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.

  • chickenf622@sh.itjust.worksEnglish
    0·
    6 days ago

    This is just marketing bullshit as OpenAI tries to hype themselves up for their IPO. Anytime an “AI” company says their models are too dangerous or did something like this it’s just to make people think the “AI” is way more powerful/intelligent than it actually is.

    • DGen@piefed.zipEnglish
      0·
      6 days ago

      Exactly.

      If you own Something “Dangerous”, would you Just warn others or just let it happen?

      If AI does this or that random Bad Thing, they wanted it to happen. They are having their hand on that big fat Red hot button saying “No”. Its a decision.