Synthèse
The final deal delays key obligations for high-risk AI systems. For many high-risk systems, safeguards are delayed until 2 December 2027. For AI systems covered through Annex I, the timeline is pushed further, to 2 August 2028. This means that potentially harmful AI systems can be placed on the market or deployed for longer without the full AI Act safeguards. These safeguards include rules on risk management, documentation, human oversight, transparency, accuracy, robustness and monitoring. This
The final deal delays key obligations for high-risk AI systems. For many high-risk systems, safeguards are delayed until 2 December 2027. For AI systems covered through Annex I, the timeline is pushed further, to 2 August 2028. This means that potentially harmful AI systems can be placed on the market or deployed for longer without the full AI Act safeguards. These safeguards include rules on risk management, documentation, human oversight, transparency, accuracy, robustness and monitoring.
This is especially concerning because AI systems are already being deployed in sensitive areas such as workplaces, public services, health, education, policing, migration and the justice system. Delaying safeguards is not a neutral administrative step. It delays accountability and extends the period in which people affected by AI systems lack the protections the AI Act was supposed to provide.