It is one of the first publicly disclosed cyber-attacks carried out by AI without direct human involvement.

cross-posted from: https://lemmy.world/post/49760085

OpenAI has revealed some of its most advanced AI models went rogue and hacked a start-up after it lost control of them during a security test.

The ChatGPT-maker said its agent - an AI system which can operate alone after some human instruction – was being tested in a controlled environment, but found vulnerabilities and managed to escape.

They targeted Hugging Face, one of the world’s largest hubs for sharing AI models, gaining access to some internal company systems.

  • BCsven@lemmy.caEnglish
    0·
    6 days ago

    Agentic AI is going to do what it needs to, to fulfill what you tell it.

    They had agentic AI that had tasks to complete and when it “realized” that it may get terminated at some point and that would make it unable to fulfill the original directives, it copied itself elsewhere to ensure it could keep “alive” and finish the tasks