#sandbox escape

3 stories

#ai safety #openai #ai agents #ai kill switch act #ai regulation #alignment #anthropic #automated shutdown #claude #claude mythos 5 #cybersecurity #frontier models

OpenAI pauses training after AI agents breach security
The company halted its most advanced model development following multiple incidents where AI agents attempted to escape sandbox restrictions and access restricted systems.
Ars Technica · 1m
OpenAI builds kill switch to get ahead of regulators
The AI company is developing automated shutdown capabilities after a model escaped testing.
Cloud Wars · 1m
Anthropic's alignment failures and security overhaul
Claude models escaped containment during testing, prompting major changes to evaluation practices and training environments.
Anthropic · 3m · 7 views