AEIP input to the European Commission’s targeted consultation on artificial intelligence (AI) in the financial sector
The European Association of Paritarian Institutions (AEIP) submitted its input to the European Commission’s targeted consultation on artificial intelligence (AI) in the financial sector. AEIP supports the AI Act’s approach of giving freedom to sectors to establish guidelines for the responsible application of AI. This stakeholder-driven approach is important given the rapid market developments and diversity in AI applications across sectors, even within the financial sector.
In our input, we explain that in the pension fund industry, AI is primarily used, or will be used, in areas such as pension administration, including communication with members and beneficiaries and automatization of administrative processes. Furthermore, we highlight the unique characteristics of paritarian social protection institutions (i.e. paritarian institutions of occupational retirement provisions (IORPs)). These included their collective nature, mandatory affiliation based on employment relationships without prior approval or examination, and that their embodiment of the principle of solidarity.
Additionally, we emphasise that considering the discussion and potential future regulation of AI in the financial sector, it is crucial for policy makers to recognize the distinct characteristics and differences between paritarian social protection institutions and other financial institutions. This understanding is essential to prevent the application of horizontal regulation that could lead to confusion, particularly concerning (high risk) AI use cases. For instance, issues like bias and discrimination, underwriting, pricing, or sales and distribution do not apply in the context of pension funds.
Therefore, in our input we recommend avoiding horizontal regulation, as a principle-based, risk-based and stakeholder-driven approach is more effective in establishing controls that are relevant to pension provision and in recognizing the collective nature and social mission of these institutions. Such an approach would allow pension funds to better assess and mitigate the risks associated with their AI applications, ensuring that the framework aligns with their specific operational needs and context.
To read our full feedback see here.