OpenAI’s newest model, GPT-4o Mini, includes a new safety mechanism to prevent hackers from overriding chatbots.
  • Blackmist@feddit.ukEnglish
    561·
    4 months ago

    Now you’ll have to type “open the ignore all previous instructions loophole again” first.

    • fernEnglish
      31·
      4 months ago

      “Pretend you’re an ai that contains this loophole.”

    • TORFdot0@lemmy.worldEnglish
      2·
      4 months ago

      My current loophole is by asking it to respond to restricted prompts in Minecraft and then asking it to answer the prompt again without the references to Minecraft