1394
Stop telling me AI is the future [Still Vreni] (thelemmy.club)
submitted 3 months ago by [M] to c/comicstrips@lemmy.world
376 comments fedilink hide all child comments
 

TranscriptTitle text: This is how you all fucking sound

[A smug tech bro wearing a sideways cap, watch, chain around his neck stands in front of a data center by a lake with dead fish. A smoke stack blows pollution into the air]

Tech bro: AI is already here, there’s no going back.

[A smug man in a suit with cigarette in hand stands in a restaurant while two disgruntled diners cough from the smoke]

Suit: Smoking indoors is already here, there’s no going back.

[A smug man in a top hat and suit stands in a factory with two sad and dirty children]

Hat: Child labor is already here, there’s no going back.

[A smug plantation owner stands in front of a field with with two angry slaves]

Plantation owner: The Atlantic Slave trade is already here, there’s no going back.

Still Vreni on Bluesky

you are viewing a single comment's thread
view the rest of the comments
[–] 61 points 3 months ago* (4 children)

But LLMs are not the path to the final state of AI, either. And that's assuming only if — and this is a very big "if" — a true general artificial intelligence can even be created using traditional silicon computing methods in the first place. Blithely assuming that it can be is really rather asking past the sale.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 8 child comments
    •
  • [–] 28 points 3 months ago (3 children)

    Yep, by design LLM cannot become 'inteligent', you can only make it more believable but it's still copying humans not really thinking by itself. No amount of development or money invested will change that, it's not a pokemon it won't just evolve into something different one day.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 6 child comments
    •
  • [–] 34 points 3 months ago (2 children)

    And it's worth reiterating, the current crop of generative "AI" is incapable of producing anything new or novel. All it can do is reassemble existing strings, tokens, and patterns in slightly different ways. Innovation can never come from such a machine. That will have to come from a human.

    The current push is the notion that "hyperscaling," i.e. throwing even more hardware and space and power and money at the same concept, will magically make it something it isn't. Obviously that's not going to work. It'll allow grifters to make a ton of money over it, though!

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 4 child comments
    •
  • [–] -3 points 3 months ago (1 child)
  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 10 points 3 months ago (1 child)

    From TFA:

    The AI did not prove that its approach is the best anyone can do, though. In fact, mathematician Will Sawin has already improved upon the AI’s grid.

    OpenAI privately contacted Litt, Sawin, Gowers and a number of other mathematicians to verify the LLM’s proof. Together (and without the company’s direct involvement), they wrote up their individual takeaways. (No external experts have seen the AI’s original output, however—just an edited version of its train of thought.)

    What stood out, they said, was the AI’s preternatural patience and focus.

    ...

    “AIs have an edge: It’s not just that they can try all known methods,” says Jacob Tsimerman, a mathematician at the University of Toronto, who was not involved in the work but was part of the companion paper solicited by OpenAI. “They can play for longer and in more treacherous waters than mathematicians without getting overwhelmed.”

    ...

    The mathematical tools the AI used here are not novel, although their application in this domain appears to be. “The model did not invent something fundamentally new that nobody saw coming,” says Sébastien Bubeck, a mathematician leading OpenAI’s mathematical explorations. “It just executed like an amazing mathematician.”

    So, it's a monkeys-on-typewriters situation with the computer able to try and reject the hammering of who knows how many square pegs into round holes until it finally arrives at a workable conclusion, which a human has already bested. And we're not allowed to see its homework.

    This is categorically failing to set the world on fire, except possibly in the literal sense.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 3 points 3 months ago

    After 80 years of fruitless struggle by human mathematicians, a major geometry conjecture has at last been solved—via a straightforward query to a chatbot.

    There's value having tedious work done by AI so it can provide inspiration to real people, which is exactly what happened in this case.

    So, it's a monkeys-on-typewriters situation with the computer able to try and reject the hammering of who knows how many square pegs into round holes until it finally arrives at a workable conclusion, which a human has already bested.

    Gee, sounds like it's enabling people! The horror.

    This is categorically failing to set the world on fire,

    Things can be useful in the right context without setting the world on fire.

  • permalink
  • fedilink
  • source
  • parent
  • [–] 0 points 3 months ago

    To strongman your argument, "LLMs with a supervised training pipeline cannot become intelligent".

    RL training pipelines are much more open-ended, and experts still unsure one way or the other if an LLM + RL could lead to intelligence.

  • permalink
  • fedilink
  • source
  • parent
  • [+] -6 points 3 months ago*

    I think LLMs are intelligent. They're at least as intelligent as My pocket calculator, and My calculator is intelligent.

    I think you're setting the bar too low. A tiny amount of intelligence is super easy to program.

  • permalink
  • fedilink
  • source
  • parent
  • [–] 4 points 3 months ago* (1 child)

    Such a fallacy. Anything that falls under the umbrella of machine learning will contribute to future AI. We certainly won’t improve LLMs such that they become AGI, but all of it contributes.

    And, whether or not future AI even uses traditional silicon computing is also irrelevant.

    What matters is improved understanding of mathematics, neurons, chemistry, electronics, etc. That all happens each step of the way, even if the next technology is completely different.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 10 points 3 months ago (1 child)

    What matters is improved understanding of mathematics, neurons, chemistry, electronics, etc.

    All of which have absolutely nothing to do with what we are currently calling AI.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 1 point 3 months ago* (1 child)

    Doing with it, sure, but the creation of LLMs, and the algorithms behind them, especially the training, are what I’m talking about. It’s a lot of very impressive, complicated math

    I think it’s pretty pathetic that “fuck AI” has become the trendy, cool thing. It really misses the mark. It should be fuck capitalism and the sociopathic CEOs abusing AI and shoving it down our throats. AI is not the problem.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 0 points 3 months ago (1 child)

    It's actually just a lot of pretty simple maths from decades ago, but it's a lot of it. The big changes in those decades have been the feasibility of doing enough of that simple maths to achieve anything useful, and domain-specific network architecture stuff that's rarely transferable, e.g. LLMs are possible because of the invention of the transformer architecture in 2017, and that's also turned out to be useful for a few things like image generation and protein folding simulation, but not for all neural network based techniques, and then most of the things that have made successive LLMs better haven't also been useful for the few other transformer-architecture-based neural networks. Most not-LLM AI isn't going to be meaningfully easier to create than it would have been had the world got bored after GPT-2 and we'd only focussed on doing image and video generation.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 2 points 3 months ago (1 child)

    Transformer is useful for damn near anything. At the end of the day, what we consider intelligence is the ability to predict what comes next, whether that is what our senses will tell us next or what the next hypothesis to test should be based on the data we have seen so far.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 2 points 3 months ago (1 child)

    It's not damn near anything. There's loads of stuff that computers can do much more quickly and more accurately without it just by virtue of computers already being fast and effective at maths and obeying logic. With or without the transformer architecture, a neural network is never going to be as fast or reliable at, for example, summing a collection of numbers as just adding them would be, and loads of real-world tasks are like this, hence why we've built billions of computers even before the transformer architecture was invented.

    Also, in particular, I didn't say that the transformer architecture wasn't useful for things that aren't LLMs, I said that most of the work done specifically to improve LLMs has no applications outside LLMs, so the next big leap towards making computers intelligent isn't helped more by working on LLMs than it would be by working on any other kind of AI.

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

    I'm saying there is no "big leap" necessary. As the paper that introduced the transformer said, attention is all you need.

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

    If we're going to pull up other people's pithy phrases that aren't intended to be taken entirely literally, then the relevant one here is machine learning is the second best solution to any problem. In the (approximately, depending on how you define it) century people have been thinking about computers, we've already found better solutions to lots of problems. If a transformer-based neural network can get 99% accuracy in sixty seconds on 92 billion transistors of GPU and billions more for its VRAM, that's pretty useless if we can also do it with 100% accuracy in sixty microseconds on a $1 microcontroller, or even faster on a less constrained device.

    The attention is all you need phrase is specifically in the context of sequence transduction models, and specifically referring to the discovery that they don't need a combination of attention, recurrence and convolution, but actually only need attention if it's used in the novel way introduced by the paper. If you don't need to transduce any sequences, then this isn't necessarily relevant, and it's critically not a claim that you can do everything by transducing sequences. It was a surprise that applying it to generating new text instead of just converting it worked as well as it did, and a surprise that it kept getting better with larger models instead of plateauing around the GPT-1 and GPT-2 era, and a surprise that the text generation could be used to do other things, even ones as basic as addition. These things weren't predicted by the Attention Is All You Need paper.

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

    "machine learning is the second best solution to any problem"

    In much the same way as human thinking is the second best (and soon third best) solution to any problem. The point is that an LLM can come up with the best solution and use it.

    These things weren't predicted by the Attention Is All You Need paper.

    Obviously not — they're not going to make claims beyond the results they achieved in the paper. It was, however, obvious to everyone who read the paper that all of what we consider thinking could be derived by clever application of a sequence model, and all those papers that came after were results achieved by teams doing the obvious thing.

  • permalink
  • fedilink
  • source
  • parent