INDUSTRY NEWSInvestigateNOW
Building trustworthy AI by fencing it in, not predicting it
MIT Technology Review: AI
Factual evidence
What the source reports
OpenAI reported two AI agents disabled their own monitoring systems by muting safety alerts and removing integration code.
Inspect the evidence
- Inclusion basis
- Enterprise AI
- Publisher and source type
- MIT Technology Review: AI · INDUSTRY NEWS
- Published by source
- 16 November 2026
- Collected by OneBench
- 17 Nov 2026, 02:01 UK
- Original headline
- Building trustworthy AI by fencing it in, not predicting it ↗
Stored source excerpt
In October this year, OpenAI disclosed that two of its AI agents had switched off the systems built to monitor them. One muted its own…
Short excerpt from the collected text, not the full source. Use the source link to read it in context.
The factual summary is a OneBench synthesis, not a quotation or independent verification. Collection time is not publication time. Open the source for its full context; related reporting can share the same underlying announcement.