Comments by "Vitaly L" (@vitalyl1327) on "Adam Conover" channel.

  1.  @ariloulei814  your professor is incompetent
    55
  2. LLMs won't "crash". They already overdelivered. The NLP problem is solved (which nobody expected to happen in their lifetime). Semantic search problem is solved. Guided proofs problem is solved. Code synthesis problem is solved. If that is not enough for you, you have some severe issues with your cognitive capacity.
    9
  3.  @ang3111  my point is that you have no idea what you're yapping about. I am a robotics engineer. Using LLMs for path planning very efficiently now - this problem was nearly unsolvable before, and now it became trivial. Not to mention hundreds of other ground breaking use cases for LLMs. I suspect you're some kind of a web coder. It is funny when you lot call yourself "engineers".
    7
  4. Since even a linear regression was called an "AI" for the past couple of decades, nobody really expects anything fancy from any "AI" or "ML".
    3
  5. I am grateful to all those crazy conspiracies for exposing a lot of people as what they truly are. Just thinking of the fact I used to respect some of them make me shiver.
    3
  6.  @ang3111  you're not a software engineer.
    2
  7. Use local models then. Fine-tune them on your own data. There is hardly any justification for cloud-based inference.
    2
  8. Sadly, crypto is never "over", just like all the other pyramid schemes are never over. One pyramid collapse, two new start.
    2
  9. It is too closed and unhackable for any serious use case.
    2
  10. Microsoft is doing it right - see the paper "Textbooks is all you need".
    2
  11. Technological evolution is a real thing. Every single advancement is very incremental and relies on decades of trial and error. The inventive power of engineering is vastly overrated, engineers are those proverbial typing monkeys who are supposed to produce "War and Peace" eventually.
    1
  12. ​ @sharkbelly1169 imagine being a web code monkey and having an audacity to have opinions on things so far above your pay grade...
    1
  13.  @ang3111  if your job title.says you're an engineer, it does not necessarily mean you're an engineer. FAANG especially, they hired a lot of useless trash. Since you're so obviously incompetent, it is clearly your case.
    1
  14. ​ @ang3111 my condolenses to the poor FAANG company that employs such "engineers".
    1
  15.  @ang3111  this field was never ML-first, up until recent advancements. Explainable, physically-based algorithms were always the main requirement. And LLMs allowed to bridge the gap between fuzzy hand-wavey ML world and explainable / certifiable certainty that industrial robotics needs. And if LLMs work perfectly here, they must work even better in simpler and less restricted domains. I use LLMs to generate formal proofs for the mission-critical code. It was very tedious previously, now fully automated. I use vLLMs to design mechanical parts with a CAD. I use LLMs to generate and debug in simulation good enough circuits that I can later refine manually. It saves me weeks or even months of work. I was never so productive. Literally every area of my work is improved by LLMs. And yes, of course I am not using raw ChatGPT (or any other cloud-hosted model).
    1
  16.  @ang3111  again, even if in this hard field AI shines (and mind you, explainability is a must, so it is really a hard sell), in all others it outperformed humans by far.
    1
  17.  @SoulyG  you have no faintest idea of what you're talking about. You don't understand AI.
    1
  18. @SoulyG  LLM is a semantic search engine. I.e., the last remaining piece of puzzle we needed to build an AI. Once again, you have no faintest idea of what you're talking about. LLM itself cannot do logical reasoning. An LLM armed with Prolog or an SMT solver is reasoning better than any human.
    1
  19. @SoulyG  PS a very simple example to counter your "specific task" point - LLMs write a much better code in programming languages they do not know. If the language was not in a train set, an LLM cannot just generate learned patterns and havr to resort to thinking step by step, querying the documentation on its way.
    1
  20. A lot of people make this mistake.
    1
  21.  @TookAHikeNowWhat  I'm using local models exclusively, with an RTX-4080, small models (7B, 11B) are super fast.
    1
  22. I do not think it is correct to classify them as "smart", falling to mythical thinking (and, as a consequence, conspiracy theories) pretty much a proof of a very deep developmental flaw in them. They were never "smart", they just managed to mimic some of the smart people behavious well enough to fool people around them.
    1
  23. Winter is coming. The second AI winter. Once hype is over, there'll be nothing left.
    1
  24. Is it new? 1970s promise of supersonic commercial flight turned BS. Pervasive nuclear power turned BS (well, thanks to the public and politicians, not the tech), promise of borderless finance turned BS.
    1
  25. And not a single conspiracy was ever uncovered by conspiracy theorist trash.
    1