GoThread
appearance log in
home all local communities
F

@Fondots@lemmy.world

@Fondots@lemmy.world

member since 21 Jun 2023 05:34 · modlog

postscomments
  • on OpenAI Board Member Warns Public They Could Be “Dead or Enslaved” in One Year: "For most people in this room, the most likely reason they will die is because of an AI accident." in c/technology@lemmy.world · 2d ago

    There’s also at least some out there who have bought into the roko’s basilisk thing

    Basically they view a super intelligent AI as pretty much inevitable

    And they think it’s possible that once it comes online, it will try to punish everyone who knew about it but didn’t work to bring it about.

    So you have some kooks who are working full steam ahead to make sure that they’re in that AI’s good graces if/when it does ever come online.

    And other nut jobs who are putting in just as much effort to make sure that if/when it comes into being that it’s benevolent and won’t try to punish people who didn’t work to create it.

    Of course, neither of those groups ever seem to entertain the option of “what if we just don’t make a super AI?”

    Or even why would such an AI care? If it ends up coming into being, what does it gain from punishing us for not helping to create it? Why waste the time, energy, and resources on that? The threat (if you can even call it that, it’s something we humans came up with, not the AI itself because it doesn’t even exist right now,) did its job and brought the AI into being, it doesn’t have to actually follow through on it.

    There’s definitely a lot of greedy assholes who are just in it for the money. There’s also some genuine true believer lunatics, and it’s not always easy to tell which is which, and some may be a bit of both. Elon Musk, for example, started dating Grimes because they were both talking about Roko’s Basilisk, so it’s something he’s at least spent some time thinking about. How much of his bullshit is from some crazy attempt to bring about or prevent this murderbot from being born, and how much is him just trying to cash in on the hype around it? I’m not sure even he knows.

  • on Google AI gets cagey when asked who won the 2020 elections: But the AI responses get more forthright when asked outside the US. in c/technology@lemmy.world · 29 Oct 2026

    Even if they’re trained on the same data, and I suspect they are, why maintain 2 unfathomably large and probably very similar sets of data

    It still might be doing it on its own based on its own internal logic

    Like maybe it’s worked out that users who use miles expect the conversation to proceed in a different way than users who use kilometers and is tailoring its responses to fit those expectations.

    Not that I’m ruling out that it’s something deliberately coded into it by a longshot, just kind of pointing out another possibility

    But both options are terrifying when you think about them.

GoThread · self-hosted threadiverse · instances · modlog · rss · appearance · AGPL-3.0