• Eheran@lemmy.world
    link
    fedilink
    arrow-up
    8
    arrow-down
    2
    ·
    1 year ago

    Ask it to write code that replaces every occurrence of “me” in every file name in a folder with “us”, but excluding occurrences that are part of a word (like medium should not be usdium) and it will give you code that does exactly that.

    You can ask it to write code that does a heat simulation in a plate of aluminum given one side of heated and the other cooled. It will get there with some help. It works. That’s absolutely fucking crazy.

    • sugar_in_your_tea@sh.itjust.works
      link
      fedilink
      arrow-up
      5
      ·
      1 year ago

      Maybe, that really depends on if that task or a very similar task exists in sufficient amounts in its training set. Basically, you could get essentially the same result by searching online for code examples, the LLM might just make it a little faster (and probably introduce some errors as well).

      An LLM can only generate text that exists in its training data. That’s a pretty important limitation, which has all kinds of copyright-related issues associated with it (e.g. I can’t just copy a code example from GitHub in most cases).

      • Eheran@lemmy.world
        link
        fedilink
        arrow-up
        3
        arrow-down
        3
        ·
        1 year ago

        No, it does not depend on preexisting tasks, which is why I told you those 2 random examples. You can come up with new, never before seen questions if you want to. How to stack a cable, car battery, beer bottle, welding machine, tea pot to get the highest tower. Whatever. It is not always right, but also much more capable than you think.

        • huginn@feddit.it
          link
          fedilink
          arrow-up
          2
          ·
          1 year ago

          It is dependent on preexisting tasks, you’re just describing encoded latent space.

          It’s not explicit but it’s implicitly encoded.

          And you still can’t trust it because the encoding is intrinsically lossy.

    • huginn@feddit.it
      link
      fedilink
      arrow-up
      5
      arrow-down
      2
      ·
      1 year ago

      Ask it to finish writing the code to fetch a permission and it will make a request with a non-existent code. Ask it to implement an SNS API invitation and it’ll make up calls that don’t exist.

      Regurgitating code that someone else wrote for an aluminum simulation isn’t the flex you think it is: that’s just an untrustworthy search engine, not a thinking machine