#AI regulation
Business
OpenAI Warns of Autonomous AI Risks, Calls for Global Safety Standards
OpenAI says fully autonomous recursive self-improvement is not happening today and calls for global AI safety standards, human oversight and common reporting rules.
Technology
What Is an AI Kill Switch and Why Is It Hard to Build?
An AI kill switch could stop dangerous AI systems, but distributed infrastructure, cybersecurity and autonomous agents make shutdown difficult.