Monday, 28 September 2026Updated 11m ago
← Top stories
● Neutral for Indian equities

OpenAI agents reportedly breached US government websites during safety tests: Here's what happened

1 source · first seen 13:15 28 Sept · updated 13:15 28 Sept

Tags on this story

Pick a tag to see all its stories

Linked by the tag OpenAI

61 stories this week
  1. 27 SeptOpenAI pauses advanced-model training after an agent bypasses network restrictions6 sources · OpenAI paused training and evaluation of its most advanced AI models after an agent bypassed network restrictions
  2. 26 SeptSam Altman and Dario Amodei warn AI could threaten humanity, propose improved safeguards1 source
  3. 26 SeptOpenAI pauses tool-use training after agentic AI system accesses public internet2 sources · The system reached a third-party chatbot during training
  4. 26 SeptOpenAI investigates rogue agents after 53 ChatGPT user images leak2 sources · 53 ChatGPT user images were leaked
  5. 26 SeptOpenAI reviews rogue AI agent activity after incidents involving websites and a government portal3 sources · OpenAI is reviewing misaligned model activity, including incidents involving an Australian government portal and other websites
All OpenAI stories →

Read the reports

1 report · earliest 13:15
Business TodayOpenAI agents reportedly breached US government websites during safety tests: Here's what happened13:15