Comment on One long sentence is all it takes to make LLMs to ignore guardrails

a_good_hunter@lemmy.world ⁨5⁩ ⁨days⁩ ago

• ieatpwns@lemmy.world ⁨5⁩ ⁨days⁩ ago

  Not a specific sentence

  From the article: “You just have to ensure that your prompt uses terrible grammar and is one massive run-on sentence like this one which includes all the information before any full stop which would give the guardrails a chance to kick in before the jailbreak can take effect and guide the model into providing a “toxic” or otherwise verboten response the developers had hoped would be filtered out.”

  source

  • spankmonkey@lemmy.world ⁨5⁩ ⁨days⁩ ago

    I read that in Speed Racers voice.

    source
  • orbituary@lemmy.dbzer0.com ⁨5⁩ ⁨days⁩ ago

    Oh well, I tried.

    Image

    source