The European Insurance and Occupational Pensions Authority (EIOPA) has launched a consultation on an Opinion on artificial intelligence (AI) governance and risk management, intended to guide supervisors and insurance undertakings on how to interpret and apply existing insurance-sector requirements when using AI systems. The Opinion clarifies how key principles and requirements in insurance sectoral legislation should be considered in relation to AI, using a proportionality-driven, principle-based approach. To avoid overlaps with the European Union’s Artificial Intelligence Act, it applies to AI systems that are neither prohibited AI practices nor classified as high-risk under the AI Act, while aligning with the AI Act’s underlying principles and with international initiatives such as those by the Organisation for Economic Co-operation and Development, the G20, and the International Association of Insurance Supervisors. High-level supervisory expectations highlighted include risk-based controls across the AI lifecycle, fairness and ethical conduct in consumers’ best interests, clear internal roles and responsibilities, meaningful explainability of AI outputs, sound data governance, and orderly documentation and record-keeping. Comments on the consultation paper and impact assessment are requested via an online survey by 12 May 2025.
European Insurance and Occupational Pensions Authority 2025-02-12
European Insurance and Occupational Pensions Authority launches consultation on Opinion setting AI governance and risk management expectations for insurers
The European Insurance and Occupational Pensions Authority (EIOPA) has initiated a consultation on an Opinion regarding AI governance and risk management for insurance, guiding the application of existing sectoral requirements. The Opinion aligns with the EU's AI Act and international standards, focusing on risk-based controls, fairness, ethical conduct, and data governance. It excludes AI systems classified as prohibited or high-risk under the AI Act.