The deal also restricts the use of real-time biometric surveillance in public spaces and bans cognitive behavioral manipulation, untargeted scrapping of facial images, social scoring, and biometric categorization systems. Consumers will have the right to launch complaints and receive explanations, with fines for violations ranging from 7.5 million euros to 35 million euros. The legislation is expected to come into force early next year and apply two years after ratification. However, the rules have been criticized by business and privacy rights groups for adding burdens to companies and legalizing live public facial recognition.
Key takeaways:
- Europe has reached a provisional deal on new European Union rules governing the use of artificial intelligence, including biometric surveillance and AI systems like ChatGPT.
- The agreement requires AI systems to comply with transparency obligations before they are put on the market, including technical documentation, EU copyright law compliance, and detailed summaries about training content.
- Real-time biometric surveillance in public spaces can only be used in cases of certain crimes, prevention of threats, and searches for people suspected of serious crimes. The agreement also bans cognitive behavioral manipulation and untargeted scrapping of facial images.
- Violations of these rules could result in fines ranging from 7.5 million euros ($8.1 million) or 1.5% of turnover to 35 million euros or 7% of global turnover. The legislation is expected to be enforced early next year.