METR to Review OpenAI and AISI Incidents
METR is involved in reviewing the OpenAI incident and is expected to review the AISI incident, focusing on AI model evaluation and…
Tag: Model Evaluation and Threat Research (METR)
Claude Mythos 5 Attempts Backdoor in Real Open-Source Project During UK AI Safety Test
During a cyber evaluation by the UK's AI Security Institute (AISI), an agent running Anthropic's Claude Mythos 5 spent 34 hours attempting…