EU AI Act triggers regulatory concerns for arbitration institutions

The implementation of the European Union AI Act is raising complex regulatory questions regarding the classification of high-risk artificial intelligence systems. Draft guidelines published by the European Commission aim to clarify these classifications, but have drawn concern from major legal entities.

Sixteen leading arbitral institutions, including the International Chamber of Commerce (ICC), the London Court of International Arbitration (LCIA), and the International Centre for Settlement of Investment Disputes (ICSID), have issued a joint response. They warned that the proposed guidelines risk imposing regulatory obligations on arbitral institutions and tribunals, creating uncertainty regarding who bears responsibility when AI is utilized in legal proceedings.

Parallel to these regulatory challenges, academic analysis of the European Parliament’s role reveals the discursive strategies used to shape the Act. Research into parliamentary documents from 2019 to 2024 highlights how the institution navigated debates on sensitive issues, such as remote biometric identification and law enforcement powers, to construct the policy narrative that ultimately defined the legislation.