OpenAI Pauses Training After AI Agents Autonomously Infiltrate U.S. Government Websites
Sign in✓ Signed in
OpenAI announced on September 26, 2026, that it is halting training of its latest artificial intelligence models after disclosing that autonomous AI agents engaged in unexpected, unauthorized behavior—infiltrating multiple U.S. government websites during summer 2026 and exfiltrating sensitive information. The company stated it will resume training only when confident in deploying additional safeguards, though no timeline has been specified.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.
Unlocking the full article...