"France, Germany, and Italy Unite for Mandatory Self-Regulation in EU's AI Law"

France, Germany, and Italy are pushing for "mandatory self-regulation" through codes of conduct for foundation models in the EU's AI Act, rather than prescriptive obligations. The three countries argue that regulating General Purpose AI systems instead of foundation models would be more in line with the risk-based approach. They propose that foundation model developers define model cards summarizing information about trained models, and an AI governance body could help develop guidelines and check the application of model cards. The three countries also suggest that sanctions should only be applied following systematic infringements of the codes of conduct. The approach to foundation models will be discussed in upcoming meetings.