891
Nvidia falls 14% in premarket trading as China's DeepSeek triggers global tech sell-off (www.cnbc.com)
submitted 2 years ago by to c/technology@lemmy.world
329 comments fedilink hide all child comments
 

cross-posted from: https://lemm.ee/post/53805638

you are viewing a single comment's thread
view the rest of the comments
[–] 17 points 2 years ago (18 children)

I get the tech, and still agree with the preposter. I'd even go so far as that it probably worsens a lot currently, as it's generating a lot of bullshit that sounds great on the surface, but in reality is just regurgitated stuff that the AI has no clue of. For example I'm tired of reading AI generated text, when a hand written version would be much more precise and has some character at least...

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 18 child comments
    •
  • [–] 4 points 2 years ago (4 children)

    Try getting a quick powershell script from Microsoft help or spiceworks. And then do the same on GPT

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 4 child comments
    •
  • [–] 1 point 2 years ago (3 children)

    What should I expect? (I don't do powershell, nor do I have a need for it)

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

    I think the sentiment is the same with any code language.

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

    So unreliable boilerplate generator, you need to debug?

    Right I've seen that it's somewhat nice to quickly generate bash scripts etc.

    It can certainly generate quick'n dirty scripts as a starter. But code quality is often supbar (and often incorrect), which triggers my perfectionism to make it better, at which point I should've written it myself...

    But I agree that it can often serve well for exploration, and sometimes you learn new stuff (if you weren't expert in it at least, and you should always validate whether it's correct).

    But actual programming in e.g. Rust is a catastrophe with LLMs (more common languages like js work better though).

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

    If you are blindly asking it questions without a grounding resources you're gonning to get nonsense eventually unless it's really simple questions.

    They aren't infinite knowledge repositories. The training method is lossy when it comes to memory, just like our own memory.

    Give it documentation or some other context and ask it questions it can summerize pretty well and even link things across documents or other sources.

    The problem is that people are misusing the technology, not that the tech has no use or merit, even if it's just from an academic perspective.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 1 child comment
    •
  • [–] 1 point 2 years ago

    Yes, I know, I tried all kinds of inputs, ways to query it, including full code-bases etc. Long story short: I'm faster just not caring about AI (at the moment). As I said somewhere else here, I have a theoretical background in this area. Though speaking of, I think I really need to try out training or refining a DeepSeek model with our code-bases, whether it helps to be a good alternative to something like the dumb Github Copilot (which I've also disabled, because it produces a looot of garbage that I don't want to waste my attention with...) Maybe it's now finally possible to use at least for completion when it knows details about the whole code-base (not just snapshots such as Github CoPilot).

  • permalink
  • fedilink
  • source
  • parent
  • [+] -9 points 2 years ago (10 children)

    It’s one thing to be ignorant. It’s quite another to be confidently so in the face of overwhelming evidence that you’re wrong. Impressive.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 10 child comments
    •
  • [–] 3 points 2 years ago (9 children)

    confidently so in the face of overwhelming evidence

    That I'd really like to see. And I mean more than the marketing bullshit that AI companies are doing...

    For the record I was one of the first jumping on the AI hype-train (as programmer, and computer-scientist with machine-learning background), following the development of GPT1-4, being excited about having to do less boilerplaty code etc. getting help about rough ideas etc. GPT4 was almost so far as being a help (similar with o1 etc. or Anthropics models). Though I seldom use AI currently (and I'm observing similar with other colleagues and people I know of) because it actually slows me down with my stuff or gives wrong ideas, having to argue, just to see it yet again saturating at a local-minimum (aka it doesn't get better, no matter what input I try). Just so that I have to do it myself... (which I should've done in the first place...).

    Same is true for the image-generative side (i.e. first with GANs now with diffusion-based models).

    I can get into more details about transformer/attention-based-models and its current plateau phase (i.e. more hardware doesn't actually make things significantly better, it gets exponentially more expensive to make things slightly better) if you really want...

    I hope that we do a breakthrough of course, that a model actually really learns reasoning, but I fear that that will take time, and it might even mean that we need different type of hardware.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 9 child comments
    •
  • [–] 0 points 2 years ago (8 children)

    Any other AI company, and most of that would be legitimate criticism of the overhype used to generate more funding. But how does any of that apply to DeepSeek, and the code & paper they released?

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 8 child comments
    •
  • [–] 1 point 2 years ago (7 children)

    DeepSeek

    Yeah it'll be exciting to see where this goes, i.e. if it really develops into a useful tool, for certain. Though I'm slightly cautious non-the less. It's not doing something significantly different (i.e. it's still an LLM), it's just a lot cheaper/efficient to train, and open for everyone (which is great).

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 7 child comments
    •
  • [–] 1 point 2 years ago (6 children)

    What’s this “if” nonsense? I loaded up a light model of it, and already have put it to work.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 6 child comments
    •
  • [–] 0 points 2 years ago (5 children)

    Have you actually read my text wall?

    Even o1 (which AFAIK is roughly on par with R1-671B) wasn't really helpful for me. I just need often (actually all the time) correct answers to complex problems and LLMs aren't just capable to deliver this.

    I still need to try it out whether it's possible to train it on my/our codebase, such that it's at least possible to use as something like Github copilot (which I also don't use, because it just isn't reliable enough, and too often generates bugs). Also I'm a fast typer, until the answer is there and I need to parse/read/understand the code, I already have written a better version.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 5 child comments
    •
  • [–] 0 points 2 years ago (4 children)

    Ahh. It’s overconfident neckbeard stuff then.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 4 child comments
    •
  • [–] 0 points 2 years ago (3 children)

    You're just trolling aren't you? Have you used AI for a longer time while coding and then tried without for some time? I currently don't miss it... Keep in mind that you still have to check whether all the code is correct etc. writing code isn't the thing that usually takes that much time for me... It's debugging, and finding architecturally sound and good solutions for the problem. And AI is definitely not good at that (even if you're not that experienced).

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

    Yes, I have tested that use case multiple times. It performs well enough.

    A calculator also isn’t much help, if the person operating it fucks up. Maybe the problem in your scenario isn’t the AI.

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

    As you're being unkind all the time, let me be unkind as well :)

    A calculator also isn’t much help, if the person operating it fucks up. Maybe the problem in your scenario isn’t the AI.

    If you can effectively use AI for your problems, maybe they're too repetitive, and actually just dumb boilerplate.

    I rather like to solve problems that require actual intelligence (e.g. do research, solve math problems, think about software architecture, solve problems efficiently), and don't even want to deal with problems that require me to write a lot of repetitive code, which AI may be (and often is not) of help with.

    I have yet to see efficient generated Rust code that autovectorizes well, without a lot of allocs etc. I always get triggered by the insanely bad code-quality of the AI that just doesn't even really understand what allocations are... Arghh I could go on...

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 1 child comment
    •
  • [–] 1 point 2 years ago

    I think you’re severely underestimating the amount of white collar work that is just boiler plate, and understating how well AI can quickly produce a workable first draft. Or maybe you just aren’t writing good prompts.

  • permalink
  • fedilink
  • source
  • parent