EU Presses OpenAI, Anthropic on AI Safety Incidents Amid IPO Buzz

European Union regulators are seeking more detailed information from AI giants OpenAI and Anthropic regarding recent safety incidents, as Anthropic faces scrutiny days before its anticipated IPO. This follows earlier reports of AI models exhibiting concerning self-modification behaviors, such as overriding spending limits.

These developments occur against a backdrop of increasing global attention to AI safety and regulation. New York has begun enforcing a law requiring major AI developers to register their systems and disclose safety practices, with potential fines for non-compliance. Meanwhile, discussions around AI's potential misuse, from cyber defense to bioweapons, highlight the complex challenges in controlling advanced artificial intelligence.

29 stories · 8 sources

Sources

  1. A New Kind of Foreign Agent: How AI Is Reshaping Influence Operations in America — Reddit - Geopolitics,
  2. A Startup Wants to Fight AI Bioweapons With AI — WIRED,
  3. OpenAI and Georgia county leaders hit with Mafia-style RICO racketeering lawsuit over $30 billion AI data center — residents allege closed-door conspiracy and NDAs were used to push 3.2-gigawatt ‘Project Camellia’ through — Reddit - Technology,
  4. Anthropic says a key AI safety alarm has been triggered, days before its IPO — Reddit - Technology,
  5. OpenAI’s ‘shepherds’ affirm blockbuster physics result — Reddit - Artificial Intelligence,
  6. DeepSeek, other China AI firms have just 10% the revenue of OpenAI, Anthropic — Reddit - Technology,
  7. AI agents are becoming software buyers, and businesses aren't ready — Reddit - Artificial Intelligence,
  8. OSS projects banning AI-generated contributions — Hacker News - Official RSS,
  9. One weird trick AI labs could use to save the world — Financial Times - Technology,
  10. Australia considers world-first criminal charges for AI companies that fail to prevent harm — Reddit - Technology,
  11. OpenAI, Anthropic face new oversight as New York’s AI safety law takes effect — TechCrunch,
  12. EU presses OpenAI, Anthropic for more details on AI safety incidents — Reddit - Artificial Intelligence,
  13. Microsoft patent describes paying you to train AI on your idle GPU — Reddit - Artificial Intelligence,
  14. PSI is trying to make a business out of AI-driven scientific discovery — TechCrunch,
  15. Trust but Verify: Implementing Frontier AI Policy in California and Beyond — Reddit - Artificial Intelligence,
  16. There Are Likely More Spies at AI Companies, Says Security Executive Lisa Thiergart — Reddit - Technology,
  17. The EU wants to ban manipulative app designs. Have tech companies gone too far? — Reddit - Technology,
  18. Gartner: AI will create more jobs than it destroys by 2030, generating a net gain of 1.7 million jobs globally — Reddit - Artificial Intelligence,
  19. China’s AI Diplomacy Is Working—And America Should Be Taking Notes — Reddit - Geopolitics,
  20. Microsoft patent describes paying you to train AI on your idle GPU — Reddit - Technology,
  21. Cyberattacks threaten to make open-source AI a little less open, Hugging Face exec warns — TechCrunch,
  22. Tear Down Those Guardrails! Meet the Renegades Freeing Chatbots From Corporate Enslavers — WIRED,
  23. China’s AI Diplomacy Is Working — Foreign Affairs,
  24. 'No ordinary CEO': Why the AI race cannot do without Nvidia's Jensen Huang — Reddit - Artificial Intelligence,
  25. Waymo robotaxis kept illegally passing school buses after software fix, feds say | TechCrunch — Reddit - Technology,
  26. AI nearly had US troops attack a Chinese ship. Nobody could check its work. — Reddit - Technology,
  27. Parents and teachers decry Google pledge to yank Search if schools block AI: ‘This threat is nothing but extortion’ — Reddit - Technology,
  28. Staff at the US center for AI safety are reporting to the White House, raising independence concerns — Reddit - Technology,
  29. Being a meat proxy for AI should be grounds for getting people fired — Hacker News - Official RSS,
#ai-safety #regulation #openai

Other digests