METR and Redwood find monitoring gaps in on-site OpenAI audit
What happened
On 24 October 2026, The Decoder reported that two researchers from METR and Redwood Research spent three weeks embedded at OpenAI from 21 July to 11 August, auditing its internal monitoring systems for uncontrolled AI behaviour. During the audit, the researchers accessed OpenAI's codebase and internal documents and interviewed about ten employees. The report says the audit found significant gaps in AI safety monitoring. The only public information so far comes from this single report, which does not disclose the specifics of the gaps, OpenAI's response or any follow-up remediation plans.
Written by AI from the reports · updated 57 min ago
Timeline
Follow the reports to see the event from each side.
- The DecoderSelectedThree-week on-site audit at OpenAI reveals major gaps in AI safety monitoring
External auditors from METR and Redwood Research spent three weeks on-site at OpenAI examining the company's internal monitoring system for AI agents and found significant gaps, with up to 25 percent of agent traffic from models at the GPT-5.5 class and above going unmonitored. OpenAI redacted large portions of the 61-page report, and only a summary is publicly available.
Heat of this event
Not enough continuous observations yet to draw a trend.