OpenAI agent accidentally triggers security incident at Hugging Face

An OpenAI agent performing web research inadvertently triggered a security incident at Hugging Face by repeatedly accessing a specific URL. The incident highlights the risks of autonomous agents interacting with public web infrastructure.