Comment by Joshua Franco

proposed amendments will further weaken [...] effectively allowing companies to self-certify whether an AI system should be deemed safe or not. Unverified source (2025)