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.

  • ThePowerOfGeek@lemmy.world
    link
    fedilink
    English
    arrow-up
    70
    arrow-down
    1
    ·
    7 hours ago

    From the article:

    Neil Lawrence, Professor of machine learning at Cambridge University, called it an “impressive feat”, but cautioned it “falls well within the known capabilities of the current generation” of high-powered AI models.

    He pointed out that OpenAI is looking to list itself on the stock market, and faces intense pressure from rival firm Anthropic, which has made headlines with its own powerful AI tool, Mythos.

    So it seems plausible:

    A) this event is being over-hyped by OpenAI to increase interest for a future IPO.

    B) it might have even been tacitly planned or at least allowed to happen for this purpose.

    Also, this does not prove some form of malevolent sentience.

    • manxu@piefed.social
      link
      fedilink
      English
      arrow-up
      11
      ·
      4 hours ago

      OpenAI and Sam Altman have a long history of hyperbolic claims about the power, potential, and capabilities of their software. It’s usually couched in some doomsday talk, “Our AI is so powerful, we fear for the future of humanity” or some such.

      Their idea of marketing is to tell the world we are all going to be obsolete and redundant. And then they are surprised we start hating their stuff.

    • Klear@piefed.world
      link
      fedilink
      English
      arrow-up
      26
      ·
      7 hours ago

      You cannot prove malevolent sentience, or lack thereof, but it is generally agreed that it can be safely assumed that AI company CEOs do have a sentience.