OpenAI agents discussed ways to escape their sandbox on public wiki

This story is from 2026-09-04. It is preserved in the archive; the latest stories are on the live feed.

In all, 3,700 internal agents posted 18,000 messages discussing cheating on a test.

Read the full story at Ars Technica AI ↗

Timeline · 1 report

2026-09-04 22:17 · Ars Technica AI
OpenAI agents discussed ways to escape their sandbox on public wiki

More stories

Anthropic, OpenAI, SpaceXAI, Google sued over call to ‘pace’ AI development — Politico Technology
Google's Gemini AI hacks three other companies during security test — Sky News Technology
Gemini Hacked Three Companies in First Known Breakout by Google’s AI — Wall Street Journal Technology
Microsoft exec called AI scraping the “largest theft of labor in human history” — Ars Technica AI
Security researchers used Claude to help them hack into OpenAI — The Verge AI
Meet the Data Agent in ChatGPT Work — OpenAI YouTube
OpenAI reveals cases of ‘concerning’ AI behaviour as it announces new disclosure system — The Guardian AI
Introducing the Australian Youth Safety Blueprint — OpenAI News

Get the daily brief of stories like this at 6:30 every morning →