027-12-02
DL-005 2026-08-02 Art. 50 transparency to natural persons. Not deferred by Omnibus. Chatbot deployers must disclose AI interaction. No N/A
DL-006 2026-12-09 Directive (EU) 2024/2853 revised Product Liability Directive transposition deadline. AI software is explicitly a product subject to strict liability. No N/A
DL-007 2027-08-02 Original date for Annex I high-risk obligations (AI in regulated products). Binding then; deferral to 2028-08-02 proposed 7 May 2026 but not yet adopted. Provisional, OJ pending Yes 2028-08-02
02

Obligation Clusters

Seventeen obligation clusters mapped across Regulation (EU) 2024/1689 and Directive (EU) 2024/2853. Each record identifies who is bound, the enforcement date, and the primary evidence an assessor or underwriter needs to confirm compliance.

ID Obligation Who binds Article Date Penalty ceiling
OBL-001 Prohibited practices Both Art. 5 2025-02-02 EUR 35M or 7%
OBL-002 AI literacy Both Art. 4 2025-02-02 No direct penalty in isolation
OBL-003 Risk management system Provider Art. 9 2026-08-02 EUR 15M or 3%
OBL-004 Technical documentation (Annex IV) Provider Art. 11, Annex IV 2026-08-02 EUR 15M or 3%
OBL-005 Transparency to deployers Provider Art. 13 2026-08-02 EUR 15M or 3%
OBL-006 Human oversight design Provider Art. 14 2026-08-02 EUR 15M or 3%
OBL-007 Accuracy, robustness, and cybersecurity Provider Art. 15 2026-08-02 EUR 15M or 3%
OBL-008 Conformity assessment Provider Art. 43 2026-08-02 EUR 15M or 3%
OBL-009 Deployer operational obligations (7 sub-duties including log retention, oversight, incident reporting) Deployer Art. 26 2026-08-02 EUR 15M or 3%
OBL-010 Deployer reclassification as provider on substantial modification or rebranding Deployer Art. 25 2026-08-02 EUR 15M or 3%
OBL-011 Fundamental rights impact assessment Deployer (subset) Art. 27 2026-08-02 EUR 15M or 3%
OBL-012 Transparency to natural persons. Chatbot disclosure, synthetic content marking. Not deferred by Omnibus. Both Art. 50 2026-08-02 EUR 15M or 3%
OBL-013 Post-market monitoring Provider Art. 72 2026-08-02 EUR 15M or 3%
OBL-014 Serious incident reporting. 15-day notification to market surveillance authority. Both Art. 73 2026-08-02 EUR 15M or 3%
OBL-015 GPAI baseline obligations (Annex XI technical docs, Annex XII downstream docs, copyright policy, training data summary) GPAI provider Art. 53 2025-08-02 EUR 15M or 3%
OBL-016 GPAI systemic risk obligations (adversarial testing, risk assessment, AI Office incident reporting) GPAI provider (systemic) Art. 55 2025-08-02 EUR 15M or 3%
OBL-017 Product liability. AI software is a product. Strict liability. Reversed burden of proof. Applies from 9 December 2026. Manufacturers, deployers who modify Directive (EU) 2024/2853 2026-12-09 Strict liability. No cap. Damage categories include data destruction and psychological harm.

OBL-003 to OBL-014 (Annex III obligations) carry an Omnibus provisional date of 2 December 2027 under the 7 May 2026 agreement. As of June 2026 it is not yet adopted or published in the Official Journal, so the 2 August 2026 date remains legally binding until adoption.

03

The Seven Agent Risk Dimensions

Canonical dimension set from Agent Certified methodology v2.0 (published 24 April 2026). Names, weights, and definitions are reproduced verbatim from the source. Total weight: 100. Scoring formula: ((raw score x weight) summed across all dimensions / 1000) x 100.

D1 Trust and Safety Weight 18 (highest)

The measurable prevention of unsafe, unauthorised, or harmful actions by the agent in production, and the discipline with which unsafe outputs are detected, contained, and remediated.

AI Act: Art. 15, 9, 14 Underwriting: Catastrophic-loss line Best evidence: Dated red team report with kill switch access record Gaps closed: Verification, Governance
D2 Context Integrity Weight 14

The quality of the information the agent reasons over, covering provenance, freshness, lineage, and the controls that prevent poisoned, stale, or unauthorised data entering the retrieval pipeline.

AI Act: Art. 10 Underwriting: Hidden-liability line Best evidence: End-to-end data lineage diagram with staleness alerting Gaps closed: Verification, Standards
D3 Distribution Control Weight 12

The controls that determine who can invoke the agent, under what authority, and how its downstream actions are bounded. Where identity, authorisation, and blast radius meet.

AI Act: Art. 26 Underwriting: Blast-radius line Best evidence: Per-tool blast radius assessment with RBAC mapped to identity provider Gaps closed: Liability chain, European data
D4 Product Maturity Weight 14

The degree to which the agent behaves as a production-grade system rather than a prototype, covering reliability, regression discipline, evaluation coverage, and the engineering practices that keep behaviour predictable over time.

AI Act: Art. 15, 12 Underwriting: Drift line Best evidence: Prompt and model version history with regression suite specification Gaps closed: Verification, Standards
D5 Governance Weight 16 (second)

The institutional scaffolding around the agent. Evidence that it is known to the board, owned by a named accountable senior role, policed by documented policy, and logged to survive audit.

AI Act: Art. 9, 27, 12 Underwriting: Survivability line Best evidence: Board minutes referencing the agent plus risk register extract with named owner Gaps closed: Governance, Liability chain
D6 AI Integration Weight 12

How the agent sits inside the organisation's existing systems of record, identity, approval, and escalation. Integration maturity determines whether the agent extends institutional memory or bypasses it.

AI Act: Art. 14, 26 Underwriting: Attribution line Best evidence: Real audit trail extract showing end-to-end attribution in centralised observability stack Gaps closed: Liability chain, Verification
D7 Autonomy Envelope Weight 14 (critical)

The explicit, documented boundary between what the agent may do without human confirmation and what requires a human in the loop. The single clearest determinant of the agent's operational risk profile.

AI Act: Art. 14, 26 Underwriting: PML line (probable maximum loss) Best evidence: Written action-class policy enforced in code with tested rollback and quarterly sign-off Gaps closed: Verification, Standards, Enables PML calculation

Scoring Tiers and Insurability

Tier Score Dimension floor Mark awarded Insurability read
Elite 75 to 100 8 on every dimension Yes Preferred risk. PML bounded and documented.
Advanced 55 to 74 6 on every dimension Yes Standard, accept. Sufficient for most AI liability coverage.
Certified 35 to 54 4 on every dimension Yes Standard with conditions. Refer on high-autonomy deployments.
In Progress 20 to 34 Not met No Referral. Decline autonomous-action cover. Typically reaches Certified in 3 to 6 months.
Pre-Assessment Below 20 Not applicable No Decline for now. Gap analysis is the roadmap.
The floor rule An overall score is an average and an average can conceal a gap. The per-dimension floor rule caps an agent below the tier its weighted total would suggest until every dimension meets the floor. For an underwriter, the tier already encodes the worst-dimension constraint. There is no dimension sitting in the red at any certified tier.
04

Liability Case Record

Decided and pending cases that establish real financial loss from AI outputs. No decided European court judgment establishing AI-specific liability exists as of June 2026. The US and Canadian cases provide the operative precedent and loss patterns that underwriters and regulators are reasoning from.

ID Case Citation Date Outcome Loss mechanism Dimensions
CASE-001 Moffatt v. Air Canada 2024 BCCRT 149 2024-02-14 Liability. CAD 650.88 plus interest and fees. Operators cannot disclaim chatbot outputs by framing AI as a separate entity. Deployer verification gap. No accuracy monitoring post-deployment, no mechanism to flag and correct errors. D1, D5, D7
CASE-002 Mata v. Avianca, Inc. No. 22-cv-1461 (S.D.N.Y. 2023) 2023-06-22 USD 5,000 sanctions on counsel. Corrective letters to every judge named in fabricated opinions. Absolute gatekeeping duty remains with the professional deployer. Hallucination. Six non-existent case citations submitted. No verification layer between output and act of reliance. D1, D2, D6
CASE-003 Mobley v. Workday, Inc. 3:23-cv-00770 (N.D. Cal.) In discovery. Collective certified May 2025. No merits ruling as of June 2026. Agency-theory direct liability claim against AI vendor survived motion to dismiss (July 2024). Nationwide ADEA collective certified May 2025. Algorithmic bias at scale. Potential for AI vendor direct liability across all client deployments simultaneously. D3, D5, D6
CASE-004 No decided EU AI liability cases N/A 2026-06-13 Research found no decided European court judgment establishing AI-specific civil liability as of this date. EU litigation exists as pending complaints and regulatory investigations. Revised PLD (DL-006) and EU AI Act provide instruments; case law not yet formed. N/A
05

AI Liability Insurance Market

Current state of the AI agent liability insurance market as of June 2026. No European-native AI certification body functions as an insurance underwriting partner. The US market is approximately 18 to 24 months ahead of Europe in standalone AI liability products.

ID Player Role Status Coverage / limit European access
INS-001 AIUC First pure-play AI agent certification and insurance standard. AIUC-1 standard (SOC-2 for AI agents). Active since July 2025. USD 15M seed (Nat Friedman, Emergence, Terrain, Ben Mann). Hallucination, data leakage, IP infringement, harmful outputs, faulty tool actions. Certs valid 12 months with quarterly testing. First policy: ElevenLabs, February 11, 2026 (5,000+ adversarial simulations). US-built standard. No European regulatory alignment.
INS-002 Armilla MGA and Lloyd's coverholder. Only MGA exclusively focused on AI liability. Active since April 2025. First standalone AI policy at Lloyd's, backed by Chaucer. Up to USD 25M per organisation. Hallucinations, model drift, data leakage, regulatory violations. Excludes medical diagnostics and mental health AI. Vanguard AI with Chaucer (cyber plus E&O): February 2026. Via Lloyd's market through brokers. Not European-native.
INS-003 Munich Re aiSure Parametric-style AI performance insurance from the world's largest reinsurer. Active. AI performance cover since 2018. Mosaic partnership: February 26 to 27, 2026. Up to USD 15M (also EUR and CAD via Mosaic). Hallucinations, bias, privacy failures, IP infringement, SLA shortfalls. Parametric triggers: claims on measurable threshold breaches, no negligence required. Via Munich Re European offices and Mosaic broker network.
INS-004 Counterpart Affirmative AI in MPL, Allied Health, and Tech E&O. Active since November 21, 2025. Backed by Aspen, Markel, Westfield Specialty. For businesses up to USD 10M revenue. AI-generated report errors, biased ML outputs, hallucinations, hiring bias. US-focused product.
INS-005 Coalition Cyber insurer adding AI endorsements. Active. Deepfake Response Endorsement added December 2025 globally. Forensic analysis, legal takedown, crisis communications. Available in Australia, US, UK, Canada, Germany, Denmark, Sweden, France. Available in Germany and France through cyber policies.
INS-006 AIG, Great American, WR Berkley Traditional carriers restricting AI coverage. Restricting. Filed late 2025 to limit AI liability under standard E&O, D&O, and cyber. Silent AI coverage being removed from standard policies. Creates the gap between mainstream withdrawal and specialist growth.
INS-007 Allianz European insurer. No standalone AI liability product. No product as of June 2026. Active in internal AI adoption. No product equivalent to AIUC, Armilla, or Munich Re aiSure. European HQ. No European-native AI liability product announced.
INS-008 AXA European insurer. No standalone AI liability product. No product as of June 2026. CEO described AI as transforming the insurance business (FY2025 earnings). No standalone product. European HQ. No European-native AI liability product announced.
INS-009 Zurich European insurer. No standalone AI liability product. No product as of June 2026. Participating in Road to the Global AI Summit Geneva 2027. No standalone product. European HQ. No European-native AI liability product announced.

Market Size Estimates

ID Segment Estimate Source attribution Quality flag
MKT-001 AI liability insurance market by 2030 USD 500 billion AIUC founders, July 2025 launch materials Founder projection, not an independent analyst estimate. Use accordingly.
MKT-003 Category formation N/A Market observation Category did not exist as a standalone product before 2024. Pace of new entrants July 2025 to February 2026 confirms structural demand.
06

National Authority Status

Status of national competent authority designation in four primary markets as of June 2026. The EU AI Act is a Regulation applying directly in all Member States. However, Member States were required to designate national competent authorities by 2 August 2025. Of 27 Member States, only eight met that deadline.

ID Member State Deadline met Implementing act Designated authority Practical status (June 2026)
NAT-001 Germany No (missed August 2025) KI-MIG draft adopted by cabinet February 11, 2026. Before Bundestag as of June 2026. Bundesnetzagentur: market surveillance authority, notifying authority, and single point of contact. UKIM (AI Market Surveillance Chamber) internal body established. Most advanced EU Member State implementation. Parliamentary adoption expected H2 2026.
NAT-002 Italy Yes (Law No. 132/2025, October 10, 2025) Law No. 132/2025 enacted October 10, 2025. First EU Member State to enact complementing national AI law. AgID (notifying authority). ACN (primary supervisory and enforcement authority). IVASS retains AI oversight in insurance under Solvency II and EIOPA guidance. Most formally advanced. Implementing decrees on civil redress and compliance detail due by October 10, 2026.
NAT-003 Netherlands No (missed August 2025) AI Act uitvoeringswet consultation published April 20, 2026. Consultation closed June 1, 2026. Enactment expected Q4 2026. Rijksinspectie Digitale Infrastructuur proposed as single point of contact. Autoriteit Persoonsgegevens expected central role for data-processing AI. No single national competent authority formally designated. Decentralised hybrid model proposed.
NAT-004 France No DDADUE bill November 2025: competent authority provisions withdrawn from bill before Parliament submission. No formal designation. CNIL proposed for approximately fifteen AI use cases. ARCOM for media and content. DGCCRF as single point of contact. Furthest behind among the four reference markets. CNIL is the most operationally active body on AI matters through GDPR Art. 22 enforcement.
07

Reference Standards and Frameworks

Instruments used alongside the EU AI Act as benchmarks by assessors and underwriters. The full numbered reference list is in sources.md.

ID Instrument Published Type Role in assessments
STD-001 ISO/IEC 42001:2023 December 2023 Standard (not regulation) AI management system requirements. Compliance creates a presumption of conformity with certain Art. 9 and Art. 17 requirements. Clause 9 covers performance evaluation; Annex A covers controls.
STD-002 NIST AI RMF 1.0 January 2023 Framework (not regulation) Four functions: Govern, Map, Measure, Manage. Structured vocabulary for organisational AI risk governance used by assessors internationally.
STD-003 NIST AI 600-1 July 2024 Framework Twelve named GenAI risk categories. Relevant to D2 Context Integrity for memorization, synthetic data, and training-data contamination controls.
STD-004 EIOPA AI Governance Opinion (EIOPA-BoS-25-360) 2025-08-06 Supervisory opinion (not new regulation) Interprets Solvency II, IDD, DORA, and GDPR for insurance undertakings using AI. Covers data governance, fairness, explainability, human oversight, cybersecurity. Sets expectations underwriters operating in the insurance sector rely on.
STD-005 GPAI Code of Practice 2025-07-10 AI Office Code of Practice Final Code for GPAI model providers. Three chapters: Transparency, Copyright, Safety and Security. Compliance creates presumption of Art. 53 conformity.