528
Don't paste the AI. (dontpastetheai.com)
submitted 5 days ago* by to c/technology@lemmy.world
148 comments fedilink hide all child comments
you are viewing a single comment's thread
view the rest of the comments
[–] 1 point 4 days ago (1 child)

Then you don't actually understand what LLMs are, and you're using the term as a short-hand to mean chat bots. I'm not trying to be rude, but you need to understand that is just a fact, if you think LLMs can only be chat-bots, and can only be corporate, and can only be trained unethically.

Would it surprise you to know that quite a lot of genuine advances have been made by using LLMs that don't speak any language you'd recognize? Evo 1 and Evo 2 for example "Speak" genome sequences. There are other models that have been used to improve weather modeling, and animal behaviour analysis.

You're right that AI is a very broad term, but so is LLM. The problem is that ignorant people see "AI" and automatically assume it's bad because of the connotation with Chat-Bots.

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

    if you think LLMs can only be chat-bots, and can only be corporate, and can only be trained unethically

    Don't put words in my mouth.

    1 and Evo 2 for example "Speak" genome sequences. There are other models that have been used to improve weather modeling, and animal behaviour analysis.

    Evo 1 and 2 are deep learning models (I think Genomic Language Models would be the correct term), but not Large Language Models. I'm willing to bet neither are the "other" models you're mentioning. They're unlikely to be trained on vast amounts of human text for purposes of natural language interaction.

    Besides, none of those "other models" are in any way applicable to either the OP or the critique of using LLMs as a source of information.

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

    This is my point. They are LLMs. That is not an opinion or a debatable point. They are categorically, definitionally, LLMs. You don't understand what LLMs are, because LLMs ARE deep learning models, and instead of taking the time to actually learn about the technology you're responding to my corrections with hostility.

    You and I are on the same side. Chat-bots ARE harmful. But putting that label on a technology as a whole is purely tribe-based fear that is already causing the spread of unfounded fear and misinformation.

    And the claim wasn't "Don't use AI as a source of information," which I agree with. It was "Don't use any AI," which you clarified to mean LLMs. So this is very much pertinent.

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

    You don't understand what LLMs are, because LLMs ARE deep learning models, and instead of taking the time to actually learn about the technology you're responding to my corrections with hostility.

    You're claiming that I don't understand technology while seemingly claiming that because LLMs are a type of deep learning, then all deep learning models are LLMs.

    Evo was trained on genomic sequences, not human text. Per Wikipedia:

    A large language model (LLM) is an AI model (typically a neural network) trained on a vast amount of text for natural language processing tasks, especially language generation.

    Genomic sequences are not natural language. Ergo, "definitionally" Evo 2 is not an LLM.

    While its StripedHyena2 architecture is very similar to LLMs, it does not use the same Generative Pretrained Transformer architecture associated with LLMs (per: https://docs.nvidia.com/bionemo-recipes/2.6.3/interactives/illustrated-evo2/index.html ).

    My hate of LLMs is certainly not misinformed. It's only tribe-based in that I care for humanity.

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

    I did not say all deep learning algorithms are LLMs, I said all LLMs are deep learning algorithms. It's a nested hierarchy, that relationship only runs one way.

    Yes. Evo was trained on text strings of genomic data. That's what I was trying to say was the misunderstanding. LLMs do not require language as you and I would recognize it. If you think that calling them "Large Language Models," is misleading, I kinda agree, but then we can start arguing semantics about why scientists name anything the way they do. Dark matter isn't actually dark, and probably isn't matter. Dark energy isn't dark. There was no explosion during the big bang.

    If you're using colloquial word usage and demanding scientific advances follow your expectations, you're going to have a bad time.

    But let's just say for the sake of argument that you're 100% correct, and that Evo is not an LLM. Rather, it's something extremely close, save for a few differences.

    People are already angry at the researchers for "Using AI to develop super-bugs," or saying the genuinely universal "AI has no uses!" And earnestly failing to differentiate between the shit-bot that makes deep-fakes, and the AI designed to identify cancers.

    That is why I absolutely reject your disrespectful framing that by defending a technology, and NOT it's worst uses, I'm somehow opposite to "care(ing) for humanity."

    The actual technology behind Evo and ChatGPT is structurally the same. The methods of training a model on genomic data or weather patterns is indistinguishable from training it on stolen media. The difference is the uses, and targets.

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

    But let's just say for the sake of argument that you're 100% correct, and that Evo is not an LLM.

    Why only assume? I cited Wikipedia. You cited nothing.

    That is why I absolutely reject your disrespectful framing that by defending a technology, and NOT it's worst uses, I'm somehow opposite to "care(ing) for humanity."

    There you go putting words in my mouth again.

    The actual technology behind Evo and ChatGPT is structurally the same.

    ...Except that it isn't. It's not a Generative Pre-Trained Transformer. It uses a Transformer-like architecture. You cannot use Evo's architecture to make a chatbot. It's a GLM.

    The methods of training a model on genomic data or weather patterns is indistinguishable from training it on stolen media.

    Go ahead and train a StripedHyena2 model to be a chatbot, then. I'm sure that will work great.

    For someone who's such a stickler for making 100% correct and unambiguous statements, you're sure keen on asserting equality where there is merely similarity.

    Nobody is claiming these technologies don't share some (or even a lot of) DNA. Being upset that people correctly use the definition of LLMs as outlined by Wikipedia, where even Evo's own Github page doesn't claim it's an LLM, is just derailing the conversation away from people's righteous objections.

    Nobody is campaigning against using non-chatbot AI for science.

    Read the room. Pay more attention to the context.

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

    Okay... You seem to get really upset when I directly quote you, and claim that I'm putting words in your mouth, so I'll do what you do and cite your phrases exactly.

    Why only assume? I cited Wikipedia. You cited nothing.

    Because whether you or I are correct in the specific definition is what is means to be an LLM is irrelevant to my purpose in engaging in this conversation. I concede that I might have overreached in my confidence of whether or not it is an LLM, because Nature itself calls it:

    "the largest-scale fully open language model[s] to date."

    And a lot of the other abstracts and comments on Evo draw the deep similarities to other LLMs. Again, this comes down to where the clean line between any given AI actually is. And my entire point behind this argument is that it doesn't matter to anybody outside the lab.

    My hate of LLMs is certainly not misinformed. It's only tribe-based in that I care for humanity.

    Stating you hate LLMs because you care for humanity directly supposes that those who do not hate LLMs, do not care for humanity.

    "I'm against X because I care about protecting children," is used ad nauseum to suggest that NOT being against X automatically means you don't care about children. I'm not putting words in your mouth by interpreting your words in the way words are typically interpreted.

    If I misunderstand your meanings, I apologize. But please stop assuming I'm intentionally misinterpreting you when I'm literally just reading your words and responding to them to the best of my ability.

    ...Except that it isn't. It's not a Generative Pre-Trained Transformer. It uses a Transformer-like architecture. You cannot use Evo's architecture to make a chatbot. It's a GLM.

    Okay. Is that the defining feature of where your problem lies? Specifically in the capacity to use human language to interact with AI? Because that's a specific issue that can be addressed, rather than the vague "I don't like this technology because it has hurt people."

    Go ahead and train a StripedHyena2 model to be a chatbot, then. I'm sure that will work great.

    Okay, second pass. So your issue is specifically with interacting with AI using natural human language?

    For someone who's such a stickler for making 100% correct and unambiguous statements, you're sure keen on asserting equality where there is merely similarity.

    Fair. I shouldn't have confidently claimed they were identical in every way. The vast majority of the time, I'm talking to people that have very little understanding of the technology, and have just pure vibe-based fears about it. I apologize for the unjustified gross-comparison.

    Nobody is claiming these technologies don't share some (or even a lot of) DNA. Being upset that people correctly use the definition of LLMs as outlined by Wikipedia, where even Evo's own Github page doesn't claim it's an LLM, is just derailing the conversation away from people's righteous objections.

    Again, my focus on the conversation was separating "This technology is universally bad," from "This technology has been used to hurt people." I wasn't trying to derail the conversation, I was trying to focus it. I am far less interested in scientific definitions of what every single model technically classifies as, than I am in addressing the widespread blanket hate against a technology that has no say in how it is used. I don't think dynamite is evil, nor do I think genetics is evil. Despite both being used for MASSIVE harm at various points in history.

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

    Stating you hate LLMs because you care for humanity directly supposes that those who do not hate LLMs, do not care for humanity.

    You accused me of tribalism. I responded that the only tribal thing about my stance is that I care about the "tribe" of humanity. How you got what you got out of it, I truly don't know. You were the one throwing out ad hominems, dude.

    So your issue is specifically with interacting with AI using natural human language?

    At this point I have no idea what you're on about. The concerns regarding LLMs are widely documented. Some examples:

    • They must be trained on stolen data to be vaguely useful. Yes, training on the Common Crawl still counts as stealing. No, there's not enough royalty free data to train on which would create a useful LLM.
    • "Useful" in this context is extremely debatable. The architecture itself makes hallucinations a mathematical certainty. The chatbot must use the internet to have general up-to-date knowledge and that opens it up to prompt injection.
    • Security is an unsolvable problem for LLMs. Even if you don't give it access to the internet, it can still be prompt injected via an external document. Prompt injection cannot be fixed because all inputs go through the same place - the prompt. 'Agentic' use cases are hilariously insecure and are, again, insecurable. A system prompt "guard rail" is not a security measure.
    • Context rot is another probably unsolvable problem which renders large context windows useless.
    • the amount of electricity and water needed to train the LLM and then infer outputs is untenable (there is no proof inference is profitable).
    • The amount of hardware necessary to power LLMs is untenable (no, local quantized models are not "almost as good" and were not trained for free). I can't buy a new PC and that's insane.
    • Because the LLMs are just fancy autocomplete (yes, they are, and no amount of backpropagation and attention heads will change that), they just produce most likely text, not actual answers. As such, the answers are often incorrect but *sound *like they are. Therefore, no LLM can be trusted to produce accurate information at any point in time.
    • The entire "AI" industry is unprofitable and propped up on debt and circular financing by an industry that's out of hypergrowth ideas, while gaslighting normal people that it's the future.
    • There is no future where this ends well. Either the bubble bursts and the economy collapses, or "AI" takes our jobs and we're left to starve.
    • LLMs are a fun toy, but no one has found a valuable use case for it that couldn't have been achieved using other means. Oh, people are vibe coding their own software with it? Then where are all the world-changing startups that completely transform our lives? Where's a single "AI" success story that people are excited to use, besides some people falling for the sycophancy of chatbots?
    • The existence od LLMs has objectively made the world a worse place, due to AI slop and misinformation. It was also used as an excuse when companies lay people off. LLMs are, terrifyingly, used in medicine, where they hallucinate patient notes saying the wrong breast has cancer, or that the patient is a drug addict (when they're not). They're being used to avoid accountability when bombing schools. CEOs and managers uncritically enforce LLM adoption despite the known and very obvious risks and limitations.

    I could go on. There's so much more. No, I'm not against machine learning. I'm not against deep learning. I'm against LLMs - a technology which takes human input and outputs text and needs enormous amounts of human text, electricity and water to produce a fancy autocomplete which has extremely narrow use cases at best and is used to enshittify the entire world while stealing our resources.

    But the original comment was about how you can't trust AI (in this context LLM) output. You still can't. You still shouldn't.

    As far as we know, no AI generated output can ever be trusted without careful verification. Only deterministic algorithms can be given that level of trust.

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

    You were the one throwing out ad hominems, dude.

    Lol what? Ad hominems? Oh Jesus Christ you're one of those... So I ignored your argument and said you didn't know what you were talking about because you were a janitor? Or that you can't possibly understand LLMs because you're too old/young/uneducated?

    No. Disagreeing is not an ad hominem. Claiming you don't understand something is not an ad hominem. Misunderstanding is not an ad hominem. I even apologized for the misunderstanding. And if you can't see how "I believe in X because I care about humanity," doesn't directly insinuate that not caring about X means you don't care about humanity, then holy fuck communication was never going to work.

    It's a shame, because you had some good points, and I was hoping that apologizing would have reduced your hostility. But I can see now that your goal has been to "Win" the argument. I've literally never seen that ad hominem claim used incorrectly, by somebody trying to have an honest discussion.

    Also. Nothing I said is ad hominem either. It's direct commentary of events within this conversation, independent of what your profession is or who you are. Even the "You're one of those," comment is not ad hominem.

    I'm done 👍.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •