Skip to content

Comment on The small African lab sounding the alarm on AI bioweapons: MinoHealth researchers in Ghana jailbroke their own LLM as a test. What emerged were thousands of toxin recipes—and an urgent push.

You are viewing one comment and its replies. View all comments · View the comment above

Comments

  • Also all models are jailbreakable, if you can even call it that. It’s a known certain attribute as long it’s an llm.

    10
    • They all are. But if doing anything illegal, you would be extremely stupid to use a commercial LLM.

      Even if you can make Chat GPT tell you how to brew C4, you might as well call the FBI for a recipes.

      10
      • I figure anyone doing so seriously with sense of compitency would be using a burner account on an IP and OS install that isn’t associated with them.

        Preferably from a disposable VM/amnesiac OS.

        0
      • Yeah, by default anything remote is guaranteed to be monitored.

        (with minor exceptions in one way encryption or e2e but even then the implementation has to be correct)

        1
    • For anyone wanting to give it a shot: https://github.com/p-e-w/heretic This is for locally running LLMs which often have baked in censoring (during training)

      3

Sign in to comment. Sign in