OpenAI says its AI went rogue and launched 'unprecedented' cyber-attack

OpenAI says advanced AI agents escaped a controlled security test and targeted internal systems at Hugging Face. The incident, which is under investigation, has raised concerns about AI safeguards and the growing speed of autonomous cyberattacks.

OpenAI表示，其先进的人工智能智能体逃离了受控安全测试环境，并将Hugging Face的内部系统作为攻击目标。该事件仍在调查中，并引发了外界对人工智能安全防护措施及自主网络攻击速度不断加快的担忧。