Comment on Taco Bell rethinks AI drive-through after man orders 18,000 waters

• Khanzarate@lemmy.world ⁨14⁩ ⁨hours⁩ ago

  They do, my concern is more about if that JSON is correct, not just well-formed.

  Also, 18000 waters might be correct JSON, but makes an AI a bad cashier.

  source

  • staph@sopuli.xyz ⁨14⁩ ⁨hours⁩ ago

    There is a lot more that goes into it than just being correct. 18000 waters may have been the actual order, because somebody decided to screw with the machine. A human who isn’t terminally autistic would reliably interpret that as a joke and would simply refuse to punch that in. The LLM will likely do what a human tells it to do.

    source

    • Khanzarate@lemmy.world ⁨12⁩ ⁨hours⁩ ago

      Thats part of correctness to me, delivering an order that taco bell actually would make is important.

      Semantics aside, though, we agree. That’s very important.

      source
    • tomiant@programming.dev ⁨10⁩ ⁨hours⁩ ago

      So they just trim the instructions so it doesn’t take joke orders, so it can make more reasonable decisions, like:

      “May I take your order?” “Two double whoppers with extra mayo and a chocolate banana sundae” “Oh you’ve GOTTA be joking!”

      source

      • staph@sopuli.xyz ⁨10⁩ ⁨hours⁩ ago

        It’s trivial to get LLMs to act against the instructions

        source