• eicker@lemmy.worldOP
    link
    fedilink
    English
    arrow-up
    209
    arrow-down
    4
    ·
    6 days ago

    Funny how AI has supposedly surpassed the greatest scientists, yet the greatest scientific discovery here is always announced by the guy selling subscriptions. If it really is that smart, perhaps it could explain to Sam the difference between a peer reviewed breakthrough and a product launch.

  • Avicenna@programming.dev
    link
    fedilink
    English
    arrow-up
    8
    ·
    edit-2
    4 days ago

    These math breakthroughs happened by throwing tens of thousands of agents at a problem to cover all possible angles (with tools developed by other mathematicians) costing the same amount of money as hiring ten top tier mathematicians for several years. Not saying it is not impressive, but not “AI can replace sciencetists” level impressive.

    • kadotux@sopuli.xyz
      link
      fedilink
      English
      arrow-up
      11
      ·
      5 days ago

      If he really believed what he said, he would. But he knows he’s just talking shit

  • rustydrd@sh.itjust.works
    link
    fedilink
    English
    arrow-up
    61
    arrow-down
    1
    ·
    6 days ago

    If it’s so smart, maybe they should ask it to come up with a viable business strategy, because he sure as hell hasn’t figured it out, lol.

    • isles@infosec.pub
      link
      fedilink
      English
      arrow-up
      19
      ·
      6 days ago

      “We have no current plans to make revenue. We have no idea how we may one day generate revenue. We have made a soft promise to investors that once we’ve built this sort of generally intelligent system, basically we will ask it to figure out a way to generate an investment return for you.” - Sam Altman

        • luciferofastora@feddit.org
          link
          fedilink
          English
          arrow-up
          2
          ·
          4 days ago

          They want to believe. They hope that, if they pump this enough, enough other people will chip in that they can make a return even if it crashes, but far more so, they want it to deliver. Imagine a super-intelligent money making machine, and you’re getting a big slice of the pie!

          They want to believe that they’ve found the free money hack, just like with Crypto.

          And just like Crypto, it’s mostly those that were rich already that’ll make bank selling shovels for the rest to dig their own graves with in search for gold.

  • The_greatest_ahole9@lemmy.world
    link
    fedilink
    English
    arrow-up
    5
    ·
    4 days ago

    Just because AI is smarter than Sam Altman doesn’t mean it’s smarter than top scientists. Especially considering Sam is just about as smart as a pile of horseshit.

  • CileTheSane@lemmy.ca
    link
    fedilink
    English
    arrow-up
    17
    arrow-down
    1
    ·
    5 days ago

    I believe AI is smarter than Sam Altman.

    Sam Altman believes he is smarter than scientists.

    • humorlessrepost@lemmy.world
      link
      fedilink
      English
      arrow-up
      3
      ·
      4 days ago

      I asked AI if it’s smarter than scientists, and it said no. So if Sam says I should trust AI, Sam is full of shit.

  • mcv@lemmy.zip
    link
    fedilink
    English
    arrow-up
    21
    ·
    5 days ago

    AI is not smarter than the best scientist. It’s just smarter than Sam Altman. But that’s partially because Sam Altman is regressing due to his over reliance on AI.

  • Treczoks@lemmy.world
    link
    fedilink
    English
    arrow-up
    6
    ·
    4 days ago

    Sam, just because the digital parrot outsmarts you, it does not need to be intelligent.

  • bashibazouk@lemmy.world
    link
    fedilink
    English
    arrow-up
    38
    arrow-down
    1
    ·
    6 days ago

    Ya, Sam, call me when the AI knows what questions to ask that it came up with on its own rather than just scraping…

  • tea_n_typewriters@lemmy.world
    link
    fedilink
    English
    arrow-up
    36
    arrow-down
    1
    ·
    6 days ago

    I’ll believe it when I hear it from the brightest human scientists, not the glorified salesman who would ironically be perfectly replaced by AI.

  • f314@lemmy.world
    link
    fedilink
    English
    arrow-up
    34
    arrow-down
    3
    ·
    6 days ago

    Is it because he thinks he is the brightest human scientist, but it turns out he’s actually pretty stupid?

    • skisnow@lemmy.ca
      link
      fedilink
      English
      arrow-up
      8
      arrow-down
      1
      ·
      6 days ago

      You can tell how competent someone is at something by how good they think an LLM is at that thing.

      • Seralth@piefed.seralth.com
        link
        fedilink
        English
        arrow-up
        3
        ·
        5 days ago

        Ai is pretty good at sucking proverbial and metaphorical dick.

        Also generating subtitles for my mother’s foreign films I guess?

        It’s painfully just ok at most tasks. It’s a someone useful tool at a local level with small models doing extremely narrow tasks.

        But anyone who thinks these giant models are remotely beyond just a glorified over priced hobby toy is… Questionably sane.

        They are fun to play with in the same sorta way playing with explosives is.

        • luciferofastora@feddit.org
          link
          fedilink
          English
          arrow-up
          1
          ·
          4 days ago

          Also generating subtitles for my mother’s foreign films I guess?

          You mean mapping learned patterns (speech) to other patterns (words)? Yes, that’s actually what it’s really good at.

          Of course, if those films invented completely new words, it would be out of its depth. Weird pronunciations may also produce weird text. It might get things wrong either way occasionally.

          But if it’s within the boundaries of its training material, it’ll generally produce reasonable output, and in that context, the occasional aberration doesn’t matter as much (particularly if the human brain can compensate).

          Just don’t ask it to generate anything where factuality matters.

    • Zwuzelmaus@feddit.org
      link
      fedilink
      English
      arrow-up
      1
      ·
      6 days ago

      He is anything else than stupid. Actually dangerous (unless they finally stop throwing insane amounts of money at him year after year)

    • phlegmy@sh.itjust.works
      link
      fedilink
      English
      arrow-up
      7
      ·
      5 days ago

      But CEO’s are much more smarterer than scientists!
      If scientists are so smart, why don’t they have any money?
      Don’t tell me they value something other than “number go up” because I won’t understand you.