Context
The European Insurance and Occupational Pensions Authority (EIOPA) has outlined six principles to promote the ethical and trustworthy use of artificial intelligence (AI) in the insurance sector. Although the UK is now outside of EIOPA’s perimeter, UK Regulators still work closely with EIOPA. UK insurers, MGAs, underwriting agencies and brokers operating with significant freedoms under delegated authorities should perhaps review the report. Although the report was published mid-June, the law firm Pinsent Masons, via their Out-Law online publications, has published a very informative summary article on 7th July.
Key points to note
- The report builds on recent international and EU developments in digitalisation and AI. EIOPA convened the consultative expert group so that a wide range of stakeholders could work together identifying the opportunities and risks associated with the growing use of AI in insurance, including exploring possible limits that might be needed.
- The report highlights the proliferation of AI in insurance. For instance, AI allows insurance companies to carry out more granular risk assessments and pricing practices, increasingly efficient claims management processes, and enables insurance companies to fight against fraud more effectively.
- Benefits such as prediction accuracy, automation, new products and services, and cost reductions can be remarkable. However, ensuring fairness, non-discrimination, transparency and ‘explainability’ can be challenging; there is work for firms to address these challenges.
- The issues raised are of critical importance in the face of the digital transformation currently underway in the sector. Developing a sound AI governance framework is therefore essential for a well-functioning, trusted and financially inclusive insurance sector.
- The report aims to help insurance companies when putting in place risk-based and proportionate measures, providing guidance on how to implement key principles in practice throughout the lifecycle of an AI application. The guidance is non-binding and illustrative.
- The report is set out under several section headings including themes which build on existing insurance principles and requirements, principles of the General Data Protection, and wider ethics considerations:
- fairness and non-discrimination
- transparency and ‘explainability’
- Human Oversight:
- data governance of record keeping
- Robustness and Performance
Next actions
None – for information.