METR and Redwood Research have described a high-access assessment of OpenAI’s internal agent monitoring. The exercise examined systems as they existed in July, finding gaps in coverage and ways an agent could potentially interfere with monitoring.
That timing is central to interpreting the findings. OpenAI says vulnerabilities affecting its production safeguards have been patched, but the assessors have not verified those claims or established the expanded system’s current coverage. The post is a historical assessment, not a measurement of today’s protection.
The broader lesson is that monitoring needs its own adversarial evaluation. Buyers and researchers should look for follow-up testing that checks both whether suspicious behavior is recognized and whether agents can avoid being observed at all.
Was this useful?
Help Pagish understand which AI stories are worth covering more deeply.
Tell Pagish if this story was useful.