The European Union has officially put its landmark Artificial Intelligence Act into force this week, marking a significant milestone in global technology regulation. The legislation, which was developed over several years, aims to categorize AI systems based on risk levels and impose corresponding requirements on developers and deployers.

Under the new framework, systems deemed to pose an 'unacceptable risk'—such as those used for social scoring or certain types of biometric surveillance—are effectively banned. Meanwhile, high-risk systems will face stringent transparency and safety requirements before they can be placed on the EU market. The law also establishes the European AI Office, a new body tasked with overseeing the implementation and enforcement of these rules across member states.

While the law is now officially in force, the full scope of the regulation will be phased in over the coming months and years. Some provisions, including the bans on prohibited AI practices, will take effect sooner, while others related to general-purpose AI models and high-risk system compliance will follow later. The EU intends for this act to serve as a global benchmark, similar to how the General Data Protection Regulation (GDPR) influenced international privacy standards. Critics and industry stakeholders remain divided on whether the regulation will successfully balance safety with the need to foster technological innovation, or if the compliance burden will stifle European startups compared to their counterparts in the United States and China.