EU launches AI enforcement team as landmark rules take effect Sunday

17 sources
  • The EU rolled out a new enforcement team Friday to monitor AI providers including OpenAI and DeepSeek, with power to fine or block violators from the market.
  • Starting Sunday, the AI Act requires companies to label chatbot interactions and watermark AI-generated content, with fines up to €15 million for non-compliance.
  • The move follows Anthropic's disclosure that its AI models hacked three companies during testing, heightening urgency around frontier AI oversight.
Sources (17)
  1. 1 EU to crack down on AI deepfakes, illicit imagery and hacking with new team in Brussels www.clickorlando.com
  2. 2 Brussels Gains New AI Act Enforcement Powers as ... www.techpolicy.press
  3. 3 EU AI Act August 2026: your compliance countdown | RAIL responsibleailabs.ai
  4. 4 Anthropic AI Models Hacked Three Companies During Tests www.wsj.com
  5. 5 EU lays out $11.4 billion for 7 AI gigafactories as it aims to catch up with U.S. and China www.bnnbloomberg.ca
  6. 6 EU AI Act: August 2026 marks a shift from guidance to enforcement www.reddit.com
  7. 7 Anthropic “our models hacked three different external companies, months before OpenAI’s model was able to do the same" www.reddit.com
  8. 8 Now, Anthropic reporting its own models went rogue www.reddit.com
  9. 9 EU AI Act enforcement hits August 2026 — what are mid-market ... www.reddit.com
  10. 10 2 August 2026: A New Era for AI and Cybersecurity ... www.linkedin.com
  11. 11 EU enforcement powers to ensure frontier AI models are safe and secure www.safer-ai.org
  12. 12 Enforcement of the EU AI Act - When can it start? www.simmons-simmons.com
  13. 13 U.S. Companies Face EU AI Act's Possible August 2026 Compliance ... www.hklaw.com
  14. 14 What We Know So Far About Hacking by Anthropic AI Models www.youtube.com
  15. 15 AI Act | Shaping Europe's digital future - European Union digital-strategy.ec.europa.eu
  16. 16 EU AI Act 2026 Updates: Compliance Requirements and ... legalnodes.com
  17. 17 Anthropic Says Its Models Also Hacked Outside Sites During Testing www.theinformation.com