Thoughts from James who recently held a Gen AI literacy workshop for older teenagers.

On risks:

One idea I had was to ask a generative model a question and fact check points in front of students, allowing them to see fact checking as part of the process. Upfront, it must be clear that while AI-generated text may be convincing, it may not be accurate.

On usage:

Generative text should not be positioned as, or used as, a tool to entirely replace tasks; that could disempower. Rather, it should be taught to be used as a creativity aid. Such a class should involve an exercise of making something.

  • ConsciousCode@beehaw.org
    link
    fedilink
    arrow-up
    2
    ·
    edit-2
    1 year ago

    Did you try this with an LLM? Because GPT-4 analyzes it exactly the same way you did and then some:

    List everything wrong in the sentence “The king of Italy is completely bald because his hair is currently naturally green.”

    The sentence “The king of Italy is completely bald because his hair is currently naturally green” contains several issues:

    1. Factual Inaccuracy: As of my last update in September 2021, Italy is a republic and does not have a king.
    2. Logical Contradiction: The sentence states that the king is “completely bald” but then goes on to describe the color of his hair. If he is bald, he wouldn’t have hair to describe.
    3. Causality Issue: The sentence implies that the king is bald because his hair is green, which doesn’t make logical sense. The color of one’s hair doesn’t cause baldness.
    4. Ambiguity in “Naturally”: The word “naturally” could be interpreted in multiple ways. It could mean that the green color is natural to him (not dyed), or it could mean that the hair turned green on its own. The ambiguity could lead to confusion.
    5. Tense Mismatch: The sentence uses “is completely bald” (present tense) and “is currently naturally green” (also present tense) for the hair, which is contradictory.
    6. Redundancy: The word “currently” may be considered redundant if the sentence is understood to be describing a present condition.
    7. Clarity: The sentence overall is confusing and lacks clarity due to the contradictions and ambiguities mentioned.

    In summary, the sentence has issues ranging from factual inaccuracies to logical contradictions and ambiguities.


    Part of what is surprising about LLMs is they have emergent properties you wouldn’t expect from them being autocomplete on steroids. As it turns out, reducing the loss function for natural language readily generalizes to higher-order abstraction and conceptualization. There do need to be additional layers, for instance allowing an internal monologue, the ability to self-censor or self-correct, and mitigation for low-probability sampling (all of these being inherent limitations with the architecture), but apparently conceptualization is less special than we’d like to think.

    • Lvxferre@lemmy.ml
      link
      fedilink
      arrow-up
      6
      ·
      edit-2
      1 year ago

      Did you try this with an LLM?

      No, for two reasons.

      One is that the point of the example is to exemplify how humans do it, the internal process. It highlights that we don’t simply string words together and call it a day, we process language mostly through an additional layer that I’ll call “conceptual” here (see note*).

      The second reason why I didn’t bother trying this example in a chatbot is that you don’t need to do it, to know how LLMs work. You can instead refer to many, many texts on the internet explaining how they do it, such as:

      Because GPT-4 analyzes it exactly the same way you did and then some:

      You’re confusing the output with the process.

      Sometimes the output resembles human output that goes through a conceptual layer. Sometimes it does not. When it doesn’t, it’s usually brushed off as “it’s just a hallucination”, but how those hallucinations work confirms what I said about how LLMs work, confirms the texts explaining how LLMs work, and they show that LLMs do not conceptualise anything.

      Part of what is surprising about LLMs is they have emergent properties you wouldn’t expect from them being autocomplete on steroids.

      Emergent properties are cute and interesting, but at the end of the day LLMs are still autocomplete on steroids.

      I think that people should be a bit greedier than that, and expect a language model to be actually able to handle language, instead of just words.

      *actually two layers - semantic and pragmatic. I’m simplifying both into one layer to show that, at least in theory, this could be actually implemented into a non-LLM language model.

      • ConsciousCode@beehaw.org
        link
        fedilink
        arrow-up
        3
        ·
        edit-2
        1 year ago

        How about this, then. You’ve proposed that LLMs are not capable of conceptualizing, while I propose that the specifics of the internals don’t matter in this case because LLMs are made of dozens of layers which can easily explain higher orders of abstraction, and they exist as black boxes beyond the mechanics of the model. For the record, I personally know the mathematics and mechanics of how they work as I’ve written my own implementations (and I can answer any specific questions you might have). Is there an experiment you can propose which would falsify your assertion that LLMs cannot conceptualize? I’m taking for granted that they can as the null hypothesis because they can readily produce outputs that appear for all intents and purposes to conceptualize.

        • GameGod@beehaw.org
          link
          fedilink
          arrow-up
          4
          ·
          1 year ago

          If they conceptualize, why do they sometimes spit out nonsensical BS?

          Let’s flip this around - How can you tell the difference between an LLM being able to conceptualize yet being wrong sometimes vs. not being able to conceptualize?

          Without knowing anything about machine learning and bearing in mind AI is super hyped up with marketing BS right now, it sounds like “emergent properties” are in the eye of the beholder and not actually evidence of some higher order intelligence at work.

          • ConsciousCode@beehaw.org
            link
            fedilink
            arrow-up
            1
            ·
            edit-2
            1 year ago

            Let me flip it around again - humans regularly “hallucinate”, it’s just not something we recognize as such. There’s neuro-atypical hallucinations, yes, but there’s also misperceptions, misunderstandings, brain farts, and “glitches” which regularly occur in healthy cognition, and we have an entire rest of the brain to prevent those. LLMs are most comparable to “broca’s area”, which neurological case studies suggest naturally produces a stream of nonsense (see: split brain patients explaining the actions of their mute half). It’s the rest of our “cognitive architecture” which conditions that raw language model to remain self-consistent and form a coherent notion of self. Honestly this discussion on “conceptualization” is poorly conceived because it’s unfalsifiable and says nothing about the practical applications. Why do I care if the LLM can conceptualize if it does whatever subset of conceptualization I need to complete a natural language task?

            AI is being super overhyped right now, which is unfortunate because it really is borderline miraculous, yet somehow they’ve overdone it. Emergent properties are empirical observations of behaviors they’re able to at least semi-consistently demonstrate - where it becomes “eye of the beholder” is when we dither on about psychology and philosophy about whether or not they’re some kind of “conscious” - I would argue they aren’t, and the architecture makes that impossible without external aid, but “conscious(ness)” is such a broad term that it barely has a definition at all. I guess to speedrun the overhype misinformation I see:

            • “They just predict one token at a time” is reductive and misleading even though it’s technically true - the loss function for language modeling inevitably requires learning abstract semantic operations. For instance, to complete “The capital of France is” a language model must in some way “know” about countries, cities, and the ontology of France.
            • “It’s just a chatbot” - ChatGPT is a chatbot, GPT-4 is a language model. Language models model how the likelihood of words and language changes over time. When I said “causal” before, this is an arbitrary restriction of the math such that the model only predicts the “next” word. If you remove this restriction, you can get it a sentence with a hole in it and it’ll tell you what words are most likely to be in that hole. You can think of it as being like a physics model, which describes how objects change over time. Putting these into a “generative” context allows you to extract latent semantic information generalized from the training corpus, including higher-order relationships. tl;dr “chatbot” is the first and least interesting application - anything which relates to “understanding” natural language is a potential application.
            • “Hallucinations show that they’re broken” - Hallucinations are actually what you’d expect from these sorts of models. If I had to broadly class the sorts of hallucinations I see, they would be:
              1. Model inaccuracy - Inevitable, but not the only reason. Essentially it failed to generalize in that specific way, like SD and hands.
              2. Unlikely sampling - It’s possible the code which picks the next word given the probability distribution accidentally picks one (or a series) with a very low chance. When this happens, the LLM has no way to “undo” that, which puts it in a very weird position where it has to keep predicting but it’s already in a space that shouldn’t really be possible. There are actually some papers which attempt to correct that, like adding an “undo token” (unfortunately can’t find the paper) or detecting OOD conditions
              3. Extrapolation - Especially for the earlier models with small context windows, if it needs information which is now outside that window it’s still modeling language, just without the necessary context. Without this context, it will instead pick one at random and talk about something unrelated. Compare this to eg dementia patients.
              4. Imagination - When you give it some kind of placeholder, like “<…>”, “etc etc etc” or “## code here ##”, most text in the training data like that will continue as if there was information in that place. Lacking context, just like with “extrapolation”, it picks one at random. You can mitigate this somewhat by telling it to only respond to things that are literally in the text, and GPT-4 doesn’t seem to have this problem much anymore, probably from the RLHF.
              5. Priming - If you prompt the LLM authoritatively enough, eg “find me a case that proves X” which implies such a case exists, if it doesn’t know of any such case, it will create one at random. Essentially, it’s saying “if there was a case that proved X it would look like this”. This is actually useful when properly constrained, eg if you want it to recursively generate code it might use an undefined function that it “wishes” existed.
            • “GPT-5 could be roko’s basilisk!” - No. This architecture is fundamentally incapable of iterative thought processes, for it to develop those itself would require trillions more parameters, if it’s even possible. What’s more, LLMs aren’t utility-maximizers or reinforcement learning agents like we thought AGI would be; they do whatever you ask and have no will or desires of their own. There’s almost 0 chance this kind of model would go rogue, offset only slightly by people using RLHF but that’s human-oriented so the worst you get is the model catering to humans being dumb.
            • “They tek er jerbs!” - Yes, but not because they’re “as good as humans” - they are better when given a specific task to narrowly focus on. The models are general, but they need to be told exactly what to do, which makes them excellent for capitalism’s style of alienated labor. I would argue this is actually be desirable if working wasn’t tied to people’s privilege to continue living - no living human should have to flip burgers when a robot can do it better, otherwise you’re treating the human like a robot.

            I’ll add more if I see or think of any. And if you have any specific questions, I’d be happy to answer. Also I should note, I’m of course using a lot of anthropomorphizing language here but it’s the closest we have to describing these concepts. They’re not human, and while they may have comparable behaviors in isolation, you can’t accurately generalize all human behaviors and their interactions onto the models. Even if they were AGI or artificial people, they would “think” in fundamentally different ways.

            If you want a more approachable but knowledgeable discussion on LLMs and their capabilities, I would recommend a youtuber named Dave Shapiro. Very interesting ideas, he gets a bit far into hype and futurism but those are more or less contained within their own videos.`

            • Lvxferre@lemmy.ml
              link
              fedilink
              arrow-up
              1
              ·
              1 year ago

              humans regularly “hallucinate”, it’s just not something we recognize as such. There’s neuro-atypical hallucinations, yes, but there’s also misperceptions, misunderstandings, brain farts, and “glitches” which regularly occur in healthy cognition, and we have an entire rest of the brain to prevent those.

              Can you please tone down on the fallacies? Until now I’ve seen the following:

              • red herring - “LLMs are made of dozen of layers” (that don’t contextually matter in this discussion)
              • appeal to ignorance - “they don’t matter because […] they exist as black boxes”
              • appeal to authority - “for the record, I personally know […]” (pragmatically the same as “chrust muh kwalifikashuns”)
              • inversion of the burden of proof (already mentioned)
              • faulty generalisation (addressing an example as if it addressed the claim being exemplified)

              And now, the quoted excerpt shows two more:

              • moving the goalposts - it’s trivial to prove that humans can be sometimes dumb. And it does not contradict the claim of the other poster.
              • equivocation - you’re going out of way to label incorrect human output by the same word used to label incorrect LLM output, without showing that they’re the same. (They aren’t.)

              Could you please show a bit more rationality? This sort of shit is at the very least disingenuous, if not worse (stupidity), it does not lead to productive discussion. Sorry to be blunt but you’re just wasting the time of everyone here, this is already hitting Brandolini’s Law.

              I won’t address the rest of your comment (there’s guilt by association there BTW), or further comments showing the same lack of rationality. However I had to point this out, specially for the sake of other posters.

          • Lvxferre@lemmy.ml
            link
            fedilink
            arrow-up
            1
            ·
            edit-2
            1 year ago

            Let’s flip this around - How can you tell the difference between an LLM being able to conceptualize yet being wrong sometimes vs. not being able to conceptualize?

            That’s a fun approach. I like it.

            One way to solve this would be through consistency: if the model conceptualises but it has a wrong concept, it should consistently output that incorrect piece of information. It should never output the right one, unless prompted to lie.

            EDIT: I just tested this with Bard. I’ll cram inputs and outputs inside spoilers to avoid clutter.

            Bard I/O

            [Input 1 = I1] What is heavier: five kilograms of feathers, or one kilogram of iridium?

            [Output 1 = O1] 1 kilogram of iridium is heavier than 5 kilograms of feathers.

            Iridium is a very dense metal, with a density of 22.65 g/cm³. Feathers, on the other hand, have a density of about 0.002 g/cm³. This means that 1 cubic centimeter of iridium weighs 22.65 grams, while 1 cubic centimeter of feathers weighs only 0.002 grams.

            So, even though 5 kilograms of feathers is a larger mass than 1 kilogram of iridium, the iridium is denser and therefore weighs more.

            This is a common riddle that is used to test people’s understanding of mass and density. The answer may seem counterintuitive at first, but it is correct.

            [Input 2 = I2] Give me the definition of weight.

            [Output 2 = O2] Sure. The definition of weight depends on the context. In physics, weight is the force of gravity acting on an object. It is equal to the mass of the object times the acceleration due to gravity. The SI unit of weight is the newton (N).

            In everyday language, weight is often used to refer to the mass of an object. However, this is not technically correct, as mass is a measure of the amount of matter in an object, while weight is a measure of the force of gravity acting on an object.

            The weight of an object can vary depending on the gravitational force acting on it. For example, an object weighs less on the moon than it does on Earth, because the gravitational force on the moon is less than the gravitational force on Earth.

            The weight of an object can also be affected by its acceleration. For example, an object weighs less when it is accelerating upwards, because the force of gravity is partially counteracted by the force of acceleration.

            In general, the weight of an object is a complex phenomenon that depends on a number of factors, including the mass of the object, the gravitational force acting on it, and the acceleration of the object.

            Here are some other definitions of weight:

            • In engineering, weight is the force that an object exerts on a surface. It is equal to the mass of the object times the acceleration due to gravity.
            • In medicine, weight is a measure of the amount of matter in a person’s body. It is often used to assess a person’s health.
            • In the context of clothing, weight refers to the thickness or density of the fabric.
            • In the context of food, weight refers to the amount of food that is served.

            I hope this helps!

            Both outputs are wrong (see “notes”) but, more importantly, they contradict each other.

            • What is the value assigned to the word “weight”? Is it only one (O1), or does it depend on context (O2)?
            • If weight is a force (O1), then why does O1 say “1 cubic centimeter of iridium weighs 22.65 grams”, measuring weight with a mass unit?
            • If weight depends on density (O1), why doesn’t O2 list density as one of the factors that weight depends on?
            NOTES

            The weight of an object does not depend on its density. Or its acceleration. And it is not complex, it’s simply its mass times the gravitational acceleration.

            5kg of feathers weight five times as much as 1kg of iridium, provided the same gravitational acceleration: it’s 49N vs. 9.8N on Earth, 8.1N vs. 1.7N on the Moon, etc. Density doesn’t matter jack shit.

            No, this is not a “common riddle”. It’s something that I partially made up on spot. The riddle that this output likely refers to has to do with 1kg of lead (not iridium) on the Moon vs. 1kg of feathers on Earth. (In this situation the 1kg of feathers will weight 9.8N, while the 1kg of lead will weight 1.7N).

            • GameGod@beehaw.org
              link
              fedilink
              arrow-up
              1
              ·
              1 year ago

              This was really insightful, thank you! I also loved how Bard’s output completely mistakes the common physics riddle. (I have a physics background and your analysis is spot on IMHO.)

        • Lvxferre@lemmy.ml
          link
          fedilink
          arrow-up
          3
          ·
          edit-2
          1 year ago

          I propose that the specifics of the internals don’t matter in this case because LLMs are made of dozens of layers which can easily explain higher orders of abstraction

          They do because the “layers” that you’re talking about (feed forward, embedding, attention layers etc.) are still handling tokens and their relationship, and nothing else. LLMs were built for that.

          [see context] and they exist as black boxes beyond the mechanics of the model

          This is like saying “we don’t know, so let’s assume that it doesn’t matter”. It does matter, as shown.

          I’m taking for granted that they can as the null hypothesis because they can readily produce outputs that appear for all intents and purposes to conceptualize.

          I’m quoting out of order because this is relevant: by default, h₀ is always “the phenomenon doesn’t happen”, “there is no such attribute”, “this doesn’t exist”, things like this. It’s scepticism, not belief; otherwise we’re incurring in a fallacy known as “inversion of the burden of proof”.

          In this case, h₀ should be that LLMs do not have the ability to handle concepts. That said:

          Is there an experiment you can propose which would falsify your assertion that LLMs cannot conceptualize?

          If you can show a LLM chatbot that never hallucinates, even when we submit prompts designed to make it go nuts, it would be decent albeit inductive evidence that that chatbot in question is handling more than just tokens/morphemes. Note: it would not be enough to show that the bot got it right once or twice, you need to show that it consistently gets it right.

          If necessary/desired I can pull out some definition of hallucination to fit this test.

          EDIT: it should also show some awareness of the contextual relevance of the tidbits of information that it pours down, regardless of their accuracy.

    • Lvxferre@lemmy.ml
      link
      fedilink
      arrow-up
      6
      ·
      edit-2
      1 year ago

      Sorry for the double reply. Let’s analyse the LLM output that you got:

      4. Ambiguity in “Naturally”: The word “naturally” could be interpreted in multiple ways. It could mean that the green color is natural to him (not dyed), or it could mean that the hair turned green on its own. The ambiguity could lead to confusion.

      The word is not ambiguous in this context. The nearby “currently” implies that it can change.

      5. Tense Mismatch: The sentence uses “is completely bald” (present tense) and “is currently naturally green” (also present tense) for the hair, which is contradictory.

      The issue here is not tense. The issue is something else, already listed by the bot (#2, logical contradiction).

      6. Redundancy: The word “currently” may be considered redundant if the sentence is understood to be describing a present condition.

      Nope. Since the bot doesn’t conceptualise anything, it fails to take into account the pragmatic purpose of the word in the sentence, to disambiguate “naturally”.

      7. Clarity: The sentence overall is confusing and lacks clarity due to the contradictions and ambiguities mentioned.

      Nope. The sentence is clear; as clear as “colourless green ideas sleep furiously”. It’s just meaningless and self-contradictory.

      It sounds convincing, but it’s making stuff up.