632
Weird Al Yankovic Turned Down “Nice Pile Of Money” For AI Ad: “I Can’t Be The Poster Boy For AI” (deadline.com)
submitted 1 month ago by to c/technology@lemmy.world
61 comments fedilink hide all child comments
you are viewing a single comment's thread
view the rest of the comments
[+] -7 points 1 month ago (2 children)

Alright please forgive me for this one transgression of using Gemini, but this is one statistical likelihood that feels creepy

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 4 child comments
    •
  • [–] 20 points 1 month ago (1 child)

    Your transgression is completely unforgiven

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 8 points 1 month ago (2 children)

    AI has evolved to the point where the first task is to manage the emotions of the human. And then hallucinate a proper sounding response.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 4 child comments
    •
  • [–] 9 points 1 month ago (1 child)

    I hate the way A.I. condescendingly talks down to me like I'm on the verge of hysterics, and then butters me up to try and make me feel good. It's so manipulative and gross and practically negging + love bombing. I try to avoid interacting with them if possible.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] -4 points 1 month ago (1 child)

    you can literally change how it talks to you in the settings...

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 1 point 1 month ago (1 child)

    In my experience that just changes the "flavour of badness" rather than making it actually good. For example, if I made it "direct", it would incessantly nitpick everything I say and tell me I'm wrong and stupid constantly, even when I'm absolutely correct. It would just make up fake facts to prove me wrong. So basically Reddit argument confidently incorrect mode. I'm not surprised people get psychosis from those things, it's probably about as good for your mental health as interacting with narcissists.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] -5 points 1 month ago (1 child)

    im not downvoting you

    but this is just stupid, there’s literally a billion people who use it daily

    claude has none of this nonsense you speak of

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 1 point 1 month ago (1 child)

    I remove my own upvote from my comments because I like things starting at 0.

    Also, billions eat fast food every day, doesn't make it good for you. I've tried all the A.I., because I believe in order to hate something, you must understand it first. Claude kept trying to do everything for me, even when I didn't want it to. As a hypothetical example, if I asked it to explain a math problem without giving me the answer, it would give me the answer anyway. I don't want something to help me that much, I like doing things mostly myself, with some assistance if needed. It causes damage to your skills, independence, and sense of value and worth in your work, if something is just like, "hey let me just quickly trivialise your entire idea and do it for you".

    From my experience every single A.I. is toxic or weird or subtly damaging to interact with, without exception. There's just something unnatural about the way it communicates.

    I won't blame anyone for using it, that's their own choice. But I really try not to for my own health.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 1 point 1 month ago

    That's funny I do the same thing but have different results to you

    That issue with ai giving the answer away was still a feature in the local models but in the last 6 months with Qwen and Gemma being released is way better

    I would ask things like 'Give me a french word to practice and don't tell me the translation' and it would say:

    Try translating Bonjour! (Hello)

    And I would have to clarify and then it would forget it was the pits but this has stopped happening with later local models like Qwen and Gemma

    Maybe give Claude another go, Sonnet 5 just came out and be strict with your question: say tell me how to answer this problem but you MUST NOT tell me the answer, you can use other examples but not this one, etc

    and see how you go

  • permalink
  • fedilink
  • source
  • parent