European Lawmakers Finalize Landmark Artificial Intelligence Regulation
EU negotiators reach a provisional agreement on the Artificial Intelligence Act, establishing risk-based rules to protect fundamental rights while fostering innovation. The legislation bans harmful practices like social scoring and places strict limits on law enforcement facial recognition.
European Parliament and Council negotiators reach a provisional agreement on the Artificial Intelligence Act to ensure AI remains safe and respects fundamental rights. This landmark regulation establishes obligations based on potential risks and impacts, aiming to protect democracy, the rule of law, and environmental sustainability while boosting innovation and enabling European businesses to thrive.
The legislation explicitly bans several AI applications that pose severe threats to citizens, including social scoring based on behavior or personal characteristics. It also prohibits emotion recognition in workplaces and schools, the exploitation of human vulnerabilities, and the untargeted scraping of facial images from the internet to create recognition databases. Additionally, consumers gain the right to launch complaints and receive meaningful explanations about AI-driven decisions.
Law enforcement agencies face strict safeguards and narrow exceptions regarding the use of biometric identification systems in public spaces. Real-time facial recognition requires prior judicial authorization and remains limited to specific locations, timeframes, and serious crimes like abduction or trafficking. Companies that violate these comprehensive rules face significant fines ranging from 7.5 million euros or 1.5% of global turnover up to 35 million euros or 7% of global turnover.