The European Insurance and Occupational Pensions Authority (EIOPA) has proposed draft guidelines on artificial intelligence use in insurance.
According to Insurance Europe, the guidelines could improve understanding of current regulatory expectations but may also raise concerns about legal consistency, proportionality, and scope.
EIOPA issued its draft opinion to clarify how existing insurance laws apply to AI systems. The opinion outlines supervisory expectations and calls for responsible use, including ethical conduct, consumer protection, and transparent decision-making (see about AI Integration in Investment).
EIOPA stated that its aim is not to introduce new requirements, but to guide how existing rules apply throughout an AI system’s lifecycle using a risk-based approach.
Insurance Europe warned, however, that insurers may view these expectations as binding. This could unintentionally create additional regulatory obligations.
The trade group also stressed that the guidelines should reflect the European Commission’s broader effort to reduce regulatory pressure on small and medium-sized companies.
It recommended clearer distinctions between internal and external AI applications, between those who develop AI systems and those who use them, and between machine learning and generative AI.
Insurance Europe also urged EIOPA to address situations where dual supervision applies—specifically in jurisdictions where separate authorities oversee insurance compliance and the AI Act.
In such cases, clear delineation of responsibilities is necessary.
The trade body further advised EIOPA to emphasize how existing insurance rules already cover governance, risk control, and consumer rights, reducing the need for duplicative oversight.
