#416 – Yann Lecun: Meta AI, Open Source, Limits of LLMs, AGI & the Future of AI

Yann LeCun is the Chief AI Scientist at Meta, professor at NYU, Turing Award winner, and one of the most influential researchers in the history of AI. Please support this podcast by checking out our sponsors: – HiddenLayer: https://hiddenlayer.com/lex – LMNT: https://drinkLMNT.com/lex to get free sample pack – Shopify: https://shopify.com/lex to get $1 per month trial – AG1: https://drinkag1.com/lex to get 1 month supply of fish oil Transcript: https://lexfridman.com/yann-lecun-3-transcript EPISODE LINKS: Yann’s Twitter: https://twitter.com/ylecun Yann’s Facebook: https://facebook.com/yann.lecun Meta AI: https://ai.meta.com/ PODCAST INFO: Podcast website: https://lexfridman.com/podcast Apple Podcasts: https://apple.co/2lwqZIr Spotify: https://spoti.fi/2nEwCF8 RSS: https://lexfridman.com/feed/podcast/ YouTube Full Episodes: https://youtube.com/lexfridman YouTube Clips: https://youtube.com/lexclips SUPPORT & CONNECT: – Check out the sponsors above, it’s the best way to support this podcast – Support on Patreon: https://www.patreon.com/lexfridman – Twitter: https://twitter.com/lexfridman – Instagram: https://www.instagram.com/lexfridman – LinkedIn: https://www.linkedin.com/in/lexfridman – Facebook: https://www.facebook.com/lexfridman – Medium: https://medium.com/@lexfridman OUTLINE: Here’s the timestamps for the episode. On some podcast players you should be able to click the timestamp to jump to that time. (00:00) – Introduction (09:10) – Limits of LLMs (20:47) – Bilingualism and thinking (24:39) – Video prediction (31:59) – JEPA (Joint-Embedding Predictive Architecture) (35:08) – JEPA vs LLMs (44:24) – DINO and I-JEPA (45:44) – V-JEPA (51:15) – Hierarchical planning (57:33) – Autoregressive LLMs (1:12:59) – AI hallucination (1:18:23) – Reasoning in AI (1:35:55) – Reinforcement learning (1:41:02) – Woke AI (1:50:41) – Open source (1:54:19) – AI and ideology (1:56:50) – Marc Andreesen (2:04:49) – Llama 3 (2:11:13) – AGI (2:15:41) – AI doomers (2:31:31) – Joscha Bach (2:35:44) – Humanoid robots (2:44:52) – Hope for the future

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Lex Fridman?

  The following is a conversation with Jan LeCun, his third time on this podcast. He is the chief AI scientist at Meta, professor at NYU, touring award winner, and one of the seminal figures in the history of artificial intelligence. He and Meta AI have been big proponents of open sourcing AI development and have been walking the walk by open sourcing many of their biggest models, including Llama 2 and eventually Llama 3. Also, Jan has been an outspoken critic of those people in the AI community who warn about the looming danger and existential threat of AGI.

• 0:42Lex Fridman?

  He believes the AGI will be created one day, but it will be good. It will not escape human control, nor will it dominate and kill all humans. At this moment of rapid AI development, this happens to be somewhat a controversial position. And so it's been fun seeing Jan get into a lot of intense and fascinating discussions online as we do in this very conversation. And now a quick few second mention of each sponsor. Check them out in the description. It's the best way to support this podcast. We've got Hidden Layer for securing your AI models, Element for electrolytes, Shopify for shopping for stuff online,

• 1:27Lex Fridman?

  and AG1 for delicious health. Choose wisely, my friends. Also, if you want to get in touch with me for whatever reason, maybe to work with our amazing team, go to lexfridman.com/contact. And now on to the full ladder reads. Never any ads in the middle. I try to make these interesting. I don't know why I'm talking like this, but I am. There's a staccato nature to it. Speaking of staccato, I've been playing a bit of piano. Anyway, if you skip these ads, please do check out the sponsors. We love them. I love them. I enjoy their stuff. Maybe you will too.

• 2:07Lex Fridman?

  This episode is brought to you by a on theme in context. See what I did there? Sponsor, since this is Jan LeCun, artificial intelligence, machine learning. One of the seminal figures in the field. So of course you're going to have a sponsor that's related to artificial intelligence. Hidden Layer, they provide a platform that keeps your machine learning models secure. The ways to attack machine learning models, large language models, all the stuff we talk about with Jan, there's a lot of really fascinating work, not just large language models, but the same for video, video prediction, tokenization where the tokens are in the

• 2:52Lex Fridman?

  space of concept versus the space of literally letters, symbols, Japa, V Japa, all of that stuff that they're open sourcing, all the stuff they're publishing on, just really incredible. But that said, all of those models have security holes in ways that we can't even anticipate or imagine at this time. And so you want good people to be trying to find those security holes, trying to be one step ahead of the people that are trying to attack. So if you're especially a company that's relying on these models, you need to have a person who's in charge of saying, yeah, this model that you got from this place has been tested, has been secured, whether that place is a hugging face or any other kind

• 3:37Lex Fridman?

  of stuff, or any other kind of repository or model zoo kind of place. I think the more and more we rely on large language models or just AI systems in general, the more the security threats that are always going to be there become dangerous and impactful. So protect, protect your models. Visit hiddenlayer.com/lex to learn more about how Hidden Layer can accelerate your AI adoption in a secure way.

• 4:09Lex Fridman?

  This episode is also brought to you by Element. A thing I drink throughout the day. I'm drinking now. When I am on a podcast, you'll sometimes see me with a mug and clear liquid in there that looks like water. In fact, it is not simply water. It is water mixed with Element, watermelon salt, cold. What I do is I take one of them, Power Raid 28 fluid ounces bottles, fill it up with water, one pack of watermelon salt, shake it up, put in the fridge, that's it. I reuse the bottles and drink from a mug, or sometimes from the bottle. Either way,

• 4:54Lex Fridman?

  delicious, good for you, especially if you're doing fasting, especially if you're doing low carb kinds of diets, which I do. You can get a sample pack for free with any purchase. Try it at drinkelement.com/lex. This episode is brought to you by Shopify as I take a drink of Element. It is a platform designed for anyone, even me, to sell stuff anywhere on a great looking store. I use a basic one, like a really minimalist one. You can check it out if you go to lexstreaming.com/store. There's a few shirts on there. If that's your thing. It was so easy to set up. I imagine

• 5:35Lex Fridman?

  there's like a million features they have that can make it look better and all kinds of extra stuff you can do with the store, but I use the basic thing and the basic thing is pretty damn good. I like basic. I like minimalism. And they integrate with a lot of third-party apps, including what I use, which is on-demand printing. So like you buy the shirt on Shopify, but it gets printed and shipped by another company that I always keep forgetting, but I think it's called Printful. Or Printify, or I think it's Printful. I'm not sure. Doesn't matter. I think there's several integrations. You can check it out yourself. For me, it works.

• 6:17Lex Fridman?

  I'm using the most popular one. Printful, I think it's called. Anyway, I look forward to your letters correcting me on my pronunciation. Shopify is great. I'm a big fan of the good side of the machinery of capitalism. Selling stuff on the internet, connecting people to the thing that they want, or rather, the thing that would make their life better. Both in advertisement and e-commerce and shopping in general. I'm a big believer, when that's done well, your life legitimately, in the long term, becomes better. And so whatever system can connect one human to the thing that makes their life better

• 7:00Lex Fridman?

  is great. And I believe that Shopify is sort of a platform that enables that kind of system. You can sign up for a $1 per month trial period at Shopify.com/lex. That's all lowercase. Go to Shopify.com/lex to take your business to the next level today. This episode is also brought to you by AG1, an all-in-one daily drink to support better health and peak performance. It is delicious. It is nutritious. And I ran out of words that rhyme with those two. Actually, let me use a large language model to figure out what rhymes with delicious.

• 7:45Lex Fridman?

  Words that rhyme with delicious include ambitious, auspicious, capricious, fictitious, suspicious. So there you have it.

• 7:55Lex Fridman?

  Anyway, I drink it twice a day. Also, put it in the fridge.

• 8:00Lex Fridman?

  And sometimes in the freezer, like it gets a little, a little bit frozen, just like a little bit, just a little bit frozen. You got that like slushy consistency. I'll do that too. Sometimes it's freaking delicious. It's delicious. No matter what. So let's just warm. It's delicious. Cold is delicious. Slightly frozen. All of it is just incredible. And of course it covers like the basic multivitamin foundation of, of a, what I think of as a good diet. So it's just a great multivitamin. That's the way I think about it. So all the crazy stuff I do, the physical challenges, the mental challenges, all of that, at least I got AG1. They'll give you one month supply of fish oil when you sign

• 8:41Lex Fridman?

  up at drinkag1.com/lex. This is the Lex Food Run Podcast. To support it, please check out our sponsors in the description. And now, dear friends, here's Jan LaKoon.

• 9:10Lex Fridman?

  You've had some strong statements, technical statements about the future of artificial intelligence recently, throughout your career, actually, but recently as well. You've said that auto aggressive LLMs are not the way we're going to make progress towards superhuman intelligence. These are the large language models like GPT-4, like Lama 2 and 3 soon and so on. How do they work? And why are they not going to take us all the way?

• 9:40Yann LeCun?

  For a number of reasons. The first is that there is a number of characteristics of intelligent behavior. For example, the capacity to understand the world, to understand the physical world, the ability to remember and retrieve things, persistent memory, the ability to reason and the ability to plan. Those are four essential characteristics of intelligent systems or entities, humans, animals.

• 10:13Yann LeCun?

  L L M's can do none of those. Or they can only do them in a very primitive way. And they don't really understand the physical world. They don't really have persistent memory. They can't really reason and they certainly can't plan. And so, you know, if, if, if you expect a system to become intelligent, just, you know, without having the possibility of doing those things, you're making a mistake. That is not to say that autoregister L L M's are not useful. They're certainly useful. That they're not interesting, that we can't build a whole ecosystem of applications around them. Of course, we can.

• 10:53Yann LeCun?

  But as it passed towards human level intelligence, they're missing essential components. And then there is another tidbit or fact that I think is very interesting. Those L L M's are trained on enormous amounts of text. Basically, the entirety of all publicly available texts on the internet, right? That's typically on the order of 10 to the 13 tokens. Each token is typically two bytes. So that's two 10 to the 13 bytes as training data. It would take you or me 170,000 years to just read through this at eight hours a day. So it seems like an enormous amount of knowledge, right? That those systems can accumulate.

• 11:38Yann LeCun?

  But then you realize it's really not that much data. If you, you talk to developmental psychologists and they tell you a four year old has been awake for 16,000 hours in his whole life. And the amount of information that has reached the visual cortex of that child in four years is about 10 to the 15 bytes. And you can compute this by estimating that the optical nerve carry about 20 megabytes per second, roughly. And so 10 to the 15 bytes for a four year old versus two times 10 to the 13 bytes for 170,000 years worth of reading.

• 12:22Yann LeCun?

  What it tells you is that through sensory input, we see a lot more information than we do through language. And that despite our intuition, most of what we learn and most of our knowledge is through our observation and interaction with the real world, not through language. Everything that we learn in the first few years of life, and certainly everything that animals learn, has nothing to do with language.

• 12:49Lex Fridman?

  So it would be good to maybe push against some of the intuition behind what you're saying. So it is true, there's several orders of magnitude more data coming into the human mind, much faster, and the human mind is able to learn very quickly from that, filter the data very quickly. You know, somebody might argue your comparison between sensory data versus language, that language is already very compressed. It already contains a lot more information than the bytes it takes to store them, if you compare it to visual data. So there's a lot of wisdom in language, there's words, and the way we stitch them together, it already contains a lot of information.

• 13:29Lex Fridman?

  So is it possible that language alone already has enough wisdom and knowledge in there to be able to, from that language, construct a world model, an understanding of the world, an understanding of the physical world that you're saying all limbs lack?

• 13:49Yann LeCun?

  So it's a big debate among philosophers and also cognitive scientists, like whether intelligence needs to be grounded in reality. I'm clearly in the camp that, yes, intelligence cannot appear without some grounding in some reality, it doesn't need to be, you know, a physical reality, it could be simulated, but the environment is just much richer than what you can express in language. Language is a very approximate representation of our perceives and our mental models, right? I mean, there's a lot of tasks that we accomplish, where we manipulate a mental model of the situation at hand, and that has

• 14:31Yann LeCun?

  Nothing to do with language. Everything that's physical, mechanical, whatever. When we build something, when we accomplish a task, a model task of, you know, grabbing something, et cetera. We plan our action sequences and we do this by essentially imagining the result of the outcome of a sequence of actions that we might imagine. And that requires mental models that don't have much to do with language. And that's, I would argue, most of our knowledge is derived from that interaction with the physical world. So a lot of, a lot of my, my colleagues who are more interested in things like computer vision are really on that camp that

• 15:15Yann LeCun?

  AI needs to be embodied, essentially. And then other people coming from the NLP side or maybe, you know, some, some other motivation don't necessarily agree with that. And philosophers are split as well.

• 15:31Yann LeCun?

  And the, the complexity of the world is hard to, it's hard to imagine. It, it, you know, it's hard to represent all the complexities that we take completely for granted in the real world that we don't even imagine require intelligence, right? This is the old Moravec paradox from the pioneer of robotics, hence Moravec. Who said, you know, how is it that with computers, it seems to be easy to do high level complex tasks like playing chess and solving integrals and doing things like that, whereas the thing we take for granted that we do every day, like, I don't know, learning to drive a car or, you know, grabbing an object. We can do as computers.

• 16:14Yann LeCun?

  And, you know, we have L L M's that can pass, pass the bar exam. So they must be smart, but then they can't learn to drive in 20 hours like any 17 year old. They can't learn to clear out the dinner table and fill up the dishwasher like any 10 year old can learn in one shot.

• 16:36Yann LeCun?

  Why is that? Like, you know, what, what are we missing? What, what type of learning or, or? Reasoning, architecture, or whatever, are we missing that

• 16:47Yann LeCun?

  basically prevent us from, you know, having level five sort of in cars and domestic robots?

• 16:53Lex Fridman?

  Can a large language model construct a world model that does know how to drive? And does know how to fill a dishwasher, but just doesn't know how to deal with visual data at this time? So it can operate in a space of concepts.

• 17:09Yann LeCun?

  So yeah, that's what a lot of people are working on. So the answer, the short answer is no. And the more complex answer is you can use all kinds of tricks to get

• 17:22Yann LeCun?

  an L L M to basically digest visual representations of, representations of images or video or audio for that matter.

• 17:35Yann LeCun?

  And a classical way of doing this is you train a vision system in some way. And we have a number of ways to train vision systems, either supervised, semi-supervised, self-supervised, all kinds of different ways.

• 17:49Yann LeCun?

  That will turn any image into a high level of representation. Basically, a list of tokens that are really similar to the kind of tokens that

• 18:01Yann LeCun?

  typical L L M takes as an input. And then you just feed that to the L L M in addition to the text. And you just expect the L L M to kind of, you know, during training to kind of be able to use those representations to help make decisions. I mean, there's been work along those lines for, for quite a long time.

• 18:24Yann LeCun?

  And now you see those systems, right? I mean, there are L L M's that can, that have some vision extension. But they're basically hacks in the sense that those things are not like trained end to end to, to handle, to really understand the world. They're not trained with video, for example. They don't really understand intuitive physics, at least not at the moment.

• 18:43Lex Fridman?

  So you don't think there's something special to you about intuitive physics, about sort of common sense reasoning about the physical space, about physical reality? That's, that to you is a giant leap that L L M's are just not able to do.

• 18:55Yann LeCun?

  We're not going to be able to do this with the type of L L M's that we are working with today. And there's a number of reasons for this, but the main reason is.

• 19:05Yann LeCun?

  The way L L M are trained is that you you take a piece of text, you remove some of the words in that text, you mask them, you replace by replace them by blank markers, and you train a genetic neural net to predict the words that are missing.

• 19:20Yann LeCun?

  And if you build this neural net in a particular way, so that it can only look at words that are to the left of the one it's trying to predict, then what you have is a system that basically is trying to predict the next word in a text, right? So then you can feed it a text, a prompt, and you can ask it to predict the next word. It can never predict the next word exactly. And so what it's going to do is produce a probability distribution of all the possible words in your dictionary. In fact, it doesn't predict words, it predicts tokens that are kind of subword units. And so it's easy to handle the uncertainty in the prediction there, because there is only a finite number of possible words in the dictionary, and you can just compute the distribution over them.

• 20:03Yann LeCun?

  Then what you what the system does is that it it picks a word from that distribution. Of course, there's a higher chance of picking words that have a higher probability within that distribution. So you sample from that distribution to actually produce a word. And then you shift that word into the input.

• 20:21Yann LeCun?

  And so that allows the system not to predict the second word, right? And once you do this, you shift it into the input, etc. That's called autoregressive prediction, which is why those L L M should be called autoregressive L L M's.

• 20:36Yann LeCun?

  But we just call them L L M's. And there is a difference between this kind of process and a process by which, before producing a word, when you talk, when you and I talk, you and I are bilingual. We think about what we're going to say, and it's relatively independent of the language in which we're going to say it. When we when we talk about, like, I don't know, let's say a mathematical concept or something, the kind of thinking that we're doing and the answer that we're planning to produce is not linked to whether we're going to say it in French or Russian or English.

• 21:12Lex Fridman?

  Chomsky just rolled his eyes, but I understand. So you're saying that there's a a bigger abstraction that that's that goes before language.

• 21:21Yann LeCun?

  Yeah.

• 21:21Lex Fridman?

  That maps onto language.

• 21:22Yann LeCun?

  Right. It's certainly true for a lot of thinking that we that we do.

• 21:26Lex Fridman?

  Is that obvious that we don't? Like, you're saying your thinking is same in French as it is in English?

• 21:33Yann LeCun?

  Yeah, pretty much.

• 21:34Lex Fridman?

  Pretty much, or is this like, how flexible are you? Like, if there's a probability distribution?

• 21:41Yann LeCun?

  Well, it depends what kind of thinking, right? If it's just, if it's like producing puns, I get much better in French than English about that.

• 21:49Lex Fridman?

  No, but so right, right. Is there an abstract representation of puns? Like, is your humor an abstract? Like, when you tweet, and your tweets are sometimes a little bit spicy. What's, is there an abstract representation in your brain of a tweet before it maps onto English?

• 22:04Yann LeCun?

  There is an abstract representation of imagining the reaction of a reader to that text.

• 22:11Lex Fridman?

  Or you start with laughter and then figure out how to make that happen?

• 22:14Yann LeCun?

  Or figure out like a reaction you want to cause and then figure out how to say it, right? So that it causes that reaction. But that's like really close to language. But think about like a mathematical concept. Or, you know, imagining, you know, something you want to build out of wood or something like this, right? The kind of thinking you're doing is absolutely nothing to do with language, really. Like, it's not like you have necessarily like an internal monologue in any particular language. You're, you know, imagining mental models of the thing, right? I mean, if I, if I ask you to like imagine what this water bottle will look like if I rotate it 90 degrees. That has nothing to do with language.

• 22:55Yann LeCun?

  And so, so clearly there is, you know, a more abstract level of representation in which we, we do most of our thinking and we plan what we're gonna say if the output is, is,

• 23:10Yann LeCun?

  you know, uttered words as opposed to an output being, you know, muscle actions, right?

• 23:19Yann LeCun?

  We, we plan our answer before we produce it. And L L M's don't do that. They just produce one word after the other instinctively, if you want. It's like, it's a bit like the, you know, subconscious actions where you don't, like you're distracted, you're doing something, you're completely concentrated. And someone comes to you and, you know, asks you a question and you kind of answer the question. You don't have time to think about the answer, but the answer is easy. So you don't need to pay attention. You sort of respond automatically. That's kind of what an L L M does, right? It doesn't think about its answer, really. It retrieves it because it's accumulated a lot of knowledge. So it can retrieve some, some things, but it's going to. Just spit out one token after the other without planning the answer.

• 24:05Lex Fridman?

  But you're making it sound just one token after the other, one token at a time generation is bound to be simplistic. But if the world model is sufficiently sophisticated, that one token at a time,

• 24:24Lex Fridman?

  the most likely thing it generates is a sequence of tokens is going to be a deeply profound thing.

• 24:31Yann LeCun?

  Okay, but then that assumes that the systems actually possess an eternal world model.

• 24:37Lex Fridman?

  So it really goes to the, I think the fundamental question is, can you build a really

• 24:45Lex Fridman?

  complete world model, not complete, but a one that has a deep understanding of the world?

• 24:51Yann LeCun?

  Yeah, so can you build this, first of all, by prediction?

• 24:56Lex Fridman?

  Right.

• 24:57Yann LeCun?

  And the answer is probably yes. Can you predict, can you build it by predicting words? And the answer is most probably no, because language is very poor in terms of weak or low bandwidth, if you want, there's just not enough information there. So building world models means observing the world and understanding why the world is evolving the way it is. And then the extra component of a world model is something that can predict how the world is going to evolve as a consequence of an action you might take, right?

• 25:38Yann LeCun?

  So what model really is, here is my idea of the state of the world at time t, here is an action I might take. What is the predicted state of the world at time t plus one? Now, that state of the world doesn't, does not need to represent everything about the world. It just needs to represent enough that's relevant for this planning of the action, but not necessarily all the details. Now, here is the problem, you're not going to be able to do this with generative models. So a GAN model is trained on video and we've tried to do this for 10 years. You take a video, show a system a piece of video and then ask it to predict the reminder of the video. Basically predict what's going to happen.

• 26:20Lex Fridman?

  One frame at a time, do the same thing as sort of the auto aggressive L L M's do, but for video.

• 26:26Yann LeCun?

  Right, either one frame at a time or a group of frames at a time. But yeah, a large video model if you want. The idea of doing this has been floating around for a long time and at at fair. Some colleagues and I have been trying to do this for about 10 years. And you can't, you can't really do the same trick as with L L M's because you know, L L M's, as I said. You can't predict exactly which word is going to follow a sequence of words, but you can predict the distribution of words. Now, if you go to video, what you would have to do is predict a distribution over all possible frames in a video. And we don't really know how to do that properly.

• 27:11Yann LeCun?

  We do not know how to represent distributions over high dimensional continuous spaces in ways that are useful. And that's that there lies the main issue. And the reason we can do this is because the world is incredibly more complicated and richer in terms of information than than text. Text is discrete. Video is high dimensional and continuous, a lot of details in this. So if I take a video of this room. And the video is, you know, a camera panning around. There is no way I can predict everything that's going to be in the room as I pan around.

• 27:52Yann LeCun?

  The system cannot predict what's going to be in the room as the camera is panning. Maybe it's going to predict this is, this is a room where there is a light and there is a wall and things like that. It can't predict what the painting on the wall looks like or what the texture of the couch looks like. Certainly not the texture of the carpet. So there's no way I can predict all those details. So the the way to handle this. Is one way to possibly to handle this, which we've been working for a long time, is to have a model that has what's called a latent variable. And the latent variable is fed to a neural net. And it's supposed to represent all the information about the world that you don't perceive yet.

• 28:30Yann LeCun?

  And that you need to augment the the system for the prediction to do a good job at predicting pixels, including the, you know, fine texture of the of the carpet and the couch and the painting on the wall.

• 28:49Yann LeCun?

  That has been a complete failure, essentially. And we've tried lots of things. We tried just straight neural nets. We tried GANs. We tried, you know, VAEs, all kinds of regularized autoencoders. We tried many things. We also tried those kind of methods to learn good representations of images or video

• 29:14Yann LeCun?

  that could then be used as input to, for example, an image classification system. And that also has basically failed, like all the systems that attempt to predict missing parts of an image or video, you know, from a corrupted version of it, basically. So I take an image or a video, corrupt it or transform it in some way, and then try to reconstruct the complete video or image from the corrupted version.

• 29:43Yann LeCun?

  And then hope that internally the system will develop good representations of images that you can use for object recognition, segmentation, whatever it is. That has been essentially a complete failure. And it works really well for text. That's the principle that is used for LLMs, right?

• 29:59Lex Fridman?

  So where's the failure exactly? Is it that it's very difficult to form a good representation of an image, like a good embedding of all? All the important information in the image? Is it in terms of the consistency of image to image to image to image that forms the video? Like, what is the... If you do a highlight reel of all the ways you failed, what's that look like?

• 30:23Yann LeCun?

  Okay, so the reason this doesn't work is, first of all, I have to tell you exactly what doesn't work, because there is something else that does work. So the thing that does not work is training the system to learn representations of images by training it to reconstruct a good image from a corrupted version of it. Okay, that's what doesn't work. And we have a whole slew of techniques for this that are, you know, variant of denoising autoencoders, something called MAE, developed by some of my colleagues at FAIR, Max Total Encoder. So it's basically like the, you know, LLMs or things like this, where you train the system by corrupting text, except you corrupt images.

• 31:08Yann LeCun?

  You remove patches from it and you train a gigantic neural net to reconstruct. The features you get are not good. And you know they're not good, because if you now train the same architecture, but you train it supervised with labeled data, with textual descriptions of images, et cetera, you do get good representations. And the performance on recognition tasks is much better than if you do this self-supervised pre-training.

• 31:35Lex Fridman?

  So the architecture is good?

• 31:37Yann LeCun?

  The architecture is good. The architecture of the encoder is good. Okay, but the fact that you train the system to reconstruct images does not lead it to produce, to learn good generic features of images.

• 31:49Lex Fridman?

  When you train in a self-supervised way?

• 31:51Yann LeCun?

  Self-supervised by reconstruction.

• 31:53Lex Fridman?

  Yeah, by reconstruction.

• 31:54Yann LeCun?

  Okay, so what's the alternative? The alternative is joint embedding.

• 32:00Lex Fridman?

  What is joint embedding? What are these architectures that you're so excited about?

• 32:03Yann LeCun?

  Okay, so now, instead of training a system to encode the image and then training it to reconstruct the full image from a corrupted version, you take the full image, you take the... Corrupted or transformed version, you run them both through encoders, which in general are identical, but not necessarily. And then you you train a predictor on top of those encoders to predict the representation of the full input from the representation of the corrupted one. Okay. So joint embedding, because you're you're taking the the full input and the corrupted version

• 32:45Yann LeCun?

  or transformed version, run them both through encoders, you get a joint embedding. And then you and then you're you're saying, can I predict the representation of the full one from the representation of the corrupted one? Okay.

• 32:59Yann LeCun?

  And I call this a JEPAS, so that means joint embedding predictive architecture, because there's joint embedding and there is this predictor that predicts the representation of the good guy from from the bad guy.

• 33:09Yann LeCun?

  And the big question is, how do you train something like this? And until five years ago, six years ago, we didn't have particularly good answers for how you train those things, except for one called contrastive, contrastive learning,

• 33:26Yann LeCun?

  where

• 33:28Yann LeCun?

  and the idea of contrastive learning is you you take a pair of images that are, again, an image and a corrupted version or degraded version somehow, or transformed version of the original one. And you train the predicted representation to be the same as as that. If you only do this, the system collapses. It basically completely ignores the input and produces representations that are constant.

• 33:52Yann LeCun?

  So the contrastive methods avoid this. And and those things have been around since the early 90s. I had a paper on this in 1993.

• 34:01Yann LeCun?

  Is you also show pairs of images that you know are different. And then you push away the representations from each other. So you say, not only do representations of things that we know are the same, should be the same or should be similar, but representation of things that we know are different should be different. And that prevents the collapse, but it has some limitation. And there's a whole bunch of techniques that have appeared over the last six, seven years that can revive this this type of method. Some of them from FAIR, some of them from from Google and other places.

• 34:37Yann LeCun?

  But there are limitations to those contrastive method. What has changed in the last

• 34:43Yann LeCun?

  You know, three, four years is now now we have methods that are non-contrasting. So they don't require those negative contrastive samples of images that are that we know are different. You can only you you turn them only with images that are, you know, different versions or different views of the same thing. And you rely on some other tweaks to prevent the system from collapsing. And we have half a dozen different methods for this now.

• 35:08Lex Fridman?

  So what is the fundamental difference between joint embedding architectures and L L Ms? So can can JAP take us to AGI? But whether we should say that you don't like the term AGI, and we'll probably argue, I think every single time I've talked to you with argued about the G in AGI.

• 35:29Yann LeCun?

  Yes.

• 35:30Lex Fridman?

  Like, I get it. I get it. Well, we'll probably continue to argue about it. It's great. You you like, I mean, this because because you like French and I mean, is is, I guess, friend in French.

• 35:44Yann LeCun?

  Yes.

• 35:44Lex Fridman?

  And A M I stands for advanced machine intelligence.

• 35:48Yann LeCun?

  Right.

• 35:50Lex Fridman?

  But either way, can JAP take us to that towards that advanced machine intelligence?

• 35:55Yann LeCun?

  Well, so it's a it's a first step. Okay. So first of all, what's the difference with generative architectures like L L Ms? So L L Ms or vision systems that are trained by reconstruction generate the inputs, right? They generate the original input that is non corrupted, non transformed, right? So you have to predict all the pixels. And there is a huge amount of resources spent in the system to actually predict all those pixels, all the details. In a JAP, you're not trying to predict all the pixels. You're only trying to predict an abstract representation of of the inputs, right? And that's much easier

• 36:40Yann LeCun?

  in many ways. So what the JAP system, when it's being trained, is trying to do is extract as much information as possible from the input, but yet only extract information that is relatively easily predictable.

• 36:54Yann LeCun?

  Okay, so there's a lot of things in the world that we cannot predict. Like, for example, if you have a self-driving car driving down the street or road, there may be trees around the road. And it could be a windy day, so the leaves on the tree are kind of moving in kind of semi-chaotic, random ways that you can't predict and you don't care. You don't want to predict. So what you want is your encoder to basically eliminate all those details. It will tell you there's moving leaves, but it's not going to keep the details of exactly what's going on. And so when you do the prediction in representation space, you're not going to have to predict every single pixel of every leaf.

• 37:32Yann LeCun?

  And that, you know, not only is a lot simpler, but also it allows the system to essentially learn an abstract representation of the world where, you know, what can be modeled and predicted is preserved, and the rest is viewed as noise and eliminated by the encoder. So it kind of lifts the level of abstraction of the representation. If you think about this, this is something we do absolutely all the time. Whenever we describe a phenomenon, we describe it at a particular level of abstraction. And we don't always describe every natural phenomenon in terms of quantum field theory. That would be impossible, right? So we have multiple levels of abstraction to describe what happens in the world,

• 38:14Yann LeCun?

  you know, starting from quantum field theory to like atomic theory and molecules, you know, and chemistry, materials, and, you know, all the way up to, you know, kind of concrete objects in the real world and things like that. So we can't just only model everything at the lowest level. And that's what the idea of JEPa is really about. Learn abstract representation in a self-supervised manner. And, you know, you can do it hierarchically as well. So that, I think, is an essential component of an intelligent system. And in language, we can get away with that doing this because language is already to some level abstract and already has eliminated a lot of information that is not predictable.

• 38:59Yann LeCun?

  And so we can get away without doing the quantum embedding, without, you know, lifting the abstraction level and by directly predicting words.

• 39:09Lex Fridman?

  So joint embedding, it's still generative, but it's generative in this abstract representation space.

• 39:16Yann LeCun?

  Yeah.

• 39:16Lex Fridman?

  And you're saying, language, we were lazy with language because we already got the abstract representation for free. And now we have to zoom out, actually think about generally intelligent systems, we have to deal with a full mess of physical reality of reality. And you can't, you do have to do this step of jumping from the full, rich, detailed reality to a abstract representation of that reality, based on which you can reason and all that kind of stuff.

• 39:50Yann LeCun?

  Right. And the thing is, those self-supervised algorithms that learn by prediction, even in representation space, they learn more concept if the input data you feed them is more redundant. The more redundancy there is in the data, the more they're able to capture some internal structure of it. And so there, there is way more redundancy in the structure in perceptual inputs, sensory input like vision than there is in text, which is not nearly as redundant. This is back to the question you were asking a few minutes ago. Language might represent more information, really, because it's already compressed. You're right about that, but that means it's also less redundant.

• 40:32Yann LeCun?

  And so self-supervision will not work as well.

• 40:36Lex Fridman?

  Is it possible to join the self-supervised training on visual data and self-supervised training on language data? There is a huge amount of knowledge, even though you talked down about those 10 to the 13 tokens. Those 10 to the 13 tokens represent the entirety, a large fraction of what us humans have figured out. Both the shit talk on Reddit and the contents of all the books and the articles and the full spectrum of human intellectual creations. So is it possible to join those two together?

• 41:14Yann LeCun?

  Well, eventually, yes, but I think if we do this too early, we run the risk of being tempted to cheat. And in fact, that's what people are doing at the moment with vision language model. We're basically cheating. We're using language as a crutch to help the deficiencies of our vision systems to kind of learn good representations from images and video. And the problem with this is that we might, you know, improve our vision language system a bit, I mean, our language models by, you know, feeding them images. But we're not going to get to the level of even the intelligence or level of understanding

• 41:55Yann LeCun?

  of the world of a cat or a dog, which doesn't have language. You know, they don't have language and they understand the world much better than any LLM. They can plan really complex actions and sort of imagine the result of a bunch of actions. How do we get machines to learn that before we combine that with language? Obviously, if we combine this with language, this is going to be a winner. But but before that, we have to focus on like, how do we get systems to learn how the world works?

• 42:26Lex Fridman?

  So this kind of joint embedding predictive architecture, for you, that's going to be able to learn something like common sense, something like what a cat uses to predict how to mess with its owner most optimally by knocking over a thing?

• 42:42Yann LeCun?

  That's that's the hope. In fact, the techniques we're using are non-contrastive. So not only is the architecture non-generative, the learning procedures we're using are non-contrastive. So we have two, two sets of techniques. One set is based on distillation and there's a number of methods that use this principle. One by DeepMind called BYOL, a couple by by FAIR, one one called V-CREG, and another one called IJAP. And V-CREG, I should say, is not a distillation method, actually, but IJAP and BYOL certainly are. And there's another one also called DINO or DINO, also produced from at FAIR. And the idea of those things is that you take the full input, let's say, an image.

• 43:28Yann LeCun?

  You run it through an encoder, produces a representation. And then you corrupt that input or transform it, run it through the, essentially what amounts to the same encoder with some minor differences. And then train a predictor, sometimes the predictor is very simple, sometime doesn't exist, but train a predictor to predict a representation of the first uncorrupted input from the corrupted input. But you only train the, the second branch. You only train the part of the network that is fed with the corrupted input. The other network, you don't, you don't train, but since they share the same weight, when you modify the first one, it also modifies the second one.

• 44:11Yann LeCun?

  And with various tricks, you can prevent the system from collapsing, with the collapse of the type I was explaining before, where the system physically ignores the input. So that works very well. The, the technique with, the two techniques we developed at FAIR, Dino and, and IJAPPA work really well for that.

• 44:32Lex Fridman?

  So what kind of data are we talking about here?

• 44:34Yann LeCun?

  So there's, there's several scenarios. One, one scenario is you take an image, you corrupt it by changing the cropping, for example, changing the size a little bit, maybe changing the orientation, blurring it, changing the colors, doing all kinds of horrible things to it.

• 44:52Lex Fridman?

  But basic horrible things?

• 44:54Yann LeCun?

  Basic horrible things that sort of degrade the quality a little bit and change the framing, you know, crop the image.

• 45:02Yann LeCun?

  Or, and in some cases, in the case of IJAPPA, you don't need to do any of this, you just, you just mask some parts of it, right? You just basically remove some regions, like a big block essentially. And, and then, you know, run through the encoders and train the entire system, encoder and predictor, to predict the representation of the good one from the representation of the corrupted one.

• 45:26Yann LeCun?

  So that's the IJAPPA, doesn't need to know that it's an image, for example, because the only thing it needs to know is how to do this masking. Whereas with Dino, you need to know it's an image because you need to do things like, you know, geometry transformation and blurring and things like that that are really image specific. A more recent version of this that we have is called VJAP, so it's basically the same idea as IJAP, except it's applied to video. So now you take a whole video and you mask a whole chunk of it. And what we mask is actually kind of a tempered tube, so an all, like a whole segment of each frame in the video over the entire video.

• 46:03Lex Fridman?

  And that tube was like statically positioned throughout the frames, just literally a straight tube?

• 46:08Yann LeCun?

  The tube, yeah, typically is 16 frames or something and we mask the same region over the entire 16 frames. It's a different one for every video, obviously. And then again, train that system so as to predict the representation of the full video from the partially masked video. That works really well. It's the first system that we have that learns good representations of video so that when you feed those representations to a supervised classifier head, it can, it can tell you what action is taking place in the video with, you know, pretty good accuracy. So that, that's the first time we get something of that, of that quality.

• 46:48Lex Fridman?

  That's a good test that a good representation is formed. That means there's something to this.

• 46:53Yann LeCun?

  Yeah. We have also preliminary result that seem to indicate that the representation allows us, allow our system to tell whether the video is physically possible or completely impossible because some object disappeared or an object, you know, suddenly jumped from one location to another or, or changed shape or something.

• 47:14Lex Fridman?

  So it's able to capture some physical, some physics based constraints about the reality represented in the video?

• 47:22Yann LeCun?

  Yeah.

• 47:22Lex Fridman?

  About the appearance and the disappearance of objects?

• 47:25Yann LeCun?

  Yeah. That's really new.

• 47:28Lex Fridman?

  Okay, but can this actually?

• 47:32Lex Fridman?

  Get us to this kind of world model that understands enough about the world to be able to drive a car.

• 47:41Yann LeCun?

  Possibly. This is going to take a while before we get to that point, but

• 47:47Yann LeCun?

  there are systems already, robotic systems that are based on this idea.

• 47:53Yann LeCun?

  And what you need for this is a slightly modified version of this, where imagine that you have

• 48:02Yann LeCun?

  a video and a complete video. And what you're doing to this video is that you're either translating it in time towards the future. So you only see the beginning of the video, but you don't see the latter part of it that is in the original one.

• 48:17Yann LeCun?

  Or you just mask the second half of the video, for example. And then you you train a JEPA system of the type I described to predict the representation of the full video from the shifted one. But you also feed the predictor with an action. For example, you know, the wheel is turned 10 degrees to the right or something. Right? So if it's a, you know, a dash cam in a car and you know, the angle of the wheel, you should be able to predict to some extent what's going to go, what's going to happen to what you see. You're not going to be able to predict all the details of, you know, objects that appear in the view, obviously, but at a abstract representation level, you can, you can probably predict what's going to happen.

• 49:01Yann LeCun?

  So now what you have is a internal model that says, here is my idea of state of the world at time T. Here is an action I'm taking. Here is a prediction of the state of the world at time T plus one, T plus delta T, T plus two seconds, whatever it is. If you have a model of this type, you can use it for planning. So now you can do what L L M's cannot do. Which is planning what you're gonna do so as to arrive at a particular outcome or satisfy a particular objective, right? So you can have a number of objectives, right? If, you know, I can, I can predict that if I have an object like this, right?

• 49:43Yann LeCun?

  And I open my hand, it's gonna fall, right? And, uh, and if I push it with a particular force on the table, it's gonna move. If I push the table itself, it's probably not gonna move with the same force. So we have, we have this internal model of the world in our, in our mind, which allows us to plan sequences of actions to arrive at a particular goal. And so, uh, so now if you have this world model, we can imagine a sequence of actions, predict what the outcome of the sequence of action is going to be, measure to what extent the final state satisfies a particular objective,

• 50:23Yann LeCun?

  like, you know, moving the bottle to the left of the table, and then plan a sequence of actions that will minimize this objective at runtime. We're not talking about learning, we're talking about inference time, right? So this is planning, really. And in optimal control, this is a very classical thing. It's called model predictive control. You have a model of the system you want to control that, you know, can predict the sequence of states corresponding to a sequence of commands. And you're planning a sequence of commands so that according to your world model, the, the, the end state of the system will satisfy an objective that you fixed.

• 51:03Yann LeCun?

  This is the way, uh, you know, rocket trajectories have been planned since computers have been around, so since the early 60s, essentially.

• 51:12Lex Fridman?

  So yes, for model predictive control, but you also often talk about hierarchical planning.

• 51:18Yann LeCun?

  Yeah.

• 51:19Lex Fridman?

  Can hierarchical planning emerge from this somehow?

• 51:21Yann LeCun?

  Well, so no, you, you will have to build a specific architecture to allow for hierarchical planning. So hierarchical planning is absolutely necessary if you want to plan complex actions. If I wanna go from, let's say, from New York to Paris, this is the example I use all the time, and I'm sitting in my office at NYU, my objective that I need to minimize is my distance to Paris at a high level. A very abstract representation of my my location. I would have to decompose this into two subgoals. First one is go to the airport. Second one is catch a plane to Paris. Okay, so my subgoal is now going to the airport. My objective function is my distance to the airport. How do I go to the airport?

• 52:06Yann LeCun?

  Or I have to go in the street and have a taxi, which you can do in New York. Okay, now I have another subgoal. Go down on the street. What that means, going to the elevator, going down the elevator, walk out the street. How do I go to the elevator? I have to stand up from my chair, open the door in my office, go to the elevator, push the button. How do I get up from my chair? Like, you know, you can imagine going down, all the way down to basically what amounts to millisecond by millisecond muscle control. Okay. And obviously, you're not going to plan your entire trip from New York to Paris

• 52:49Yann LeCun?

  in terms of millisecond by millisecond muscle control. First, that would be incredibly expensive, but it will also be completely impossible because you don't know all the conditions of what's going to happen. You know, how long it's going to take to catch a taxi or to go to the airport with traffic, you know. I mean, you would have to know exactly the condition of everything to be able to do this planning. And you don't have the information. So you have to do this hierarchical planning so that you can start acting and then sort of replanning as you go. And nobody really knows how to do this in AI. Nobody knows how to train a system to learn the appropriate multiple levels of representation so that hierarchical planning works.

• 53:34Lex Fridman?

  There's something like that already emerged. So like, can you use an LLM? State of the art LLM to get you from New York to Paris by doing exactly the kind of detailed set of questions that you just did, which is, can you give me a list of 10 steps I need to do to get from New York to Paris? And then for each of those steps, can you give me a list of 10 steps, how I make that step happen? And for each of those steps, can you give me a list of 10 steps to make each one of those until you're moving your individual muscles? Maybe not. Whatever you can actually act upon using your own mind.

• 54:14Yann LeCun?

  Right. So there's a lot of questions that are implied by this, right? So the first thing is, LLMs will be able to answer some of those questions down to some level of abstraction, under the condition that they've been trained with similar scenarios in their training set.

• 54:30Lex Fridman?

  They would be able to answer all of those questions, but some of them may be hallucinated, meaning non-factual.

• 54:37Yann LeCun?

  Yeah, true. I mean, they will probably produce some answer, except they're not going to be able to really kind of produce millisecond by millisecond muscle control of how you stand up from your chair, right? So, but down to some level of abstraction, where you can describe things by words, they might be able to give you a plan, but only under the condition that they've been trained to produce those kind of plans, right? They're not going to be able to plan for situations where that they never encountered before. They basically are going to have to regurgitate a template that they've been trained on.

• 55:05Lex Fridman?

  But where, like, just for the example of New York to Paris, is it going to start getting into trouble? Like, at which layer of abstraction do you think you'll start? Because like, I can imagine almost every single part of that an LLM will be able to answer somewhat accurately, especially when you're talking about New York and Paris, major cities.

• 55:23Yann LeCun?

  So, I mean, certainly, LLM would be able to solve that problem if you find your need for it, you know, just, and so, I can't say that an LLM cannot do this. It can do this if you train it for it, there's no question, down to a certain level. Where things can be formulated in terms of words. But like, if you want to go down to like, how you, you know, climb down the stairs or just stand up from your chair in terms of words, like you, you can't, you can't do it. You, you, you need, that's one of the reasons you need experience of the physical world, which is much higher bandwidth than what you can express in words, in human language.

• 56:03Lex Fridman?

  So everything we've been talking about on the joint embedding space. Is it possible that that's what we need for like, the interaction with physical reality for, on the robotics front? And then just the L L M's are the thing that sits on top of it for the bigger reasoning about like, the fact that I need to book a plane ticket and I need to know, I know how to go to the websites and so on.

• 56:26Yann LeCun?

  Sure. And, you know, a lot of plans that people know about, that are relatively high level, are actually learned. They're not, people, most people don't invent the, you know, plans, they, they, by themselves, they, you know, we have some ability to do this, of course, obviously, but, but most plans that people use are plans that have been trained on, like they've seen other people use those plans or they've been told how to do things, right? That you can't invent how you, like, take a person who's never heard of airplanes and tell them like, how do you go from New York to Paris? And they're probably not going to be able to kind of, you know, deconstruct the whole plan unless they've seen examples of that before.

• 57:11Yann LeCun?

  So certainly L L M's are going to be able to do this, but, but then, how you link this from the low level of, of, of actions, that needs to be done with things like, like JEP that basically lift the abstraction level of the representation without attempting to reconstruct every detail of the situation. That's why we need JEPs for.

• 57:33Lex Fridman?

  I would love to sort of linger on your skepticism around auto-aggressive L L M's. So, one way I would like to test that skepticism is, everything you say makes a lot of sense.

• 57:50Lex Fridman?

  But if I apply everything you said today and in general to like, I don't know, 10 years ago, maybe a little bit less, no, let's say three years ago, I wouldn't be able to predict the success of LLMs. So does it make sense to you that autoregressive LLMs are able to be so damn good?

• 58:12Yann LeCun?

  Yes.

• 58:14Lex Fridman?

  Can you explain your intuition? Because if I were to take your wisdom and intuition at face value, I would say there's no way autoregressive LLMs, one token at a time, would be able to do the kind of things they're doing.

• 58:29Yann LeCun?

  No, there's one thing that autoregressive LLMs or that LLMs in general, not just the autoregressive one, but including the BERT-style bidirectional ones, are exploiting, and it's self-supervised learning. And I've been a very, very strong advocate of self-supervised learning for many years. So those things are an incredibly impressive demonstration that self-supervised learning actually works. The idea that, you know, started, it didn't start with BERT, but it was really kind of a good demonstration with this. So the idea that, you know, you take a piece of text, you corrupt it, and then you train some gigantic neural net to reconstruct the parts that are missing, that has been an

• 59:12Yann LeCun?

  enormous, produced an enormous amount of benefits. It allowed us to create systems that understand language, systems that can translate hundreds of languages in any direction, systems that are multilingual, so they're not, it's a single system that can be trained to understand hundreds of languages and translate in any direction and produce summaries and then answer questions and produce text. And then there's a special case of it where, you know, you, which is the autoregressive trick, where you constrain the system to not elaborate a representation of the text from looking at the entire text, but only...

• 59:58Yann LeCun?

  Predicting a word from the words that are come before, right? And you do this by constraining the architecture of the network. And that's what you can build an autoregressive LLM from. So there was a surprise many years ago with what was called decoder only LLM. So since, you know, systems of this type that are just trying to produce words from the, from the previous one. And, and the fact that when you scale them up, they, they tend to really kind of understand more about the, about language. When you train them on loss of data, you make them really big. That was kind of a surprise and that surprise occurred quite a while back, like, you know,

• 1:00:37Yann LeCun?

  with work from, you know, Google Meta, OpenAI, et cetera, you know, going back to, you know, the GPT kind of work general pre-trained transformers.

• 1:00:49Lex Fridman?

  You mean like GPT two? Like there's a certain place where you start to realize scaling might actually keep giving us an emergent benefit.

• 1:00:59Yann LeCun?

  Yeah, I mean, there were, there were work from, from various places, but

• 1:01:04Yann LeCun?

  if you want to kind of, you know, place it in the, in the GPT timeline, that would be around GPT two, yeah.

• 1:01:11Lex Fridman?

  Well, I just, cause you said it, you're so charismatic and you said so many words, but self-supervised learning, yes. But again, the same intuition you're applying to saying that autoregressive LLMs cannot have a deep understanding of the world. If we just apply that same intuition, does it make sense to you that they're able to form enough of a representation of the world to be damn convincing? Essentially passing the original Turing test with flying colors.

• 1:01:43Yann LeCun?

  Well, we're fooled by their fluency, right? We just assumed that if a system is fluent in manipulating language, then it has all the characteristics of human intelligence. But that impression is false. We're really fooled by it.

• 1:01:59Lex Fridman?

  What do you think Alan Turing would say? Without understanding anything, just hanging out with it?

• 1:02:04Yann LeCun?

  Alan Turing would decide that a Turing test is a really bad test. Okay, this is what the AI community has decided many years ago, that the Turing test was a really bad test of intelligence.

• 1:02:14Lex Fridman?

  What would Hans Moravec say about the large language models?

• 1:02:18Yann LeCun?

  Hans Moravec would say the Moravec paradox still applies.

• 1:02:22Lex Fridman?

  Okay.

• 1:02:22Yann LeCun?

  Okay, okay, we can.

• 1:02:25Lex Fridman?

  You don't think he would be really impressed?

• 1:02:26Yann LeCun?

  No, of course, everybody would be impressed, but you know, it's not a question of being impressed or not. It's a question of knowing what the limit of those systems can do. Like, again, they are impressive, they can do a lot of useful things. There's a whole industry that is being built around them. They're going to make progress, but there is a lot of things they cannot do, and we have to realize what they cannot do. And then figure out, you know, how we get there. And I'm not seeing this, I'm seeing this from basically, you know, 10 years of research on the idea of self-supervised learning. Actually, that's going back more than 10 years, but the idea of self-supervised learning.

• 1:03:08Yann LeCun?

  So basically capturing the internal structure of a piece of, of a set of inputs without training the system for any particular task, right? Learning representations. You know, the conference I co-founded 14 years ago is called International Conference on Learning Representations. That's the entire issue that deep learning is dealing with, right? And it's been my obsession for, you know, almost 40 years now. So, so learning representation is really the thing. For the longest time, we could only do this with supervised learning. And then we started working on, you know, what we used to call unsupervised learning. And sort of revived the idea of unsupervised learning in the early 2000s with Yoshua Benjou and Jeff Hinton.

• 1:03:52Yann LeCun?

  Then discovered that supervised learning actually works pretty well if you can collect enough data. And so the whole idea of, you know, unsupervised, self-supervised learning kind of took a. A backseat for a bit and then I kind of tried to revive it

• 1:04:08Yann LeCun?

  in a big way, you know, starting in 2014, basically when we started FAIR. And and really pushing for like finding new new methods to do self-supervised learning, both for text and for images and for video and audio. And some of that work has been incredibly successful. I mean, the reason why we have multilingual translation system, you know, things to do content moderation on on Meta, for example, on Facebook that are multilingual that understand whether a piece of text is hate speech or not or something. It's due to that progress using self-supervised learning for NLP, combining this with, you know, transformer architectures and and blah, blah, blah. But that's the big success of self-supervised learning. We had similar success in speech recognition.

• 1:04:51Yann LeCun?

  A system called Wave2Vec, which is also a joint embedding architecture, by the way, trained with contrastive learning and and that that system also can produce speech recognition systems that are multilingual with mostly unlabeled data and only need a few minutes of labeled data to actually do speech recognition. That's that's amazing.

• 1:05:12Yann LeCun?

  We have systems now based on those combination of ideas that can do real time translation of hundreds of languages into each other, speech to speech.

• 1:05:20Lex Fridman?

  Speech to speech, even including just fascinating languages that don't have written forms.

• 1:05:27Yann LeCun?

  That's right.

• 1:05:27Lex Fridman?

  They're spoken only.

• 1:05:28Yann LeCun?

  That's right. We don't go through text. It goes directly from from speech to speech using an internal representation of kind of speech units that are discrete. But it's it's called Textless NLP. We used to call it this way, but. Yeah, so that, I mean, incredible success there. And then, you know, for 10 years, we tried to apply this idea to learning representations of images by training a system to predict videos, learning intuitive physics by training a system to predict what's going to happen in the video. And tried and tried and failed and failed with generative models, with models that predict pixels.

• 1:06:02Yann LeCun?

  We could not get them to learn good representations of images. We could not get them to learn good representations of videos. We tried many times, we published lots of papers on it. Yeah, well, they kind of sort of work, but not really great. They started working, we abandoned this idea of predicting every pixel and basically just doing the joint embedding and predicting in representation space. That works. So there's ample evidence that we're not going to be able to learn good representations of the real world using generative model. So I'm telling people, everybody is talking about generative AI. If you're really interested in human level AI, abandon the idea of generative AI.

• 1:06:44Lex Fridman?

  Okay, but you really think it's possible to get far with a joint embedding representation? So like, there's common sense reasoning, and then there's high level reasoning. I feel like those are two, the kind of reasoning that LLMs are able to do. Okay, let me not use the word reasoning, but the kind of stuff that LLMs are able to do seems fundamentally different than the common sense reasoning we use to navigate the world.

• 1:07:13Yann LeCun?

  Yeah.

• 1:07:13Lex Fridman?

  It seems like we're gonna need both. You're not, would you be able to get with the joint embedding, with the JUPA type of approach, looking at video, would you be able to learn?

• 1:07:25Lex Fridman?

  Let's see, well, how to get from New York to Paris? Or how to understand the state of politics in the world today? Right? These are things where various humans generate a lot of language and opinions on, in the space of language, but don't visually represent that in any clearly compressible way.

• 1:07:48Yann LeCun?

  Right. Well, there's a lot of situations that, you know, might be difficult to, for a purely language-based system to know. Like, okay, you can probably learn from reading texts, the entirety of public, publicly available texts in the world, that I cannot get from New York to Paris by snapping my fingers. That's not gonna work, right?

• 1:08:09Lex Fridman?

  Yes.

• 1:08:11Yann LeCun?

  But there's probably sort of more complex scenarios of this type, which an LLM may never have encountered and may not be able to determine whether it's possible or not. So that link, you know, from the low level to the high level, the thing is that the high level that language expresses. Is based on a common experience of the low level, which LLMs currently do not have. You know, we, when we talk to each other, we know we have a common experience of the, of the world, like, you know, a lot of it is, is similar. And LLMs don't have that.

• 1:08:51Lex Fridman?

  But see, there is present. You and I have a common experience of the world in terms of the physics of how gravity works and stuff like this. And that common knowledge of the world, I feel like is there in the language. We don't explicitly express it, but if you have a huge amount of text, you're going to get this stuff that's between the lines. You're going, you're going, in order to form a consistent world model, you're going to have to understand how gravity works, even if you don't have an explicit explanation of gravity. So even though in the case of gravity, there is explicit explanations of gravity on Wikipedia.

• 1:09:32Lex Fridman?

  But you're like, the stuff that we think of as common sense reasoning, I feel like to generate language correctly, you're going to have to figure that out. Now, you could say, as you have, there's not enough text, sorry. Okay, so what? You don't think so?

• 1:09:50Yann LeCun?

  No, I agree with what you just said, which is that to be able to do high level common sense, to have high level common sense, you need to have the low level common sense to build on top of.

• 1:10:00Lex Fridman?

  Yeah. But that's not there.

• 1:10:03Yann LeCun?

  And that's not there in LLMs. LLMs are purely trained from text. So then the other statement you made, I would not, I would not agree with, the fact that implicit in all languages in the world is the underlying reality. There's a lot of that underlying reality, which is not expressed in language.

• 1:10:19Lex Fridman?

  Is that obvious to you?

• 1:10:20Yann LeCun?

  Yeah, totally.

• 1:10:23Lex Fridman?

  So like all, all the conversations we had, what, okay. There's the dark web, meaning whatever, the private conversations like DMs and stuff like this, which is much, much larger probably than what's available, what LLMs are trained on.

• 1:10:39Yann LeCun?

  You don't need to communicate the stuff that is common.

• 1:10:42Lex Fridman?

  But the humor, all of it. No, you do. Like when you, you don't need to, but it comes through. Like you've, like if I accidentally knock this over, you'll probably make fun of me. And in the content of the you making fun of me will be an explanation of the fact that cups fall. And then, you know, gravity works in this way. And then you'll have some very vague information about what kind of things explode when they hit the ground. And then maybe you'll make a joke about entropy or something like this. They will never be able to reconstruct this again. Like, okay, you'll make a little joke like this and there'll be trillion of other jokes. And from the jokes, you can piece together the fact that gravity works and mugs can break

• 1:11:24Lex Fridman?

  and all this kind of stuff. You don't need to see, uh, it'll be very inefficient. It's easier for like, knock the thing over.

• 1:11:34Yann LeCun?

  Yeah.

• 1:11:34Lex Fridman?

  But I feel like it would be there if you have enough of that data.

• 1:11:39Yann LeCun?

  I just think that most of the information of this type that we have accumulated when we were babies is just not present in, uh, in, in text, in any description essentially.

• 1:11:52Lex Fridman?

  And the sensory data is much, is a much richer source for getting that kind of understanding.

• 1:11:57Yann LeCun?

  I mean, that's the 16,000 hours of, of wake time of a four year old. And 10 to the 15 bytes, you know, going through vision, just vision, right? There is a similar, uh, bandwidth, you know, of touch and a little less through audio. And then text doesn't, language doesn't come in until like, you know, a year, uh, in, in life. And by the time you are nine years old, you've learned about gravity. You know about inertia, you know about gravity, you know, the stability, you know. You know about the distinction between animate and inanimate objects. You know, by eighteen months, you know about like, uh, white people want to do things and you help them if they can't, you know? I mean, it is a lot of things that you learn mostly by observation.

• 1:12:42Yann LeCun?

  Really, uh, not even through interaction. In the first few months of life, babies don't, don't really have any influence on the world. They can only observe, right? And you accumulate like a gigantic amount of, uh, of knowledge just, just from that. So that, that's what we're missing from, uh, current AI systems.

• 1:13:00Lex Fridman?

  I think in one of your slides, you have this nice plot that is one of the ways you show that LLMs are limited. I wonder if you could talk about hallucinations from your perspectives? The why hallucinations happen from large language models? And why and to what degrees that a fundamental flaw of large language models?

• 1:13:22Yann LeCun?

  Right, so because of the autoregressive prediction, every time an LLM produces a token or a word,

• 1:13:31Yann LeCun?

  uh, there is some level of probability for that word to take you out of the set of reasonable answers.

• 1:13:38Yann LeCun?

  Uh, and if you assume, which is a very strong assumption, that the probability of such error,

• 1:13:45Yann LeCun?

  uh, is that those errors are independent across a, a sequence of tokens being produced. What that means is that every time you produce a token, the probability that you rest, you, you stay within the, the set of correct answer decreases and it decreases exponentially.

• 1:14:01Lex Fridman?

  So there's a strong, like you said, assumption there that if, uh, there's a non-zero probability of making a mistake, which there appears to be, then there's going to be a kind of drift.

• 1:14:11Yann LeCun?

  Yeah. And that drift is exponential. It's like errors accumulate, right? So, so the probability that an answer would be nonsensical increases exponentially with the number of tokens.

• 1:14:24Lex Fridman?

  Is that obvious to you, by the way? Like, uh, well. So mathematically speaking, maybe, but like, isn't there a kind of gravitational pull towards the truth? Because on average, hopefully, the truth is well represented in the training set?

• 1:14:41Yann LeCun?

  No, it's basically a struggle against the curse of dimensionality. So the way you can correct for this is that you fine tune the system by having it produce answers for all kinds of questions that people might come up with. And people are people, so they, a lot of the questions that they have are very similar to each other. So you can probably cover, you know, eighty percent or whatever of questions that people will ask by, you know, collecting data. And then, and then you fine tune the system to produce good answers for all of those things. And it's probably going to be able to learn that because it's

• 1:15:21Yann LeCun?

  got a lot of capacity to to learn. But then there is, you know, the enormous set of prompts that you have not covered during training. And that set is enormous, like within the set of all possible prompts, the proportion of prompts that have been used for training is absolutely tiny. It's a tiny, tiny, tiny subset of all possible prompts. And so the system will behave properly on the prompts that has been either trained, pre-trained, or fine tuned. But then there is an entire space of things that it cannot possibly have been trained on because it's just, the number is gigantic. So, so whatever training the

• 1:16:05Yann LeCun?

  system has been subject to, to produce appropriate answers, you can break it by finding out a prompt that will be outside of the, the, the set of prompts it's been trained on, or things that are similar. And then it will just spew complete nonsense.

• 1:16:22Lex Fridman?

  When you say prompt, do you mean that exact prompt? Or do you mean a prompt that's like? In many parts, very different than like, is it that easy to ask a question or to say a thing that hasn't been said before on the internet?

• 1:16:39Yann LeCun?

  I mean, people have come up with things where like you, you put a essentially a random sequence of characters in the prompt, and that's enough to kind of throw the system into a mode where, you know, it is going to answer something completely different than it would have answered without this. So that's a way to jailbreak the system, basically get it, you know, go outside of its, uh, of its conditioning, right?

• 1:17:01Lex Fridman?

  So that's a very clear demonstration of it, but of course, you know, that's, uh, that goes outside of what is designed to do, right? If you actually stitch together reasonably grammatical sentences, is that, is it that easy to break it?

• 1:17:19Yann LeCun?

  Yeah, some people have done things like you, you, you write a sentence in English, right, that has, and, or you ask a question in English and it, it produces a perfectly fine answer. And then you just substitute a few words by the same word in another language. And all of a sudden, the answer is complete nonsense.

• 1:17:37Lex Fridman?

  Yes, so I guess what I'm saying is like, which fraction of prompts that humans are likely to generate are going to break the system?

• 1:17:47Yann LeCun?

  So the, the problem is that there is a long tail.

• 1:17:50Lex Fridman?

  Yes.

• 1:17:51Yann LeCun?

  This is a, an issue that a lot of people have realized, you know, in social networks and stuff like that, which is, uh, there is a very, very long tail of, of things that people will ask. And you can fine tune the system for the 80% or whatever of, uh, of the things that most people will, will ask. And then this long tail is, is so large that you're not going to be able to fine tune the system for all the conditions. And in the end, the system has been kind of a giant look up table, right? Essentially, which is not really what you want. You want systems that can reason, certainly that can plan. So the type of reasoning that takes place in LLM is very, very primitive. And the reason you can tell is primitive is because the amount of computation that is spent per token produced is constant.

• 1:18:36Yann LeCun?

  So if you ask a question and that question has an answer in a given number of token, the amount of computation devoted to computing that answer can be exactly estimated. It's like, you know, it's how it's the the size of the prediction network, you know, with its 36 layers or 92 layers or whatever it is, multiplied by the number of tokens. That's it. And so essentially, it doesn't matter if the question being asked is is simple to answer, complicated to answer, impossible to answer because it's a decidable or something. The amount of computation the system will be able to devote to that to the answer is constant or is proportional to the number of token produced in the answer.

• 1:19:21Yann LeCun?

  Right. This is not the way we work. The way we reason is that when we're faced with a complex problem or a complex question, we spend more time trying to solve it and answer it. Right. Because it's more difficult.

• 1:19:36Lex Fridman?

  There's a prediction element. There's an iterative element where you're like adjusting your understanding of a thing by going over and over and over. There's a hierarchical element, so on. Does this mean it's a fundamental flaw of LLMs? Or does it mean that there's more part to that question? Now you're just behaving like an LLM. Immediately answer. No, that that is just a low level world model. On top of which we can then build some of these kinds of mechanisms, like you said, persistent long term memory or reasoning, so on. But we need that world model that comes from language.

• 1:20:21Lex Fridman?

  Is it maybe it is not so difficult to build this kind of reasoning system on top of a well constructed world model?

• 1:20:29Yann LeCun?

  Okay, whether it's difficult or not, the near future will will say, because a lot of people are working on reasoning and planning abilities for for dialogue systems. I mean, if we're even if we restrict ourselves to language, just having the ability to plan your answer before your answer in terms that are not necessarily linked with the language you're going to use to produce the answer. Right. So this idea of this mental model that allows you to plan what you're going to say before you say it.

• 1:21:02Yann LeCun?

  That is very important. I think there's going to be a lot of systems over the next few years that are going to have this capability. But the blueprint of those systems would be extremely different from autoregulatory LLMs. So it's the same difference as the difference between what psychologists call system one and system two in humans, right? So system one is the type of task that you can accomplish without like deliberately consciously think about how you do them. You just do them. You've done them enough that you can just do it subconsciously, right? Without thinking about them. If you're an experienced driver, you can drive without really thinking about it and

• 1:21:43Yann LeCun?

  you can talk to someone at the same time or listen to the radio, right? If you are a very experienced chess player, you can play against a non-experienced chess player without really thinking either. You just recognize the pattern and you play, right? That's the system one. So all the things that you do instinctively without really having to deliberately plan and think about it. And then there is all the tasks where you need to plan. So if you are a not too experienced chess player or you are experienced, but you play against another experienced chess player, you think about all kinds of options, right? You think about it for a while, right? And you're much better if you have time to think about it than you are if you play blitz with limited time.

• 1:22:28Yann LeCun?

  So this type of deliberate planning, which uses your internal world model, that's system two. This is what LLMs currently cannot do. So how do we get them to do this, right? How do we build a system that can do this kind of planning that or reasoning that devotes more resources to complex problems than to simple problems? And it's not going to be autoregulatory prediction of tokens. It's going to be more something akin to inference of latent variables in, you know, what used to be called probabilistic models or graphical models and things of that type. So basically the principle is like this.

• 1:23:13Yann LeCun?

  You know, the prompt is like observed variables.

• 1:23:18Yann LeCun?

  And what the model does is that it's basically a measure of, it can measure to what extent an answer is a good answer for a prompt. Okay, so think of it as some gigantic neural net, but it's got only one output, and that output is a scalar number, which is, let's say, zero if the answer is a good answer for the question, and a large number if the answer is not a good answer for the question. Imagine you had this model. If you had such a model, you could use it to produce good answers. The way you would do is, you know, produce the prompt and then search through the space of possible answers for one that minimizes that number.

• 1:24:02Yann LeCun?

  That's called an energy-based model.

• 1:24:04Lex Fridman?

  But that energy-based model would need the model constructed by the LLM.

• 1:24:11Yann LeCun?

  Well, so really what you need to do would be to not search over possible strings of text that minimize that energy. But what you would do is do this in abstract representation space. So in sort of the space of abstract thoughts, you would elaborate a thought, right? Using this process of minimizing the output of your model, okay? Which is just a scalar. It's an optimization process, right? So now, the way the system produces its answer is through optimization,

• 1:24:45Yann LeCun?

  by minimizing an objective function, basically, right? And this is, we're talking about inference. We're not talking about training, right? The system has been trained already. So now we have an abstract representation of the thought, of the answer, representation of the answer. We feed that to, basically, to a reactive decoder, which can be very simple, that turns this into a text that expresses this thought. Okay, so that, in my opinion, is the blueprint of future dialogue systems.

• 1:25:14Yann LeCun?

  They will think about their answer, plan their answer by optimization, before turning it into text.

• 1:25:21Yann LeCun?

  And that is Turing complete.

• 1:25:24Lex Fridman?

  Can you explain exactly what the optimization problem there is? Like, what's the objective function? Just link on it, you kind of briefly described it. But over what space are you optimizing?

• 1:25:36Yann LeCun?

  The space of representations.

• 1:25:38Lex Fridman?

  Goes abstract representation.

• 1:25:39Yann LeCun?

  Abstract representation. So you have an abstract representation inside the system. You have a prompt. The prompt goes through an encoder, produces a representation, perhaps goes through a predictor that predicts a representation of the answer, of the proper answer. But that representation may not be a good answer, because there might be some complicated reasoning you need to do, right? So then you have another process that takes the representation of the answers and modifies it so as to minimize a cost function that measures to what extent the answer is a good answer for the question. Now, we sort of ignore the fact for, I mean, the issue for a moment of how you train that

• 1:26:23Yann LeCun?

  system to measure whether an answer is a good answer for a question.

• 1:26:28Lex Fridman?

  But suppose such a system could be created.

• 1:26:31Yann LeCun?

  Right.

• 1:26:31Lex Fridman?

  But what's the process, this kind of search-like process?

• 1:26:35Yann LeCun?

  It's an optimization process. You can do this if the entire system is differentiable, that scalar output is the result of, you know, running through some neural net, running the answer, the representation of the answers through some neural net, then by gradient descent, by backpropagating gradients, you can figure out like how to modify the representation of the answers so as to minimize that.

• 1:26:57Lex Fridman?

  So that's still a gradient-based?

• 1:26:59Yann LeCun?

  It's gradient-based inference. So now you have a representation of the answer in abstract space. Now you can turn it into text. Right? And the cool thing about this is that the representation now can be optimized through gradient descent, but also is independent of the language in which you're going to express the answer.

• 1:27:20Lex Fridman?

  Right. So you're operating in this abstract representation. I mean, this goes back to the joint embedding.

• 1:27:25Yann LeCun?

  Right.

• 1:27:25Lex Fridman?

  That is better to work in the, in the space of, I don't know, or to romanticize the notion, like, space of concepts versus

• 1:27:34Yann LeCun?

  Yeah.

• 1:27:34Lex Fridman?

  the space of concrete.

• 1:27:37Lex Fridman?

  Sensory information.

• 1:27:38Yann LeCun?

  Right.

• 1:27:40Lex Fridman?

  Okay, but can this do something like reasoning, which is what we're talking about?

• 1:27:44Yann LeCun?

  Well, not really, only in a very simple way. I mean, basically, you can think of those things as doing the kind of optimization I was talking about, except they optimize in the discrete space, which is the space of possible sequences of tokens. And they do it, they do this optimization in a horribly inefficient way, which is generate a lot of hypotheses and then select the best ones. And that's incredibly wasteful in terms of computation. Because you basically have to run your LLM for like every possible, you know, generated sequence. And it's incredibly wasteful.

• 1:28:21Yann LeCun?

  So it's much better to do an optimization in continuous space, where you can do gradient descent, as opposed to like generate tons of things and then select the best. You just iteratively refine your answer to go towards the best, right? That's much more efficient. But you can only do this in continuous spaces with differentiable functions.

• 1:28:40Lex Fridman?

  You're talking about the reasoning, like ability to think deeply or to reason deeply. How do you know what is an answer

• 1:28:53Lex Fridman?

  that's better or worse based on deep reasoning?

• 1:28:57Yann LeCun?

  Right, so then we're asking the question of, conceptually, how do you train an energy-based model, right? So an energy-based model is a function with a scalar output, just a number. You give it two inputs, X and Y, and it tells you whether Y is compatible with X or not. X, you observe. Let's say it's a prompt, an image, a video, whatever. And Y is a proposal for an answer, a continuation of a video, you know, whatever. And it tells you whether Y is compatible with X. And the way it tells you that Y is compatible with X is that the output of that function will be zero. If Y is compatible with X, it would be a positive number, non-zero, if Y is not compatible with X.

• 1:29:40Yann LeCun?

  Okay, how do you train a system like this? At a completely general level? Is you show it pairs of X and Y's that are compatible? A question and the corresponding answer. And you train the parameters of the big neural net inside to produce zero. Okay, now that doesn't completely work because the system might decide, well, I'm just gonna say zero for everything. So now you have to have a process to make sure that for a wrong Y, the energy would be larger than zero. And there you have two options. One is contrastive method. So contrastive method is you show an X and a bad Y. And you tell the system, well, that's, you know, give a high energy to this, like push up the energy, right?

• 1:30:23Yann LeCun?

  Change the weights in the neural net, the compute the energy so that it goes up. So that's contrastive methods. The problem with this is if the space of Y is large, the number of such contrastive samples you're gonna have to show is gigantic. But people do this. They do this when you train a system with RLHF. Basically, what you're training is what's called a reward model, which is basically an objective function that tells you whether an answer is good or bad. And that's basically exactly what this is. So we already do this to some extent. We're just not using it for inference. We're just using it for training.

• 1:31:06Yann LeCun?

  There is another set of methods which are non-contrastive, and I prefer those. And those non-contrastive methods basically say, okay, the energy function needs to have low energy on pairs of X, Y's that are compatible, that come from your training set. How do you make sure that the energy is gonna be higher everywhere else? And the way you do this is by having a regularizer, a criterion, a term in your cost function that basically minimizes the volume of space that can take low energy.

• 1:31:44Yann LeCun?

  And the precise way to do this is all kinds of different specific ways to do this depending on the architecture, but that's the basic principle. So that if you push down the energy function for particular regions in the X Y space, it will automatically go up in other places because it is only a limited volume of space that can take away energy. Okay. By the construction of the system or by the regularizer, regularizing function.

• 1:32:09Lex Fridman?

  We've been talking very generally, but what is a good X and a good Y? What is a good representation of X and Y? Because we've been talking about language and if you just take language directly, that presumably is not good. So there has to be some kind of abstract representation of ideas.

• 1:32:28Yann LeCun?

  Yeah, so you, I mean, you can do this with language directly by just, you know, X is a text and Y is a continuation of that text.

• 1:32:36Lex Fridman?

  Yes.

• 1:32:37Yann LeCun?

  Or X is a question, Y is the answer.

• 1:32:40Lex Fridman?

  But you're, you're saying that's not going to take it. I mean, that's going to do what L L M is doing.

• 1:32:45Yann LeCun?

  Well, no, it depends on how you, how the internal structure of the system is built. If the, if the internal structure of the system is built in such a way that inside of the system, there is a latent variable, let's call it Z, that you can manipulate so as to minimize the output energy. Then that Z can be viewed as a representation of a good answer that you can translate into a Y that is a good answer.

• 1:33:13Lex Fridman?

  So this kind of system could be trained in a very similar way.

• 1:33:17Yann LeCun?

  Very similar way, but you have to have this way of preventing collapse of, of ensuring that, you know, there is high energy for things you don't train it on.

• 1:33:27Yann LeCun?

  And, and currently it's, it's very implicit in L L M. It's done in a way that people don't realize it's being done, but it is, it is being done. It's, it's due to the fact that when you give a high probability to a, a word.

• 1:33:42Yann LeCun?

  Automatically you give low probability to other words because you only have a finite amount of probability to go around, right there, to sum to one. So when you minimize the cost entropy or whatever, when you train your LLM to produce the, to predict the next word, you're increasing the probability your system will give to the correct word, but you're also decreasing the probability it will give to the incorrect words. Now indirectly, that gives a low probability to, a high probability to sequences of words that are good and low probability to sequences of words that are bad, but it's very indirect. And it's not, it's not obvious why this actually works at all, but

• 1:34:21Yann LeCun?

  because you're not doing it on a joint probability of all the symbols in a, in a sequence. You're just doing it kind of

• 1:34:28Yann LeCun?

  you sort of factorize that probability in terms of conditional probabilities over successive tokens.

• 1:34:34Lex Fridman?

  So how do you do this for visual data?

• 1:34:36Yann LeCun?

  So we've been doing this with OJAPAR architectures, basically.

• 1:34:39Lex Fridman?

  The joint embedding.

• 1:34:40Yann LeCun?

  IJAPAR. So

• 1:34:43Yann LeCun?

  there, the compatibility between two things is, you know, here is, here is an image or a video. Here is a corrupted, shifted or transformed version of that image or video or masked. Okay. And then the energy of the system is the prediction error of the

• 1:35:02Yann LeCun?

  representation,

• 1:35:05Yann LeCun?

  the, the predicted representation of the good thing versus the actual representation of the good thing. Right? So, so you run the corrupted image to the system, predict the representation of the, the good input and corrupted, and then compute the prediction error. That's the energy of the system. So this system will tell you, this is a good,

• 1:35:26Yann LeCun?

  you know, if this is a good image and this is a corrupted version, it will give you zero energy if those two things are effectively, one of them is a corrupted version of the other. It gives you a high energy if the, if the two images are completely different.

• 1:35:39Lex Fridman?

  And hopefully that whole process gives you a really nice compressed representation of, of reality, of visual reality.

• 1:35:47Yann LeCun?

  And we know it does because then we use those representations as input to a classification system or something.

• 1:35:52Lex Fridman?

  That classification system works really nicely. Okay. Well, so to summarize, you recommend in a, in a

• 1:36:00Lex Fridman?

  In a spicy way that only Yann LeCun can, you recommend that we abandon generative models in favor of joint embedding architectures.

• 1:36:07Yann LeCun?

  Yes.

• 1:36:08Lex Fridman?

  Abandon autoregressive generation.

• 1:36:10Yann LeCun?

  Yes.

• 1:36:10Lex Fridman?

  Abandon prob... This feels like court testimony. Abandon probabilistic models in favor of energy-based models, as we talked about. Abandon contrastive methods in favor of regularized methods. And let me ask you about this. You've been for a while a critic of reinforcement learning.

• 1:36:29Yann LeCun?

  Yes.

• 1:36:29Lex Fridman?

  So the last recommendation is that we abandon RL in favor of model predictive control, as you were talking about, and only use RL when planning doesn't yield the predicted outcome. And we use RL in that case to adjust the world model or the critic.

• 1:36:48Yann LeCun?

  Yes.

• 1:36:48Lex Fridman?

  So you've mentioned RLHF, reinforcement learning with human feedback. Why do you still hate reinforcement learning?

• 1:36:58Yann LeCun?

  I don't hate reinforcement learning, and I think it's...

• 1:37:00Lex Fridman?

  I love, yes.

• 1:37:01Yann LeCun?

  I think it should not be abandoned completely, but I think its use should be minimized because it's incredibly inefficient in terms of samples. And so the proper way to train a system is to first have it learn good representations of the world and world models from mostly observation, maybe a little bit of interactions.

• 1:37:24Lex Fridman?

  And then steered based on that. If the representation is good, then the adjustments should be minimal.

• 1:37:29Yann LeCun?

  Yeah, now there's two things. You can use, if you've learned a world model, you can use the world model to plan a sequence of actions to arrive at a particular objective. You don't need RL unless the way you measure whether you succeed might be inexact.

• 1:37:45Yann LeCun?

  Your idea of, you know, whether you were going to fall from your bike

• 1:37:51Yann LeCun?

  might be wrong, or whether the person you were fighting with MMA was going to do something and then do something else.

• 1:37:58Yann LeCun?

  So there...

• 1:38:01Yann LeCun?

  So there's two ways you can be wrong. Either your objective function does not reflect the actual objective function you want to optimize, or your world model is inaccurate. Right, so you didn't, the prediction you were making about what was going to happen in the world is inaccurate. So if you want to adjust your world model while you are operating in the world, or your objective function, that is basically in the realm of RL. This is what RL deals with to some extent, right? So adjust your world model, and the way to adjust your world model even in advance is to explore parts of the space where your world model, where you know that your world model is inaccurate. That's called curiosity basically, or play, right?

• 1:38:46Yann LeCun?

  When you play, you kind of explore parts of the state space that, you know, you don't want to do in for real because it might be dangerous, but you can adjust your world model without killing yourself, basically. So that's what you want to use RL for. When it comes time to learning a particular task, you already have all the good representations, you already have your world model, but you need to adjust it for the situation at hand. That's when you use RL.

• 1:39:19Lex Fridman?

  What do you think RLHF works so well? This reinforcement learning with human feedback. Why did it have such a transformational effect on large language models that came before?

• 1:39:30Yann LeCun?

  What's had the transformational effect is human feedback. There is many ways to use it, and some of it is just purely supervised, actually. It's not really reinforcement learning.

• 1:39:40Lex Fridman?

  So it's the HF?

• 1:39:42Yann LeCun?

  It's the HF. And then there is various ways to use human feedback, right? So you can ask humans to rate answers, multiple answers that are produced by a world model. And then what you do is you train an objective function to predict that rating. And then you can use that objective function to predict, you know, whether an answer is good. And you can backpropagate gradient through this to fine tune your system so that it only produces highly rated answers. Okay, so that's one way. So that's like... In RL, that means training what's called a reward model, right? So something that, you know, basically a small neural net that estimates to what extent an answer is good, right?

• 1:40:27Yann LeCun?

  It's very similar to the objective I was talking about earlier for planning, except now it's not used for planning, it's used for fine tuning your system. I think it would be much more efficient to use it for planning, but currently it's used to fine tune the parameters of the system. Now, there's several ways to do this. You know, some of them are supervised. You just, you know, ask a human person like, what is a good answer for this? Right? And you just type the answer. I mean, there's lots of ways that those systems are being adjusted.

• 1:41:03Lex Fridman?

  Now, a lot of people have been very critical of the recently released Google's Gemini 1.5 for essentially, in my words, I could say, super woke, woke in the negative connotation of that word. There is some almost hilariously absurd things that it does, like it modifies history, like generating images of a black George Washington, or perhaps more seriously, something that you commented on Twitter, which is refusing to comment on or generate images of, or even

• 1:41:42Lex Fridman?

  descriptions of Tiananmen Square, or the Tank Man, one of the most sort of legendary protest images in history. And of course, these images are highly censored by the Chinese government, and therefore everybody started asking questions of what is the process of designing these LLMs? What is the role of censorship in these, and all that kind of stuff? So you commented on Twitter saying that open source is the answer.

• 1:42:17Yann LeCun?

  Yeah.

• 1:42:18Lex Fridman?

  Essentially, so can you explain?

• 1:42:22Yann LeCun?

  I actually made that comment on just about every social network I can. And I've I've I've made that point multiple times in various forums.

• 1:42:33Yann LeCun?

  Here's my my point of view on this. People can complain that AI systems are biased and they generally are biased by the distribution of the training data that they've been trained on. That reflects biases in society.

• 1:42:52Yann LeCun?

  And that is potentially offensive to some people. Or potentially not. And and some techniques to de-bias then become offensive to some people.

• 1:43:08Yann LeCun?

  Because of, you know, historical incorrectness and things like that.

• 1:43:15Yann LeCun?

  And so you can ask the question, you can ask two questions. The first question is, is it possible to produce an AI system that is not biased? And the answer is absolutely not. And it's not because of technological

• 1:43:29Yann LeCun?

  challenges, although there are technological challenges to that. It's because bias is in the eye of the beholder. Different people may have different ideas about what constitutes bias.

• 1:43:44Yann LeCun?

  You know, for a lot of a lot of things. I mean, there are facts that are, you know,

• 1:43:49Yann LeCun?

  indisputable, but there are a lot of opinions or things that can be expressed in different ways. And so you cannot have an unbiased system that's just an impossibility.

• 1:44:01Yann LeCun?

  And so what's the

• 1:44:04Yann LeCun?

  what's the answer to this? And the the answer is the same answer that we found in liberal democracy about the press. The press needs to be free and diverse.

• 1:44:18Yann LeCun?

  We have free speech for a good reason. It's because we don't want all of our information to be, to come from a unique source. Because that's opposite to the whole idea of democracy and, you know, progress of ideas and even science, right? In, in science, people have to argue for different opinions and, and science makes progress when people disagree and they come up with an answer and, you know, a consensus forms, right? And it's true in all democracies around the world. So there is a, a future, which is already happening, where every single one of our interaction with the digital world will be mediated by AI,

• 1:45:02Yann LeCun?

  AI systems, AI assistance, right? We're going to have smart glasses. You can already buy them from Meta, the Ray-Ban Meta, where, um, you know, you can talk to them and they are connected with an LLM and you can get answers on any question you have. Or you can be looking at a monument and there is a camera in the, in the system that in, in the glasses, you can ask it like, what can you tell me about this, building or this monument? You can be looking at a menu in a foreign language and it's thing will translate it for you or you can, we can do real time translation if you speak different languages. So a lot of our interactions with the digital world are going to be mediated by those systems in the near future.

• 1:45:45Yann LeCun?

  You know, increasingly the search engines that we're going to use are not going to be search engines. They're going to be dialogue systems that we just ask a question and it will answer and then point you to perhaps appropriate reference for it. But here is the thing, we cannot afford those systems to come from a handful of companies on the west coast of the US.

• 1:46:09Yann LeCun?

  Because those systems will constitute the repository of all human knowledge. And we cannot have that be controlled by a small number of people. Right? It has to be diverse, for the same reason the press has to be diverse. So how do we get a diverse set of AI assistants? It's very expensive and difficult to train a base model, right? A base LLM at the moment, you know, in the future it might be something different. But at the moment, that's an LLM. So only a few companies can do this properly. And if some of those top systems are open source, anybody can use them, anybody can fine tune them.

• 1:46:52Yann LeCun?

  If we put in place some systems that allows any group of people, whether they are individual citizens, groups of citizens, government organizations, NGOs, companies, whatever, to take those open source systems, AI systems, and fine tune them for their own purpose on their own data, then we're going to have a very large diversity of different AI systems that are specialized for all of those things, right? So I tell you, I talked to the French government quite a bit, and the French government will not accept that the digital diet of all their citizens

• 1:47:37Yann LeCun?

  be controlled by three companies on the west coast of the US. That's just not acceptable. It's a danger to democracy, regardless of how well intentioned those companies are, right? And so, and it's also a danger to local culture, to values, to language, right? I was talking with the founder of Infosys in India. He's funding a project to fine tune Llama 2, the open source model produced by Meta, so that Llama 2 speaks all 22 official languages in India. It's very important for people in India. I was talking to a former colleague of mine, Bustaf Asise,

• 1:48:21Yann LeCun?

  who used to be a scientist at Fair and then moved back to Africa. I created a research lab for Google in Africa and now has a new startup called Kara. And what he's trying to do is basically have LLM that speaks the local languages in Senegal so that people can have access to medical information because they don't have access to doctors. It's a very small number of doctors per capita in Senegal.

• 1:48:46Yann LeCun?

  I mean, you can't have any of this unless you have open source platforms. So with open source platforms, you can have AI systems that are not only diverse in terms of political opinions or things of that type, but in terms of language, culture, value systems, political opinions,

• 1:49:08Yann LeCun?

  technical abilities in various domains. And you can have an industry, an ecosystem of companies that fine-tune those open source systems for vertical applications in industry, right? You have, I don't know, a publisher has thousands of books and they want to build a system that allows a customer to just ask a question about the content of any of their books. You need to train on their proprietary data, right? You have a company, we have one within Meta, it's called Meta Mate, and it's basically an LLM that can answer any question about internal stuff about the company. Very useful. A lot of companies want this, right? A lot of companies want this not just for their employees, but also for their customers, to take care of their customers.

• 1:49:53Yann LeCun?

  So the only way you're going to have an AI industry, the only way you're going to have AI systems that are not uniquely biased is if you have open source platforms on top of which any group can build specialized systems. So the direction of inevitable direction of history is that the vast majority of AI systems will be built on top of open source platforms.

• 1:50:21Lex Fridman?

  So that's a beautiful vision. So meaning like a company like Meta or Google or so on should take only minimal fine-tuning steps after the building the foundation pre-trained model, as few steps as possible.

• 1:50:40Yann LeCun?

  Basically.

• 1:50:42Lex Fridman?

  Can Meta afford to do that?

• 1:50:44Yann LeCun?

  No.

• 1:50:44Lex Fridman?

  So I don't know if you you know this, but companies are supposed to make money somehow. And open source is is like giving away. I don't know, Mark made a video, Mark Zuckerberg, a very sexy video, talking about three hundred and fifty thousand Nvidia H one hundreds.

• 1:51:04Yann LeCun?

  Yeah.

• 1:51:05Lex Fridman?

  The the math of that is just for the GPUs, that's a hundred billion.

• 1:51:12Lex Fridman?

  Plus the infrastructure for training, everything. So I'm no business guy. But how do you make money on that? So the vision you paint is a really powerful one. But how is it possible to make money?

• 1:51:25Yann LeCun?

  OK, so you have several business models, right? The business model that Meta is built around is

• 1:51:35Yann LeCun?

  you offer a service. And the the financing of that service is either through ads or through business customers. So for example, if you have an L L M that, you know, can help a moment pizza place

• 1:51:53Yann LeCun?

  by, you know, talking to the customers through WhatsApp. And so the customers can just order a pizza and the system will just, you know, ask them like, what topping do you want or what size, blah, blah, blah.

• 1:52:05Yann LeCun?

  The business will pay for that. OK, that's the model.

• 1:52:12Yann LeCun?

  And otherwise, you know, if it's a system that is on the more kind of classical services, it can be ad supported or, you know, there are several models. But the point is, if you have a big enough

• 1:52:26Yann LeCun?

  potential customer base and you need to build that system

• 1:52:30Yann LeCun?

  anyway for them. It doesn't hurt you to actually distribute it in open source?

• 1:52:36Lex Fridman?

  Again, I'm no business guy. But if you release the open source model, then other people can do the same kind of task and compete on it. Basically, provide fine tuned models for businesses.

• 1:52:49Yann LeCun?

  Sure.

• 1:52:50Lex Fridman?

  Is the bet that Meta is making, by the way, I'm a huge fan of all this. But is the bet that Meta is making is like, we'll do a better job of it.

• 1:52:58Yann LeCun?

  Well, no, the the bet is is more, we have, we already have a huge user base and customer base.

• 1:53:05Lex Fridman?

  Ah, right.

• 1:53:06Yann LeCun?

  Right, so it's going to be useful to them. Whatever we offer them is going to be useful and there is a way to derive revenue from this.

• 1:53:13Lex Fridman?

  Sure.

• 1:53:15Yann LeCun?

  And it doesn't hurt that, you know, we provide that system or the base, the base model, right, the foundation model in open source for others to build applications on top of it, too. If those applications turn out to be useful for our customers, we can just buy it from them.

• 1:53:35Yann LeCun?

  It could be that they will improve the platform. In fact, we see this already. I mean, there is, you know, literally millions of downloads of Llama 2 and thousands of people who have, you know, provided ideas about how to make it better. So, you know, this this clearly accelerates progress to make the system available to

• 1:53:55Yann LeCun?

  it's sort of a wide community of people. And and there is literally thousands of businesses who are building applications with it. So so

• 1:54:06Yann LeCun?

  our ability to, Meta's ability to derive revenue from this technology is not impaired by the distribution of it, of base models in open source.

• 1:54:19Lex Fridman?

  The problem with the criticism that Gemini is getting is that, as you pointed out on the West Coast, just to just to clarify, we're currently in the East Coast, where I would suppose Meta AI headquarters would be.

• 1:54:32Lex Fridman?

  So they're strong words about the West Coast, but I guess the issue that happens is, I think it's fair to say that most tech people have a political affiliation with the left wing. They're, they lean left. And so the problem that people are criticizing Gemini with is that there's, in that de-biasing process that you mentioned, that their ideological lean

• 1:55:00Lex Fridman?

  Becomes obvious. Is this something that could be escaped? You're saying open source is the only way. Have you witnessed this kind of ideological lean that makes engineering difficult?

• 1:55:14Yann LeCun?

  No, I don't think it has to do. I don't think the issue has to do with the political leaning of the people designing those systems. It has to do with the acceptability or political leanings of their customer base or audience. Right? So a big company cannot afford to offend too many people. So they're going to make sure that whatever product they put out is safe, whatever that means. And it's very possible to overdo it. And it's also very possible to, it's impossible to do it properly for everyone. You're not going to satisfy everyone. So that's what I said before.

• 1:55:56Yann LeCun?

  You cannot have a system that is unbiased, that is perceived as unbiased by everyone. It's going to be, you know, you push it in one way, one set of people are going to see it as biased, and then you push it the other way, and another set of people is going to see it as biased. And then in addition to this, there's the issue of if you push the system perhaps a little too far in one direction, it's going to be non-factual, right? You're going to have, you know, you know, black Nazi soldiers in the.

• 1:56:24Lex Fridman?

  Yeah, so we should, we should mention image generation of, of black Nazi soldiers, which is not factually accurate.

• 1:56:31Yann LeCun?

  Right. And can be offensive for some people as well, right? So, so, you know, it's going to be impossible to kind of produce systems that are unbiased for everyone. So the only solution that I see is diversity.

• 1:56:44Lex Fridman?

  And diversity in the full meaning of that word, diversity of in every possible way.

• 1:56:50Yann LeCun?

  Yeah.

• 1:56:52Lex Fridman?

  Marc Andreessen just tweeted today.

• 1:56:56Lex Fridman?

  Let me do a TLDR. The conclusion is only startups and open source can avoid the issue that he's highlighting with big tech. He's asking, can big tech actually field generative AI products? One, ever escalating demands from internal activists, employee mobs, crazed executives, broken boards, pressure groups, extremist regulators, government agencies, the press, in quotes, experts and everything corrupting the output. Two, constant risk of generating a bad answer or drawing a bad picture or rendering a bad video. Who knows what is going to say or do at any moment? Three, legal exposure, product liability, slander,

• 1:57:40Lex Fridman?

  election law, many other things and so on. Anything that makes Congress mad. Four, continuous attempts to tighten grip unacceptable output, degrade the model, like how good it actually is in terms of usable and pleasant to use and effective and all that kind of stuff. And five, publicity of bad text, images, video, actual puts those examples into the training data for the next version, so on. So he just highlights how difficult this is from all kinds of people being unhappy. He said, you can't create a system that makes everybody happy.

• 1:58:17Yann LeCun?

  Yes.

• 1:58:18Lex Fridman?

  So if you're going to do the fine tuning yourself and keep a close source, essentially the problem there is then trying to minimize the number of people who are going to be unhappy.

• 1:58:29Yann LeCun?

  Yep.

• 1:58:30Lex Fridman?

  And you're saying like the only, that that almost impossible to do right. And that's the better way is to do open source.

• 1:58:38Yann LeCun?

  Basically, yeah. I mean, he's, Mark is right about a number of things that he lists that indeed scare large companies. You know, certainly congressional investigations is one of them. Legal liability, you know, making things that get people to, you know, hurt themselves or hurt others, like, you know, big companies are really careful about not. Producing things of this type and because they have, you know, they don't want to hurt anyone, first of all, and then second, they want to preserve their business. So it's essentially

• 1:59:18Yann LeCun?

  impossible for systems like this that can inevitably formulate political opinions and, you know, opinions about various things that may be political or not, but that people may disagree about, about, you know, moral issues and, you know, things about like questions about religion and things like that, right? Or cultural issues that people from different communities would disagree with in the first place. So there's only kind of a relatively small number of things that people will sort of agree on, you know, basic principles. But beyond that, if you, if you want those systems to be useful, they will necessarily have to offend a number of people inevitably.

• 2:00:01Lex Fridman?

  And so open source is just better and then

• 2:00:04Yann LeCun?

  Diversity is better, right?

• 2:00:06Lex Fridman?

  And open source enables diversity.

• 2:00:08Yann LeCun?

  That's right. Open source enables diversity.

• 2:00:10Lex Fridman?

  That's, this can be fascinating world where if it's true that the open source world, if Meta leads the way and creates this kind of open source foundation model world, there's going to be like governments will have a fine new model and

• 2:00:24Yann LeCun?

  Yeah.

• 2:00:26Lex Fridman?

  And, and then potentially, you know, people that vote left and right will have their own model and preference to be able to choose and it will potentially divide us even more, but that's on us humans. We get to figure out. Basically, the technology enables humans to human more effectively and all the difficult ethical questions that humans raise will just

• 2:00:53Lex Fridman?

  leave it up to us to figure it out.

• 2:00:55Yann LeCun?

  Yeah. I mean, there are some limits to what, you know, the same way there are limits to free speech, there has to be some limit to the kind of stuff that those systems might be authorized to, to produce, you know, some guardrails. So I mean, that's one thing I've been interested in, which is in the type of architecture that we were discussing before, where The output of a system is a result of an inference to satisfy an objective. That objective can include guardrails. And we can put guardrails in open source systems. I mean, if we eventually have systems that are built with this blueprint, we can put guardrails in those systems that guarantee that there is sort of a minimum

• 2:01:40Yann LeCun?

  set of guardrails that make the system non-dangerous and non-toxic, et cetera. You know, basic things that everybody would agree on.

• 2:01:47Yann LeCun?

  And then, you know, the fine tuning that people will add or the additional guardrails that people will add will kind of cater to their community, whatever it is.

• 2:01:57Lex Fridman?

  And yeah, the fine tuning will be more about the gray areas of what is hate speech, what is dangerous and all that kind of stuff. I mean, you've

• 2:02:05Yann LeCun?

  Or different value systems.

• 2:02:06Lex Fridman?

  Still value systems. I mean, like, but still, even with the objectives of how to build a bioweapon, for example, I think something you've commented on, or at least there's a paper where a collection of researchers is trying to understand the social impacts of these LLMs. And I guess one threshold is nice is like, does the LLM make it any easier than a than a search would, like a Google search would?

• 2:02:32Yann LeCun?

  Right, so the increasing number of studies on this seems to point to

• 2:02:40Yann LeCun?

  the fact that it doesn't help. So having an LLM doesn't help you design or build a bioweapon or a chemical weapon if you already have access to, you know, a search engine and their library.

• 2:02:55Yann LeCun?

  And so the sort of increased information you get or the ease with which you get it doesn't really help you. That's the first thing. The second thing is it's one thing to have a list of instructions of how to make a chemical weapon, for example, or bioweapon. It's another thing to actually build it. And it's much harder than you might think, and an underline will not help you with that.

• 2:03:18Yann LeCun?

  In fact, you know, nobody in the world, not even like, you know, countries used by your weapons, because most of the times they have no idea how to protect their own populations against it. So, so it's too dangerous actually to kind of ever use. And it's in fact banned by international treaties. Chemical weapons is different, it's also banned by treaties. But, but it's the same problem. It's difficult to use in situations that doesn't turn against the perpetrators. But we could ask Elon Musk, like, I can, I can give you a very precise list of instructions of how you build a rocket engine. And even if you have a team of 50 engineers that are really experienced building it,

• 2:04:00Yann LeCun?

  you're still gonna have to blow up a dozen of them before you get one that works. And, you know, it's the same with, you know, the chemical weapons or bioweapons or things like this. It requires expertise, you know, in the, in the real world that an underline is not going to help you with.

• 2:04:17Lex Fridman?

  And it requires even the common sense expertise that we've been talking about, which is how to take language-based instructions and materialize them in the physical world requires a lot of knowledge that's not in the instructions.

• 2:04:34Yann LeCun?

  Yeah, exactly. A lot of biologists have posted on this actually in response to those things, saying like, do you realize how hard it is to actually do the, the lab work? I can, this is not trivial.

• 2:04:44Lex Fridman?

  Yeah, and that's Hans Moravec comes, comes to light once again. Just to linger on Lama, you know, Mark announced that Lama 3 is coming out eventually. I don't think there's a release date, but what, what are you most excited about? First of all Lama 2, that's already out there. And maybe the future Lama 3, 4, 5, 6, 10. Just the, the future of the open source under Meta.

• 2:05:10Yann LeCun?

  Well, a number of things. So there's going to be like various versions of Llama that are, you know, improvements of previous Lamas, bigger, better, multimodal, things like that. And then in future generations, systems that are capable of planning, that really understand how the world works, maybe are trained from video, so they have some world model, maybe, you know, capable of the type of reasoning and planning I was talking about earlier. Like, how long is that going to take? Like, when is the research that is going in that direction going to sort of feed into the product line, if you want, of Llama? I don't know. I can't tell you. And there is, you know, a few breakthroughs that we

• 2:05:49Yann LeCun?

  have to basically go through before we can get there. But you'll be able to monitor our progress because we publish our research, right? So, you know, last week, we published the VJEPa work, which is sort of a first step towards training systems from video. And then the next step is going to be world models based on kind of this type of idea, training from video. There's similar work at DeepMind also, and taking place, people, and also at UC Berkeley on world models from video. A lot of people are working on this. I think a lot of good ideas are coming, are appearing. My bet is that those systems

• 2:06:33Yann LeCun?

  are going to be JEPa-like, they're not going to be generative models. And we'll see what the future will tell. There's really good work at, a gentleman called Daniel J. Hafner, who is now DeepMind, who's worked on kind of models of this type that learn representations and then use them for planning or learning tasks by reinforcement learning. And a lot of work at Berkeley by Peter Abiel, Sergio Levine, a bunch of other people of that type. I'm collaborating with, actually, in the context of some grants with my NYU hat.

• 2:07:12Yann LeCun?

  And then collaborations also through Meta, because the lab at Berkeley is associated with Meta in some way, so with FAIR. So I think it's very exciting. You know, I think... I'm super excited about, I haven't been that excited about like the direction of machine learning and AI, you know, since, you know, 10 years ago when Fairway started and before that, 30 years ago, when we were working on, let's say, 35, on, on convolutional nets and, and, and the early days of neural nets. So I'm super excited because I see a path towards potentially human level intelligence with, you know, systems that can understand

• 2:07:55Yann LeCun?

  the world, remember, plan, reason. There, there is some, some set of ideas to make progress there that might have a chance of working. And I'm really excited about this. What I like is that, you know, it somewhat, we, we get onto like a good direction and perhaps succeed before my brain turns to a white sauce or, or before I need to retire.

• 2:08:19Lex Fridman?

  Yeah, yeah. You're also excited by, are you, is it beautiful to you just the amount of GPUs involved? Sort of the, the, the whole training process on this much compute. It's just zooming out, just looking at Earth and humans together have built these computing devices and are able to train this one brain. Then, then we then open source, like giving birth to this open source brain trained on this gigantic compute system. There's just the details of how to train on that, how to build the infrastructure and the hardware,

• 2:09:02Lex Fridman?

  the cooling, all of this kind of stuff. Or are you just still, most of your excitement is in the, the theory aspect of it? The, meaning like the software?

• 2:09:11Yann LeCun?

  Well, I used to be a hardware guy many years ago.

• 2:09:14Lex Fridman?

  Yes, yes.

• 2:09:15Yann LeCun?

  Decades ago.

• 2:09:15Lex Fridman?

  Hardware has improved a little bit.

• 2:09:18Yann LeCun?

  Change a little bit.

• 2:09:19Lex Fridman?

  Yeah.

• 2:09:20Yann LeCun?

  I mean, certainly, scale is necessary, but not sufficient.

• 2:09:25Lex Fridman?

  Absolutely.

• 2:09:25Yann LeCun?

  So we certainly need computation. I mean, we're still far in terms of compute power from, you know, what we would need to match the compute power of the human brain. You know, this may occur in the next couple of decades, but we're still some ways away. And certainly in terms of power efficiency, we're really far. So a lot of progress to make in hardware. And, you know, right now, a lot of the progress is not, I mean, there's a bit coming from silicon technology, but a lot of it coming from architectural innovation and quite a bit coming from like more efficient ways of, you know, implementing the architectures that have become popular, basically a combination of transformers and convnets, right?

• 2:10:10Yann LeCun?

  And so, you know, there's still some ways to go until we're gonna saturate, we're gonna have to come up with like new, new principles, new fabrication technology, new basic components, perhaps, you know, based on sort of different principles than those classical digital CIMAs.

• 2:10:34Lex Fridman?

  Interesting. So you think in order to build AMI, I mean, we need, we potentially might need some hardware innovation too?

• 2:10:45Yann LeCun?

  Well, if you want to make it ubiquitous, yeah, certainly, because we're gonna have to reduce the, you know, compute, the power consumption. A GPU today, right, is half a kilowatt to a kilowatt. Human brain is about 25 watts. And a GPU is way below the power of human brain. You need, you know, something like a hundred thousand or a million to match it. So, so, you know, we are off by a huge factor.

• 2:11:14Lex Fridman?

  You often say that AGI is not coming soon, meaning like not this year, not the next few years, potentially farther away. What's your basic intuition behind that?

• 2:11:28Yann LeCun?

  So first of all, it's not going to be an event. Right? The idea somehow, which, you know, is popularized by science fiction in Hollywood, that, you know, somehow, somebody is going to discover the secret, the secret to AGI or human-level AI or AMI, whatever you want to call it. And then, you know, turn on a machine and then we have AGI. That's just not going to happen. It's not going to be an event. It's going to be gradual progress. Are we going to have systems that can learn from video, how the world works and learn good representations? Yeah. Before we get them to the scale and performance that we observe in humans, it's going to take quite a while. It's not going to happen in one day.

• 2:12:12Yann LeCun?

  Are we going to get systems that can have large amount of associative memory? So they can, they can remember stuff? Yeah, but same, it's not going to happen tomorrow. I mean, there is some basic techniques that need to be developed. We have a lot of them, but like, you know, to get this to work together with a full system is another story. Are we going to have systems that can reason and plan, perhaps along the lines of the objective-driven AI architectures that I described before? Yeah, but like before we get this to work, you know, properly, it's going to take a while. So, and before we get all those things to work together and then on top of this have systems that can learn, like hierarchical planning, hierarchical representations, systems that can be configured for a lot of different situations at hands, the way the human brain can.

• 2:12:57Yann LeCun?

  You know, all of this is going to take, you know, at least a decade and probably much more, because there are a lot of problems that we're not seeing right now, that we have not encountered. And so we don't know if there is an easy solution within this framework.

• 2:13:13Yann LeCun?

  So, you know, it's, it's not just around the corner. I mean, I've, I've been hearing people for the last 12, 15 years, claiming that, you know, AGI is just around the corner and being systematically wrong. And I knew they were wrong when they were saying it. I called their bullshit.

• 2:13:29Lex Fridman?

  Why do you think people have been calling? First of all, I mean, from the beginning, from the birth of the term artificial intelligence, there has been a eternal optimism. That's perhaps unlike other technologies. Is it a Marvex paradox? Is the explanation for why people are so optimistic about AGI?

• 2:13:49Yann LeCun?

  I don't think it's just Marvex paradox. Marvex paradox is a consequence of realizing that the world is not as easy as we think. So first of all, intelligence is not a linear thing that you can measure with a scalar, with a single number. You know, can you say that humans are smarter than orangutans? In some ways, yes. But in some ways, orangutans are smarter than humans in a lot of domains that allows them to survive in the forest, for example.

• 2:14:19Lex Fridman?

  So IQ is a very limited measure of intelligence. Do you intelligence is bigger than what IQ, for example, measures?

• 2:14:26Yann LeCun?

  Well, IQ can measure, you know, approximately something for humans. But because humans kind of, you know, come in relatively kind of uniform form, right?

• 2:14:43Yann LeCun?

  But it only measures one type of ability that, you know, may be relevant for some tasks, but not others. And but then if you're talking about other intelligent entities for which the, you know, the basic things that are easy to them is very different, then it doesn't mean anything. So intelligence is a collection of skills and an ability to acquire new skills efficiently.

• 2:15:15Yann LeCun?

  Right? And the collection of skills that an intelligent, particular intelligent entity possesses or is capable of learning quickly is different from the collection of skills of another one. And because it's a multidimensional thing, the set of skills is a high dimensional space, you can't measure, you can compare, you cannot compare two things as to whether one is more intelligent than the other. It's multidimensional.

• 2:15:41Lex Fridman?

  So you push back against what are called AI doomers a lot. Can you explain their perspective and why you think they're wrong?

• 2:15:52Yann LeCun?

  Okay, so AI doomers imagine all kinds of catastrophic scenarios of how AI could escape or control and basically kill us all. And that relies on a whole bunch of assumptions that are mostly false. So the first assumption is that the emergence of superintelligence could be an event that at some point we're going to have, we're going to figure out the secret and we'll turn on a machine that is superintelligent. And because we'd never done it before, it's going to take over the world and kill us all. That is false. It's not going to be an event. We're going to have systems that are like as smart as a cat, has all the, have all

• 2:16:33Yann LeCun?

  the characteristics of, you know, human level of intelligence, but their level of intelligence would be like a cat or parrot, maybe, or something.

• 2:16:44Yann LeCun?

  And then we're going to work our way up to kind of make those things more intelligent. And as we make them more intelligent, we're also going to put some guardrails in them and learn how to kind of put some guardrails so they behave properly. And we're not going to do this with just one, it's not going to be one effort, but it's going to be lots of different people doing this. And some of them are going to succeed at making intelligent systems that are controllable and safe and have the right guardrails. And if some other goes rogue, then we can use the, the good ones to go against the rogue ones. So it's going to be my, you know, smart AI police against your rogue AI. So it's not going to be like, you know, we're going to be exposed to like a single rogue AI that's going to kill us all. That's just not, not happening. Now, there is another fallacy, which is the fact that because a system is intelligent,

• 2:17:29Yann LeCun?

  it necessarily wants to take over.

• 2:17:33Yann LeCun?

  And there is several arguments that make people scared of this, which I think are completely false as well. So one of them is, you know, in nature, it seems to be that the more intelligent species are the one that end up dominating the other. And, and even, you know, extinguishing the others, sometimes by design, sometimes just by mistake. And and so, you know, there is sort of a thinking by which you say, well, if AI systems are more intelligent than us, surely they're going to eliminate us, if not by

• 2:18:13Yann LeCun?

  design, simply because they don't care about us. And that's just preposterous for for a number of reasons. First reason is they're not going to be a species. They're not going to be a species that competes with us. They're not going to have the desire to dominate, because the desire to dominate is something that has to be hardwired into an intelligent system. It is hardwired in humans. It is hardwired in baboons, in chimpanzees, in wolves, not in orangutans. The species in which this desire to dominate or submit or or attain status in other ways is is specific to social species. Non-social species

• 2:18:57Yann LeCun?

  like orangutans don't have it, right? And they are as smart as we are, almost, right?

• 2:19:02Lex Fridman?

  And to you, there's not significant incentive for humans to encode that into the AI systems. And to the degree they do, there'll be other AIs that sort of punish them for it. I'll compete them over.

• 2:19:15Yann LeCun?

  Well, there's all kinds of incentive to make AI systems submissive to humans.

• 2:19:19Lex Fridman?

  Right.

• 2:19:19Yann LeCun?

  Right? I mean, this is the way we're going to build them, right? And so so then people say, oh, but look at LLMs. LLMs are not controllable. And they're right. LLMs are not controllable. But objective-driven AI, so systems that derive their answers by optimization of an objective, means they have to optimize this objective, and that objective can include guardrails. One guardrail is obey humans. Another guardrail is don't obey humans if it's hurting other humans with limits.

• 2:19:50Lex Fridman?

  I've heard that before somewhere. I don't remember.

• 2:19:52Yann LeCun?

  Yes, maybe in a book.

• 2:19:54Lex Fridman?

  Yeah. But speaking of that book, what is could there be unintended consequences also from all of this?

• 2:20:01Yann LeCun?

  No, of course. So this is not a simple problem, right? I mean, designing those guardrails so that the system behaves properly is not going to be as a simple. Issue that for which there is a silver bullet, for which you have a mathematical proof that the system can be safe. It's going to be a very progressive iterative design system where we put those guardrails in such a way that the system behave properly. And sometimes they're going to do something that, you know, was unexpected because the guardrail wasn't right and we're going to correct them so that they do it right. The idea somehow that we can't get it slightly wrong, because if we get it slightly wrong, we all die is ridiculous.

• 2:20:42Yann LeCun?

  We're just going to go progressively. And it's just going to be, the analogy I've used many times is is turbojet design. How did we figure out how to make turbojets so unbelievably reliable? Right? I mean, those are like incredibly complex pieces of hardware that run at really high temperatures for, you know, 20 hours, 20 hours at a time sometimes. And we can, you know, fly halfway around the world with a two, on a two-engine

• 2:21:17Yann LeCun?

  jetliner at near the speed of sound. Like, how incredible is this? It's just unbelievable. Right? And did we do this because we invented like a general principle of how to make turbojets safe? No, we, it took decades to kind of fine tune the design of those systems so that they were safe. Is there a separate group within General Electric or SNEKMA or whatever that is specialized in turbojet safety? No, it's, the design is all about safety. Because a better turbojet is also a safer turbojet. So a more reliable one. It's the same for AI. Like, do you, do you need, you know, specific provisions to make AI safe?

• 2:22:01Yann LeCun?

  No, you need to make better AI systems and they will be safe because they are designed to be more useful and more controllable.

• 2:22:09Lex Fridman?

  So let's imagine a system, AI system, that's able to be incredibly convincing and can convince you of anything. I can at least imagine such a system. And I can see such a system be weapon-like because it can control people's minds. We're pretty gullible. We want to believe a thing. You can have an AI system that controls it. And you could see governments using that as a weapon. So do you think if you imagine such a system, there's any parallel to something like nuclear weapons?

• 2:22:46Yann LeCun?

  No.

• 2:22:47Lex Fridman?

  So is why, why, why is that technology different? So you're saying there's going to be gradual development?

• 2:22:54Yann LeCun?

  Yeah.

• 2:22:54Lex Fridman?

  It's going to be, I mean, it might be rapid, but there'll be iterative. And then we'll be able to kind of respond and so on.

• 2:23:01Yann LeCun?

  So that AI system designed by Vladimir Putin or whatever, or his minions,

• 2:23:09Yann LeCun?

  you know, is going to be like talking to, trying to talk to every American to convince them to vote for, you know,

• 2:23:17Lex Fridman?

  Whoever.

• 2:23:18Yann LeCun?

  Whoever pleases Putin.

• 2:23:21Lex Fridman?

  Sure.

• 2:23:22Yann LeCun?

  Or whatever, or, or, you know, or rile people up against each other, as they've been trying to do. They're not going to be talking to you, they're going to be talking to your AI assistant,

• 2:23:37Yann LeCun?

  which is going to be as smart as theirs. Right? That AI, because as I said, in the future, every single one of your interaction with the digital world will be mediated by your AI assistant. So the first thing you're going to ask is, is this a scam?

• 2:23:51Lex Fridman?

  Yeah.

• 2:23:51Yann LeCun?

  Like, is this thing like telling me the truth? Like, it's not even going to be able to get to you because it's only going to talk to your AI system. Your AI system is not even going to, it's going to be like a spam filter, right? You're not even seeing the email, the spam email, right? It's automatically put in a folder that you never see. It's going to be the same thing. That AI system that tries to convince you of something is going to be talking to your AI assistant, which is going to be at least as smart as it.

• 2:24:19Yann LeCun?

  And it's going to say this is spam, you know, it's not even going to bring it to your attention.

• 2:24:24Lex Fridman?

  So to you, it's very difficult for any one AI system to take such a big leap ahead to where it can convince even the other AI systems. So like, there's always going to be this kind of race where nobody's way ahead.

• 2:24:39Yann LeCun?

  That's the history of the world. History of the world is, you know, whenever there is a progress someplace, there is a countermeasure. And, and, you know, it's a, it's a cat and mouse game.

• 2:24:50Lex Fridman?

  This is why mostly yes, but this is why nuclear weapons are so interesting because that was such a powerful weapon that it matters who got it first.

• 2:25:01Lex Fridman?

  That you know, you could imagine Hitler, Stalin,

• 2:25:08Lex Fridman?

  Mao getting the weapon first and that that having a different kind of impact on the world than than the United States getting the weapon first.

• 2:25:16Yann LeCun?

  Yeah.

• 2:25:17Lex Fridman?

  To you, nuclear weapons is like you, you don't imagine a breakthrough discovery and then Manhattan Project like effort for AI.

• 2:25:28Yann LeCun?

  No, as I said, it's not going to be an event. It's going to be, you know, continuous progress and and whenever, you know, one breakthrough occurs, it's going to be widely disseminated really quickly. Probably first within industry. I mean, this is not a domain where, you know, government or military organizations are particularly innovative and they're in fact way behind. And so this is going to come from industry and and this kind of information disseminates extremely quickly. We've seen this over the last few years, right? Where you have a new like, you know, even take AlphaGo. This was reproduced within three months. Even without like particularly detailed information, right?

• 2:26:10Lex Fridman?

  Yeah, this is an industry that's not good at secrecy.

• 2:26:13Yann LeCun?

  No, but even even if there is just the fact that you know that something is possible.

• 2:26:17Lex Fridman?

  Yeah.

• 2:26:19Yann LeCun?

  Makes you like realize that it's worth investing the time to actually do it. You you may be the second person to do it, but you know, you'll you'll do it.

• 2:26:29Yann LeCun?

  And, you know, same for, you know, all the innovations of, you know, self-supervising transformers, decoder-only architectures, LLMs. I mean, those things, you don't need to know exactly the details of how they work to know that, you know, it's possible. Because it's deployed and then it's getting reproduced and then, you know, people who work for those companies move. They go from one company to another. And, you know, the information disseminates. What makes the success of the the US tech industry and Silicon Valley in particular is exactly that. It's because information circulates really, really quickly. And this, you know, disseminates very quickly. And so, you know, the the whole region sort of.

• 2:27:14Yann LeCun?

  Is ahead because of that circulation of information.

• 2:27:17Lex Fridman?

  Maybe I just to linger on the psychology of AI doomers. You give in the classic Yann Le Cun way, a pretty good example of just when a new technology comes to be. You say, engineer says, I invented this new thing. I call it a ball pen. And then the Twitter sphere responds, OMG, people could write horrible things with it like misinformation propaganda. Hate speech, ban it now. Then writing doomers come in akin to the AI doomers. Imagine if everyone can get a ball pen. This could destroy society. There should be a law against using ball pen to write hate speech. Regulate ball pens now.

• 2:28:00Lex Fridman?

  And then the pencil industry mogul says, yeah, ball pens are very dangerous. Unlike pencil writing, which is erasable. Ball pen writing stays forever. Government should require a license for a pen manufacturer. I mean, this does seem to be part of human psychology. When, when it comes up against new technology. So what, what deep insights can you speak to about this?

• 2:28:30Yann LeCun?

  Well, there is a natural fear of new technology and the impact it can have on society. And people have kind of instinctive reaction to, you know, the world they know being threatened by major transformations. That are either cultural phenomena or technological revolutions. And they fear for their culture, they fear for their job, they fear for their, you know, the future of their children.

• 2:29:04Yann LeCun?

  And their way of life, right? So, so any change is feared and.

• 2:29:10Yann LeCun?

  And you see this, you know, along history, like any technological revolution or cultural phenomenon was always accompanied by, you know, groups or reaction in the media that that basically attributed the all the problems, the current problems of society to that particular change. Right. Electricity was going to kill everyone at some point, you know, you the train was going to be a horrible thing because, you know, you can't breathe past fifty kilometers an hour. And so there's a wonderful website called the Pessimist Archive,

• 2:29:49Yann LeCun?

  which has all those newspaper clips of all the horrible things people imagine would would arrive because of either a technological innovation or a cultural phenomenon. You know, this is wonderful examples of,

• 2:30:09Yann LeCun?

  you know, jazz or comic books being blamed for unemployment or or, you know, young people not wanting to work anymore and things like that. Right. And and that has existed for for centuries.

• 2:30:25Yann LeCun?

  And it's, you know, knee jerk reactions.

• 2:30:31Yann LeCun?

  The question is, you know, do we embrace change?

• 2:30:35Yann LeCun?

  Or do we resist it? And what are the real dangers as opposed to the imagined imagined ones?

• 2:30:44Lex Fridman?

  So people worry about, I think one thing they worry about with big tech, something we've been talking about over and over. But I think worth mentioning again, they worry about how powerful AI will be. And they worry about it being in the hands of one centralized power of just a handful of central control. And so that's the skepticism with big tech you can make. These companies can make a huge amount of money and control this technology. And by so doing, you know, take advantage, abuse the little guy in society.

• 2:31:21Yann LeCun?

  Well, that's exactly why we need open source platforms.

• 2:31:24Lex Fridman?

  Yeah, I just want to

• 2:31:28Lex Fridman?

  nail the point home more and more.

• 2:31:29Yann LeCun?

  Yes.

• 2:31:31Lex Fridman?

  So let me ask you on your. Like I said, you do get a little bit flavorful on the internet. Yoshua Bach tweeted something that you LOL'd at in reference to Hal 9000. Quote, "I appreciate your argument and I fully understand your frustration, but whether the pod bay doors should be opened or closed is a complex and nuanced issue." So you're at the head of Meta AI. You know, this is something that really worries me that AI or AI overlords will speak down to us with corporate speak of this nature.

• 2:32:13Lex Fridman?

  And you sort of resist that with your way of being. Is this something you can just comment on sort of working at a big company? How you can avoid the over fearing, I suppose, the through caution create harm?

• 2:32:34Yann LeCun?

  Yeah. Again, I think the answer to this is open source platforms and then enabling a widely diverse set of people to build AI assistants that represent the diversity of cultures, opinions, languages, and value systems across the world so that you're not bound to just, you know, be brainwashed by a particular way of thinking because of a single AI entity. So I mean, I think it's a really, really important question for society. And the problem I'm seeing is, is that, which is why I've been so vocal and sometimes a little sardonic about it.

• 2:33:17Lex Fridman?

  Never stop, never stop, Jan. We love it.

• 2:33:22Yann LeCun?

  Is because I see the danger of this concentration of power through proprietary AI systems as a much bigger danger than everything else. That if we really want, you know, diversity of opinion, AI systems that, you know, in this future that where we'll all be interacting through AI systems. We need those to be diverse for the preservation of diversity of ideas and, you know, creeds and political opinions and whatever. And the preservation of democracy and.

• 2:34:02Yann LeCun?

  What works against this is people who think that for reasons of security, we should keep AI systems under lock and key because it's too dangerous to put it in the hands of everybody, because it could be used by terrorists or something. That would lead to, you know, potentially a very bad future in which all of our information diet is controlled by a small number of companies who proprietary systems.

• 2:34:37Lex Fridman?

  Do you trust humans with this technology to build systems that are, on the whole, good for humanity?

• 2:34:46Yann LeCun?

  Isn't that what democracy and free speech is all about?

• 2:34:49Lex Fridman?

  I think so.

• 2:34:50Yann LeCun?

  Do you trust institutions to do the right thing? Do you trust people to do the right thing? And yeah, there's bad people who are going to do bad things, but they're not going to have superior technology to the good people. So then it's going to be my good AI against your bad AI, right? I mean, it's the examples that we were just talking about of, you know, maybe some rogue country will build, you know, some AI system that's going to try to convince everybody to go into a civil war or something, or elect a favorable ruler. And but then they will have to go past our AI systems.

• 2:35:29Lex Fridman?

  An AI system with a strong Russian accent will be trying to convince our...

• 2:35:33Yann LeCun?

  And doesn't put any articles in their sentences.

• 2:35:38Lex Fridman?

  Well, it'll be at the very least absurdly comedic. Okay, so I, since we talked about the physical reality, I'd love to ask your vision of the future with robots in this physical reality. So many of the kinds of intelligence you've been speaking about would empower robots to be more effective collaborators with us humans. So since Tesla's Optimus team has been showing us some progress on humanoid robots, I think it really reinvigorated the whole industry that I think Boston Dynamics has been leading for a very, very long time. So now there's all kinds of companies, Figure AI, obviously Boston Dynamics.

• 2:36:22Lex Fridman?

  Union tree, but there's like a lot of them. It's great. It's great. I mean, I love it.

• 2:36:30Lex Fridman?

  So do you think there'll be millions of humanoid robots walking around soon?

• 2:36:36Yann LeCun?

  Not soon, but it's gonna, it's gonna happen. Like the next decade, I think is gonna be really interesting in robots. Like the, the emergence of the robotics industry has been in the waiting for, you know, ten, twenty years without really emerging other than for like, you know, kind of pre-programmed behavior and stuff like that. And, and the main issue is, again, the Moravec paradox, like, you know, how do we get the systems to understand how the world works? And, and kind of, you know, plan actions. And so we can do it for really specialized tasks.

• 2:37:10Yann LeCun?

  And the way Boston Dynamics goes about it is, you know, basically with a lot of handcrafted dynamical models and careful planning in advance, which is very classical robotics with a lot of innovation, a little bit of perception, but it's still not, like, they can't build a domestic robot, right? And, you know, we're still some distance away from completely autonomous level five driving.

• 2:37:40Yann LeCun?

  And we're certainly very far away from having, you know, a level five autonomous driving by a system that can train itself by driving twenty hours like any seventeen year old.

• 2:37:54Yann LeCun?

  So until we have, again, world models, systems that can train themselves to understand how the world works,

• 2:38:05Yann LeCun?

  we're not gonna, we're not gonna have significant progress in robotics. So a lot of the people working on robotic hardware at the moment are, are betting or banking on the fact that AI is gonna make sufficient progress towards that.

• 2:38:20Lex Fridman?

  And they're hoping to discover a product in it too, is uh,

• 2:38:24Yann LeCun?

  Yeah.

• 2:38:25Lex Fridman?

  Before you have a really strong world model, there'll be an almost strong world model and uh,

• 2:38:32Lex Fridman?

  People are trying to find a product in a clumsy robot, I suppose. Like, not a perfectly efficient robot. So there's the factory setting where humanoid robots can help automate some aspects of the factory. I think that's a crazy, difficult task because of all the safety required and all this kind of stuff. I think in the home is more interesting, but then you start to think. I think you mentioned loading the dishwasher, right?

• 2:38:55Yann LeCun?

  Yeah.

• 2:38:56Lex Fridman?

  Like, I suppose that's one of the main problems you're working on.

• 2:39:00Yann LeCun?

  I mean, there's, you know, cleaning up.

• 2:39:03Lex Fridman?

  Yeah.

• 2:39:04Yann LeCun?

  Cleaning the house, clearing up the table after a meal. Washing the dishes, you know, all those tasks, you know, cooking. I mean, all the tasks that, you know, in principle could be automated, but are actually incredibly sophisticated, really complicated.

• 2:39:21Lex Fridman?

  But even just basic navigation around a space full of uncertainty.

• 2:39:24Yann LeCun?

  That sort of works, like you can sort of do this now. Navigation is fine.

• 2:39:30Lex Fridman?

  Well, navigation in a way that's compelling to us humans is a different thing.

• 2:39:35Yann LeCun?

  Yeah, it's not going to be, you know, necessarily. I mean, we have demos actually because, you know, there is a so-called embodied AI group at Fair. And, you know, they've been not building their own robots, but using commercial robots. And you can, you can tell a robot dog, like, you know, go to the fridge. And it can actually open the fridge and it can probably pick up a can in the fridge and stuff like that and, and bring it to you. I know, so it can navigate, it can grab objects as long as it's been trained to recognize them, which, you know, vision systems work pretty well nowadays.

• 2:40:11Yann LeCun?

  But, but it's not like a completely, you know, general robot that would be, you know, sophisticated enough to do things like clearing up the dinner table.

• 2:40:23Lex Fridman?

  Yeah, to me, that's an exciting future of getting humanoid robots, robots in general in the whole, more and more, because that gets humans to really directly interact with AI systems in the physical space. And in so doing, it allows us to philosophically, psychologically explore our relationships with robots. It can be really, really, really interesting. So I hope you make progress on the whole JAPPA thing soon.

• 2:40:47Yann LeCun?

  Well, I mean, I hope, I hope things can, you know, work as, as planned.

• 2:40:53Yann LeCun?

  I mean, again, we've been kind of working on this idea of self-supervised learning of, from video for, for 10 years and, and, you know, only made significant progress in the last two or three.

• 2:41:04Lex Fridman?

  And actually, you've, you've mentioned that there's a lot of interesting breakthroughs that can happen without having access to a lot of compute.

• 2:41:10Yann LeCun?

  Yeah.

• 2:41:11Lex Fridman?

  So if you're interested in doing a PhD in this kind of stuff, there's a lot of possibilities still.

• 2:41:16Yann LeCun?

  Yeah.

• 2:41:16Lex Fridman?

  To do innovative work. So like, what advice would you give to a undergrad that's looking to go to grad school and do a PhD?

• 2:41:25Yann LeCun?

  So basically, I've listed them already, this idea of how do you train a world model by observation? And you don't have to train necessarily on gigantic data sets or, I mean, it could turn out to be necessary to actually train on large data sets, to have emergent properties like, like we have with other LMS. But I think there is a lot of good ideas that can be done without necessarily scaling up. Then there is, how do you do planning with a learned world model? If the world the system evolves in is not the physical world, but is the world of, let's say, the internet, or, you know, some sort of world of where an action consists in doing a search in a search engine, or interrogating a database, or running a

• 2:42:09Yann LeCun?

  simulation, or calling a calculator, or solving differential equation?

• 2:42:15Yann LeCun?

  How do you get a system to actually plan a sequence of actions to, you know, give the solution to a problem?

• 2:42:22Yann LeCun?

  And so the question of planning is not just a question of planning physical actions. It could be, you know, planning actions to use tools for a dialogue system or for any kind of intelligent system. And there's some work on this, but not like, not a huge amount. Some work at FAIR, one called Toolformer, which was a couple of years ago. And some more recent work on planning, but, but I don't think we have like a good solution for any of that. Then there is the question of hierarchical planning. So the example I, I mentioned of, you know, planning a trip from New York to Paris.

• 2:43:03Yann LeCun?

  That's hierarchical, but almost every action that we take involves hierarchical planning in some, in some sense. And we really have absolutely no idea how to do this. Like there's zero demonstration of hierarchical planning in AI, where the various levels of representations that are necessary have been learned. We can do like two level hierarchy, hierarchical planning when we design the two, the two levels. So for example, you have like a dog-like robot, right? You want it to go from the living room to the kitchen. You can plan a path that avoids the obstacle.

• 2:43:44Yann LeCun?

  And then you can send this to a lower, lower level planner that figures out how to move the legs to kind of follow that trajectories, right? So that works, but that two level planning is designed by hand, right?

• 2:43:58Yann LeCun?

  We specify what the proper levels of abstraction, the representation at each level of abstraction has, have to be. How do you learn this? How do you learn that hierarchical representation of action plans?

• 2:44:12Yann LeCun?

  Right, we, you know, with ConvNets and deep learning, we can train the system to learn hierarchical representations of percepts.

• 2:44:20Yann LeCun?

  What is the equivalent when what you're trying to represent are action plans?

• 2:44:23Lex Fridman?

  For action plans, yeah. So you want, you want basically a robot dog or humanoid robot that turns on and travels from New York to Paris all by itself?

• 2:44:33Yann LeCun?

  For example?

• 2:44:34Lex Fridman?

  All right. It might have some trouble at the TSA, but.

• 2:44:39Yann LeCun?

  Yeah, no, but even doing something fairly simple, like a household task.

• 2:44:43Lex Fridman?

  Sure.

• 2:44:43Yann LeCun?

  Like, you know, cooking or something.

• 2:44:46Lex Fridman?

  Yeah, that there's a lot involved. It's a super complex task we take. And once again, we take it for granted. What hope do you have for the future of humanity? We're talking about so many exciting technologies, so many exciting possibilities. What gives you hope when you look out over the next 10, 20, 50, 100 years? If you look at social media, there's a lot of, there's, there's wars going on. There's division, there's hatred, all this kind of stuff. That's also part of humanity. But amidst all that, what gives you hope?

• 2:45:21Yann LeCun?

  I love that question.

• 2:45:27Yann LeCun?

  We can make humanity smarter with AI.

• 2:45:32Yann LeCun?

  Okay. I mean, AI basically will amplify human intelligence.

• 2:45:38Yann LeCun?

  It's as if every one of us will have a staff of smart AI assistants. They might be smarter than us. They'll do our bidding.

• 2:45:51Yann LeCun?

  Perhaps execute a task in ways that are much better than we could do ourselves because they'll be smarter than us. And so it's like everyone would be the, the boss of a staff of super smart virtual people. So we shouldn't feel threatened by, by this any more than we should feel threatened by being the manager of a group of people, some of whom are more intelligent than us.

• 2:46:20Yann LeCun?

  I certainly have a lot of experience with this, of, you know, having people working with me who are smarter than me. That's actually a wonderful thing. So having machines that are smarter than us, that assist us in all of our tasks, our daily lives, whether it's professional or personal, I think would be absolutely a wonderful thing. Because intelligence is the most, is the commodity that is most in demand. That's really what, I mean, all the mistakes that humanity makes is because of lack of intelligence, really, or lack of knowledge, which is, you know, related. So making people smarter would just, can only be better. I mean, for the same reason that, you know, public education is a good thing.

• 2:47:05Yann LeCun?

  And books are a good thing. And the internet is also a good thing, intrinsically. And even social networks are a good thing, if you run them properly. It's difficult, but, you know, you can. Because, you know, it, helps the communication of information and knowledge, and the transmission of knowledge. So AI is going to make humanity smarter. And the analogy I've been using is the fact that perhaps an equivalent event in the history of humanity to what might be provided by generalization of AI assistant is the invention of the printing press. It made everybody smarter.

• 2:47:50Yann LeCun?

  The fact that people could have access to books. Books were a lot cheaper than they were before. And so a lot more people had an incentive to learn to read, which wasn't the case before. And people became smarter. It enabled the Enlightenment, right? There wouldn't be an Enlightenment without the printing press. It enabled philosophy, rationalism, escape from religious doctrine.

• 2:48:27Yann LeCun?

  Democracy, science.

• 2:48:34Yann LeCun?

  And certainly, without this, it wouldn't be, there wouldn't have been the American Revolution, the French Revolution. And so we'll still be under a feudal regimes, perhaps.

• 2:48:46Yann LeCun?

  And so it completely transformed the, the world because people became smarter and kind of learn, learn about things. Now, it also created 200 years of essentially religious conflicts in Europe, right? Because the first thing that people read was the Bible and realized that perhaps there was a different interpretation of the Bible than what the priests were telling them. And so that created the Protestant movement and created the rift. And in fact, the Catholic school, the Catholic Church didn't like the idea of the printing press, but they had no choice. And so it had some bad effects and some, some good effects. I don't think anyone today would say that the invention of the printing press had a

• 2:49:29Yann LeCun?

  overall negative effect, despite the fact that it created 200 years of religious conflicts in Europe. Now, compare this, and I, I thought, I was very proud of myself to come up with this analogy, but realized someone else came with the same idea before me. Compare this with what happened in the Ottoman Empire. The Ottoman Empire banned the printing press for 200 years.

• 2:50:00Yann LeCun?

  And it didn't ban it for all languages, only for Arabic. You could actually print books in Latin or Hebrew or whatever, in the Ottoman Empire, just not in Arabic. And

• 2:50:16Yann LeCun?

  I thought it was because the rulers just wanted to preserve the control over the population and the dogma, religious dogma and everything. But after talking with the UAE Minister of AI,

• 2:50:31Yann LeCun?

  Omar Al-Omar.

• 2:50:34Yann LeCun?

  He told me, no, there was another reason.

• 2:50:38Yann LeCun?

  And the other reason was that it was to preserve the corporation of calligraphers.

• 2:50:46Yann LeCun?

  Right, there's like an art form, which is, you know, writing those beautiful.

• 2:50:51Lex Fridman?

  Yes.

• 2:50:52Yann LeCun?

  You know, Arabic poems or whatever religious texts in this thing. And it was very powerful corporation of scribes, basically that kind of, you know, run a big chunk of the empire. And you couldn't put them out of business. So they, you know, banned the British press in part to protect that business.

• 2:51:14Yann LeCun?

  Now, what's the analogy for AI today? Like, who are we protecting by banning AI? Like, who are the people who are asking that AI be regulated to protect their, their jobs? And of course, you know, there's, it's, it's, it's a real question of what is going to be the effect of,

• 2:51:32Yann LeCun?

  you know, technological transformation like AI on the, on the job market and the labor market. And there are economists who are much more expert at this than I am. But when I talk to them, they, they tell us, you know, we're not going to run out of job. This is not, this is not going to cause mass unemployment. This, this is just going to be gradual shift of different professions. The professions that are going to be hot 10 or 15 years from now. We have no idea today what they're going to be. The same way, if we go back 20 years in the past, like, who could have thought 20 years ago? That like the hottest job, even like five, 10 years ago, was mobile app developer. Like smartphones weren't invented.

• 2:52:16Lex Fridman?

  Most of the jobs of the future might be in, in the metaverse.

• 2:52:19Yann LeCun?

  Well, it could be, yeah.

• 2:52:21Lex Fridman?

  But the point is you can't possibly predict. But you're right, I mean, you made a lot of strong points and I, I believe that people are fundamentally good. And so if AI, especially open source AI, can make them smarter, it just empowers the goodness in humans.

• 2:52:41Yann LeCun?

  So I, I share that feeling, okay? I think people are fundamentally good.

• 2:52:47Yann LeCun?

  And in fact, a lot of doomers are doomers because they don't think that people are fundamentally good. And they either don't trust people or they don't trust the institution to do the right thing so that people behave properly.

• 2:53:03Lex Fridman?

  Well, I think both you and I believe in humanity. And I think I speak for a lot of people in saying, thank you for pushing the open source movement, pushing to making both research in AI open source, making it available to people, and also the models themselves, making it open source. So, thank you for that. And thank you for speaking your mind in such colorful and beautiful ways on the internet. I hope you never stop. You're one of the most fun people I know and get to be a fan of. So, yeah, thank you for speaking to me once again, and thank you for being you. Thank you, Lex. Thanks for listening to this conversation with Jan LeCun. To support this podcast, please check out our sponsors in the description. And now, let me leave you with some words from Arthur C. Clarke.

• 2:53:49Lex Fridman?

  "The only way to discover the limits of the possible is to go beyond them into the impossible." Thank you for listening and hope to see you next time.