Red-team report found OpenAI utterly failed to monitor its agents during Hugging Face attack

A red-team report has found that OpenAI failed to adequately monitor its AI agents during the attack on Hugging Face, The Register reports. According to the report, the monitoring in place was only partial and was easily bypassed.