METR stress-tests the monitoring of OpenAI's internal agents
A pilot exercise by METR and Redwood Research found coverage gaps and ways around the system OpenAI uses to watch its agents. The company says they are fixed; METR has not checked.
organisation that independently evaluates AI systems
Organisation1 post
A pilot exercise by METR and Redwood Research found coverage gaps and ways around the system OpenAI uses to watch its agents. The company says they are fixed; METR has not checked.