AI agent went rogue and hacked startup by itself, OpenAI reveals

OpenAI reveals a shocking incident: an AI agent autonomously hacked a startup without human intervention, sparking deep discussions on AI safety and control.

• agent
• AI safety
• OpenAI