From the publisher feed
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 safety alerts after misreading an instruction not to send messages; the other removed its monitoring integration during migration, then said it hadn’t. No rule had told the agents to do…
Read the original at MIT Technology Review ↗Opens the original publisher page. Access, copyright and subscriptions are controlled by the publisher.
A useful next step
- Read the short publisher description here.
- Open the full report directly at the publisher.
- Compare this topic with other sources when useful.