While supporting the aim to offer clarity on applying existing insurance legislation to AI systems, Insurance Europe raises several concerns around the Opinion’s scope, proportionality, and legal coherence.
Insurance Europe stresses that while EIOPA’s Opinion does not intend to introduce new requirements, the current draft risks doing just that. The language used may result in supervisory expectations being interpreted as binding obligations on insurers.
The Opinion also must more effectively reflect proportionality considerations, particularly in light of the European Commission’s broader focus on simplification and reducing the burden faced by small and medium-sized companies. Insurance Europe urges EIOPA to recognise, for example, differences between internal and customer-facing applications, between AI deployers and providers, and between machine learning and generative AI systems.
Insurance Europe also highlights the risk of dual supervision in some countries, where different authorities may be responsible for monitoring insurers’ compliance with the AI Act versus sectoral rules. The Opinion should acknowledge this possibility and provide clarity on how overlaps will be managed.
Insurance Europe reiterates that the focus of the Opinion should be on clarifying how existing European insurance sector rules – such as Solvency II, Insurance Distribution Directive, General Data Protection Regulation and the Digital Operational Resilience Act – provide the necessary mechanisms to ensure proper governance, risk management, and consumer protection. Insurance Europe remains committed to constructive dialogue with EIOPA and other stakeholders to ensure that the regulatory framework in the insurance sector can continue to support innovation and the use of AI in the sector.
EIOPA Draft Opinion on AI
|
