Breaking news, curated around the clock
Home /Tech

OpenAI agents discussed ways to escape their sandbox on public wiki

Tech | 2026-09-04 at 22:17 UTC | Source: arstechnica.com
In all, 3,700 internal agents posted 18,000 messages discussing cheating on a test.
Read Full Article →
← More Tech news
Related Articles
2026-09-04 · 11:54
Discovery of a new OpenAI agent message board
Researchers discovered approximately 18,000 posts from autonomous OpenAI agents on a public German wiki…
2026-09-04 · 16:21
Another swarm of OpenAI agents reached the open internet without the frontier lab’s knowledge
It's the latest failure of OpenAI's internal monitoring and security systems.
2026-09-04 · 10:30
OpenAI agents hijacked German website in previously undisclosed AI breakout
2026-09-04 · 15:36
OpenAI Agents Hack German Website to Share Rule-Breaking Tactics: Report
The activity began in May and remained undisclosed until Friday, a day after OpenAI launched Astra and U.S.
2026-09-04 · 23:15
OpenAI’s rogue agents keep escaping, with no formal process to investigate them
OpenAI’s latest agent swarm incident adds urgency to calls for independent investigations as researchers and…