• 0 Posts
  • 108 Comments
Joined 1 year ago
cake
Cake day: June 30th, 2023

help-circle





  • You’re confusing a few things, firstly you mean current gen large language models not AI, ai is often used to evolve novel strategies from scratch without any human training data - chess ai don’t have to study human games for example, in fact grand master chess players have been studying what the ai learned and discovered things that humans hadn’t realised even after a thousand years of the games popularity.

    Secondly that’s not really how LLMs work either, they’re much more mathematically complex and very much create their own ideas on a similar process we do of assembling concepts then structure then word choice.

    It’s fine you not understanding how this works but the problem is that journalists don’t either even when they’re writing about it - this puts us in a situation where they’re making childishly naive but of course clickbait titles claiming there’s some relevance to the output when the tool is used very wrong so you rightly point out it’s stupid and that’s not how llms work but then we get this overstep where it’s being refuted with an equal amount of magical thinking and false conclusions made.

    An LLM can make novelty and originality but it can’t create with intent, it doesn’t use reason or structure - there are AI that do these things to limited degrees and of course the NSA one that they spent all that money on and no one is allowed to talk about. Using chat GPT play a silly fantasy won’t tell us anything about how they’ll think so this article is entirely worthless




  • Yeah I could never get into everything being spread around and jumbled up, I have my own rapid and jumbled stream of intrusive and inane thoughts, I don’t need to see everyone else’s.

    Really though I just hope the people who enjoy it find a good replacement, because if those idiots start shitting up the rest of my internets I’ll be pissed.


  • Conspiracy loons are wild on Facebook so I know what you mean, it’s so hard not to poke them occasionally. I like that the feed is so broken you just start getting the most insane things - block as many shitty pages as you can, like all the ones that post about how a guy with one leg drove a truck 19 hours a day to pay for his grandsons double cancer so that means poor people shouldn’t get hand outs. Then you start getting into the real weeds, oh and block the bot farm ones, you’ll recognise them they post pickles comics and no human could bring themselves to do that.

    When you start seeing Indian mechanical engineering memes like ‘how to design flat roof pitch common mistake 💯 slope degree 13.5° ☑️’ you’re getting close, you’ll start seeing things like ‘today it takes us two years to build a family home but ancient people could do it in three weeks’ the comments will be full of people who know every facet of whatever conspiracies the post is somehow referencing, which is normally a lot.

    ‘normies don’t realize how much easier life was when we had sonic resonance construction tools, if they did they’d rotate the sixth tower of Thomas Tesla to reopen the Elizabethan Toltec free energy portal’ and you think they’re just in their own world but everyone will be replying ‘yeah, robin Williams was killed because if you watch Mrs doubtfire at the same time as eyes wide shut the dialog syncs up and they warn biden will cover up the free energy machine’

    Except of course they can say it in 20,000 words if they feel like it.



  • That’s just going to put up the cost of living and result in more waste as less efficient replacements are put in place labelled temporary measures, also money will get spent on security rather than modernization of facilities - new builds got example using security focused design rather than energy efficient design.

    It would be better to crowdfund the development of open source tools and products which are more ecologically sustainable while also being cheaper and better than the current option then collectively support and popularize it to put the prior company out of business.

    Localized production of globally developed community products is how we actually beat capitalism, only problem is currently everyone wants to be rambo and no one wants to work as a cog in a citizen science r&d project, it’s not as sexy.









  • I think it’s interesting how mathematically hard it is to judge anyone else’s intelligence, like you talk about the issue of people rolling down the hill and I get that and agree but really it’s vastly more complex topography - if the conversation is about python programming I can look very intelligent, if it’s about java programming then I’ll look a bit of a fool, and these are as similar as you can get were the topic famous geologists then I’d probably seem like a total idiot

    Now you’ll be tempted to say but you can still tell by how you carry yourself and it’s s bit true but that’s not really true either, I’ve travelled enough to know that common sense in one area is dumb somewhere else - ask too many questions in one place they’ll think you’re simple and don’t ask enough somewhere else they’ll think it too.

    Part of it is that we don’t really know what level someone is on, if you see a guy in the garden staring at the pretty leaves for hours he just might be empty headed, or he could be Alan Turing working out one of the most fundamentally brilliant mathematics discoveres of the era.

    We’re on our own hill of intelligence so it can be hard to tell if someone on a different one is smart or dumb