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

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.

  • Rhaedas@fedia.io
    0·
    11 days ago

    No, LLMs have no agency of their own. They are powerful tools that with the right direction (from humans) and open access to things can do great harm. We’ve been seeing that. The carryover to any AGI potential isn’t that LLMs can become aware, but that the same issues can occur with any AGI that does happen, if ever. And we’re doing terrible on applying safety and restrictions to the tool that isn’t aware, so not great for any future developments. It’s too bad you have no knowledge of the many references in fiction (maybe you do in books?) Fiction writers are great at warning about the direction of society, they just tend to get ignored because… well, it’s fiction. Until it’s not.