In short
Podcast Summary: The Twenty Minute VC (20VC) - Episode with Yann LeCun
Episode Overview Title: 20VC: Yann LeCun on Why Artificial Intelligence Will Not Dominate Humanity, Why No Economists Believe All Jobs Will Be Replaced by AI, Why the Size of Models Matters Less and Less & Why Open Models Beat Closed Models Guest: Yann LeCun, VP & Chief AI Scientist at Meta Host: Harry Stebbings Release Date: Not specified in the transcript
Guest Background
- Yann LeCun is a prominent figure in the AI field and the recipient of the 2018 ACM Turing Award for his work in deep neural networks.
- Previously held positions at AT&T Bell Labs and NYU, where he was a founding director of the NYU Center for Data Science.
Key Topics Discussed
- The Road to AI OG
- Initial Interest in AI: LeCun's fascination with machine learning began during his undergraduate studies in France, fueled by debates on language acquisition.
- Persistence Through Disinterest: He reflects on a decade of diminished interest in machine learning and staying motivated during that period.
- The Next Five Years of AI: Hope or Horror
- AI as a Non-Threat: LeCun argues against the fear that AI poses a danger to humanity, stating that AI will not seek to dominate humans.
- Future of Digital Assistants: He predicts that digital assistants will streamline interactions and potentially replace traditional search methods like Google.
- Employment in an AI World
- Job Replacement Concerns: Contrary to common fears, LeCun mentions that no economists believe AI will completely replace jobs.
- Job Creation: He forecasts the creation of new jobs in the AI economy and expects the transition to be slower than anticipated.
- Critique of Calls to Pause AI Development: LeCun challenges the notion of pausing AI advancements, citing the slow pace of societal adaptation to new technologies.
- Open vs. Closed Models
- Preference for Open Models: LeCun asserts that open models will outperform closed systems due to their collaborative nature, which fosters innovation and knowledge sharing.
- Historical Precedents: He provides examples from history where open-source approaches triumphed over closed models.
- Startups vs. Incumbents
- Future Competition: Discussion on who will dominate the AI landscape—startups or established companies.
- Regulatory Challenges: LeCun explains how regulations can hinder incumbents, impacting their agility in the rapidly evolving AI market.
- Impartiality in His Role: He addresses how his position at Meta has not compromised his objectivity.
Key Takeaways
- AI as an Enhancer: LeCun envisions AI as a tool for boosting human intelligence and creativity, rather than a threat.
- Open Infrastructure: Emphasizes the benefits of open-source models and collaborative developments as essential for progress.
- Adaptive Workforce: While AI will transform job landscapes, it will also create new opportunities, supporting a shift in job types rather than a complete elimination.
Conclusion Yann LeCun's insights provide a forward-looking perspective on AI's role in society, emphasizing collaboration and adaptation rather than fear and stagnation. He encourages a focus on the positive potential of AI technologies while acknowledging necessary regulatory frameworks. This episode serves as a significant resource for understanding the future trajectory of AI and its implications for humanity and the workforce.
Additional Resources For more information about the podcast and to access full episodes, visit [The Twenty Minute VC](http://www.20vc.com).
Written by AI. May contain mistakes. Listen to the episode to check what was said.
Transcript
Automatic transcript. May contain errors.0:00AI is going to bring a new renaissance for humanity. A new form of enlightenment, if you want, because AI is going to amplify everybody's intelligence. Every one of us will have a staff of people who are smarter than us and know most things about most things. So it's going to empower every one of us. Welcome back. This is 20VC with me, Harry Stubbins, and stay. I'm joined by one of the greats, Jan LeCoon. Jan is VP and Chief AI Scientist at Metta, and Silver Professor at NYU. He was the founding director of Fair and of the NYU Centre for Data Science. He's the recipient of the 2018 ACM Turing Award for Conceptual and Engineering breakthroughs that have made deep neural networks a critical component of computing, and I want to say huge thank you to David Marcus and Mathieu at Photo Room for helping make this one happen.
0:45This really was a huge honour for me to do. But before we dive into the show's day, are you building enough conviction to outpace other investors in this changing ecosystem? system, Tegas helps VCs get under the hood of their investment's quicker. Map of all the markets with deeper research, monitor the financial health of portfolio companies and even source new deals. Tegas is trusted by top firms like Red Point and Spark Capital for everything from expert calls and transcripts to quickly surfacing insights across filings, comps and banish marking. Find out why a majority of the top venture funds are using Tegas on a daily basis to differentiate, consolidate and accelerate their research, had two teagus .com forward slash growth for your free trial, and speaking of tools we cannot live without there, Secure Frame is the leading all -in -one platform for automated security and privacy compliance.
1:34Secure Frame simplifies and streamlines the process of getting and staying compliant to the most rigorous global privacy and security standards, like SOC2 and others, and Secure Frame's industry leading compliance automation platform, paired with their in -house compliance experts and former auditors helps you get all it ready in weeks or not months so you can close more deals faster. Automate your security and privacy compliance with secure frame. Schedule a demo today at secureframe .com that really is a must. And finally are you tired of using the same old tools and platforms in an attempt to manage your global team?
2:09Well us too, why use 16 different tools? When you could just use deal, deal is the only global first HR tool with unmatched automation, customizations and integration making it the only platform for building contracts, visas, background checks, payroll equipment, benefits and beyond. Are you ready to see how Deal can scale with your global and local teams? Visit Deal .com forthslash20VC and book your demo today. That's Deal .com forthslash20VC to book your demo.
2:44You are now arrived at your destination. Jan, I am so excited for this. I had so many great things from our mutual friends. Obviously, David Marcus and then Mathieu at PhotoRoom. So thank you so much for joining me today. And it's a pleasure. Now, I heard some of the early stories, but I want to start with one from David Marcus. How did you first end to the world of AI and make that first for a? We were still on the undergraduate engineering student in France and I stumbled on the philosophy book which was a debate between Jean -Pierre Gé and the cognitive psychologist and Noël Chomsky, the famous linguist.
3:17And they were arguing about nature versus nurture for language, whether language is acquired or innate. So Chomsky was on the side of innate and Pierre Gé on the side of acquired with some innate structure. And on the side of Piaget was a guy called Simor Papper, who was a professor in MIT. In his argument, he talked about something called the perceptron, which was an early machine learning system. I read this and discovered that people had been working on machines that could learn, and I was fascinated, and I started digging the literature. Soon discovered that much of that literature was in the 1950s and 60s, and basically stopped in the late 60s, because of a book that they killed it.
3:54And Simor Papper was a co -author of that book. And here we were 10 years later actually praising the perceptron as an amazing concept. So I was hooked. I started getting interested in what was not yet called machine learning, but eventually became neural nets and now deep learning. Can we ask you David asked this as well? How long it takes you to get to the major breakthroughs that you're the origin of when you look back over that time to get to those major breakthroughs? So the first one was when I was still not so glad, basically finishing my engineering studies, figured out that the way forward to lift the limitations of the old systems that were abandoned in the 60s was to find learning algorithms that could train multi -layer neural nets.
4:31And people had all but abandoned this type of research, except for a handful of people in Japan. And one guy I heard about called Jeff Hinton, who published a paper in 1983, so this was just the year I graduated, on something called a bosom machine, which was clearly a method to go beyond those limitations. And so I had developed a method for training multilayer nets, which was very close to what we now call back propagation, but not exactly the same. It was closer to what we call target prop actually nowadays. And then published a few papers in French and eventually you met Jeff at a meeting in France in 1985.
5:08And we realized we'd been working on the same thing and we're seeking a like. But I was in the middle of my PhD and he was an associate professor at County Gamelon. So we started a discussion and then you know, visited him at Carnegie Mellon for a summer school he organized and then when I finished my G .O .G. I did a post -doc with him and they joined Bell Labs and when I was in Toronto I developed what was called computational nets, now computational networks which is a major method for image and speech processing. And so that's what I'm based known for, but it started much earlier. I have to ask you, you're sure described the hype cycles within AI a neural net like deserts when you're not in them.
5:44And he asked the question, how did Jan not get discouraged? When for a solid decade we were in a desert where no one really cared about neural nets. How did you keep the enthusiasm bluntly when Azoshu said no one really cared? Both Azoshu and Jai Fanai had in the back of our minds that those methods would eventually come to the fore and that we would have to snap people out of their preconceived ideas about neural nets. So you're trying to actually work together AT &T Bell Labs in the early 90s and then the interest of the community for those methods started waning around 1995 or so. And it was indeed about 10 years when not only nobody was interested in your own S but people were even making fun of it and talking about it in disparaging terms.
6:27Now there is something though, in 1996 I can't change the job. I stayed in the same company, I was still working at AT &T in the research labs, but I became a department head and it was the early days of the internet and my group and I started working on something completely different. That had not much to do at least with machine learning, with image compression. I had this idea that with the internet, coming up we should have a way of scanning existing paper documents and then put them on the internet so that everybody could have access to them. So I worked on this for five or six years and that project ended when all of us basically left AT &T.
7:00That's when I restarted working on on deep learning and Jeff also came back to Canada He had been in the UK for a while Yoshwai Jeff and I decided in the early 2000 to basically start a conspiracy They will revive the interests of the community in your nets by making them work discovering new algorithms and It took almost 10 years, but it succeeded be all the while this dream was basically So I'm gonna ask you a range of varying questions in terms of depth, breath and obvious and non -obvious So forgive me if some are obvious. I just want to ask when I hear the historical context there from you over Many decades.
7:35How do you feel today when we look at what's happening today? Are we at a new inflection point in development? Or is this merely the continuation of what we've seen for many decades? It's a combination of the two. On the one hand, a lot of what we see today when you are down in the trenches of research looks a logical extension. I was not as enthralled by the sort of recent progress as the public was because I've seen this progress happening all the last several years. Now there are things that have been very surprising. The fact that self -supervised learning methods applied to transformer architectures work amazingly well and it worked way beyond what we could have expected.
8:15The fact that we can do basically train systems to understand language, translate language in multiple languages and then continue text if you try them to do this or answer questions if you try them to do this. Works amazing you will to an extent that people didn't quite expect what's going to happen by just making them bigger and training them on more data. So that's certainly been surprising for everybody but that revolution occurred two years ago, right? So whereas the wider public has a lot about it through tragedy that was made available for us, you know, it's been more continuous and you see this in a lot of marking events in technological progress or in AI in particular are marked by kind of splashy events that the public pays attention to.
8:53But to many of us, it looks like more like a continuous thing. And generally, what those progress require is a bunch of people to take the techniques that already exist, push them a little further, do a bit of engineering, and then make a demo that demonstrated that it works. So that was the case for Deep Blue, the chess player, that IBM built in the mid 90s, that Beat Gary Casparov, you know, was something with the DAPA Grand Challenge for that Sebastian Tuen team at Stanford won, you know, a car that could drive itself in the desert for a hundred miles. And then, I'll go and the IBM Drupody and there's a number of those things, right?
9:27And Dbt just being the latest one. It looks like kind of jumps when you look at it from far away, but when you're in the field, it's more like a continuous evolution. Can I ask, have there been any other surprising on the positive side? Developmental things you've seen in the last year or so you said about self -supervised learning and the efficiencies there? Is there anything else where you're like I didn't expect it to go as well as it has done in the last year? Yeah, so I already mentioned it the fact that merely training a Language model to predict that the last word in the sequence of words if you do it properly You get a system that has capabilities that are somewhat unexpected and they emerge as you make those systems bigger and you train them on not your amounts of data That's really been the surprise for everyone now the thing is as Researchers and scientists were always looking for the next thing So what I'm interested in at the moment is what goes beyond that a lot of people are going to work on applications of To receive large language models, which is great.
10:21There's going to be a lot of products and new ways for people to do things and it's going to be wonderful But I've already been thinking about the next stage for the last four or five years even actually more Which is like what's missing from those systems? What are your thoughts on what's missing from my systems in that logical next step? Where does that lead you in your thinking? So those systems do not have anywhere close to human level intelligence, okay? Despite what you might think, we are kind of fooled into thinking it because those systems are really fluent with language. But their ability to think, to understand how the world works, to plan, are very limited.
10:53And they understand you, the world is very superficial. And the reason for it is that they are strictly trained on language. And language only contains a small proportion of whole human knowledge. Most of human knowledge is not linguistic at all. And all of animal knowledge is non -negustic. We take it for granted. You know, this is the more of that paradox, right? All the capabilities and abilities that we take for granted, planning a motion or something, or very simple things that everyone can do. A 10 -year -old can clear the dinner table and fill up the dishwasher. Any 17 -year -old can learn to drive.
11:26We still don't have some very cars. We don't have domestic robots. If they're non -linguistic, like the majority, I'm sorry for the base questions, but then what are they? And is that that we don't have able to be ingested by AI models and engines over time. So first of all, there is no question that eventually AI systems will understand the world in similar ways that humans do, but has better ways. But they will not be autoregressive large language models as a type that we're now talking about. There will be different, for a number of different reasons. But to answer your question more directly, anything that has to do with sort of an intuition of the real world requires an experience of the real world or a simulated version of it, which those large language models don't have.
12:06They purely train from text. There's a number of questions about the physical world that they'll be able to answer because there's a template for it in the, or something very similar in the data that they've been trained on. But same for planning, you can ask them to plan a trip or something and they will adapt a template that they've been trained on. But they don't really have a mental model of how the world works and allows them to plan complex action sequences or use tools or things like that. Is that why you said that AI research is face palm when they hear prophecies of doom. No, that's a different question.
12:36Those are kind of orthogonal concepts. So, I mean, there is some some weak connection. There is a flaw in current auto -reversive elements, which is that you can only control their answer in two ways. The first way is you modify the statistics of the training data that you train them on, possibly using a human feedback for specific answers. And the second one is you change the point. And the combination of the point that, you know, the question you you ask them, the form in which you ask the question and those statistics of the training data entirely determines the answer to the system we produce.
13:08So there is no persistent memory, first of all, but second of all, you cannot control the system. You cannot impose constraints on it like be factual, be understandable by a certain year old. You can try to put this in a prompt, but then, you know, you rely on whether the statistics of the training data is appropriate it for taking that into account. But there's no direct way to constrain the answer of those systems to satisfy certain objectives. That makes them very difficult to control and steer. And so that creates some fears because people are kind of extrapolating if we let those systems do whatever we connect them to a internet and they can do whatever they want.
13:40They're going to do crazy things and stupid things and they have dangerous things and we're not going to be able to control them and they're going to escape or control and they're going to become intelligent just because they're bigger. And that's nonsense. First of all, because this is not the type of system that we are going to give agency to. The system that will eventually be given agency that are going to be able to plan secrets of actions. Our systems are going to have objectives that are going to have to satisfy. And because of those objectives, they're going to be controllable. So they're going to be much more controllable than the current systems.
14:09Okay, so my prediction is that within a few years, nobody in their right mind would use autoregressive internet apps. They'll go away in favor of something more sophisticated and controllable. They can plan its answer as opposed to just produce one word up to the other reactively. That's the first fallacy. The second fallacy is that there is this idea of somehow that the desire to and the ability to dominate is linked with intelligence. So this is a state that a lot of people are making including, you know, my friend Jeff interned recently that somehow as soon as a machine becomes intelligent it becomes uncontrollable because it's being smarter than us It can influence us in ways that we can't even imagine now I think this is a gigantic fantasy because even within the human species It is not the smartest among us that want to dominate the others, okay to dominate other entities You don't necessarily need to be smarter than them But you need to walk to dominate them This is not something that every intelligent entity is going to do spontaneously.
15:08We do it as humans because the desire to influence others was built into us by evolution because we are a social species. Okay, same as baboons and chimpanzees and wolves and dogs and etc. It's not the case for wrong autons. Long autons don't have the desire to dominate anybody because they are non -social animals. They are solitary animals. They are territorial in fact. We need to separate those two concepts, the will, the desire and the ability to dominate on one hand and intelligence on the other hand. The fact that we're going to have separate television machines at our disposal means that every one of us is going to be like a business leader, politician or an academic with a staff of people working for them that are more intelligent than themselves.
15:48It's great. If you feel threatened by being the boss of other people who work with you but are smarter than you, you're not being a good leader. And can I ask you how do we still values within models where they don't have a desired stonminase? OK, so let me describe the sort of architecture or future AI systems as I see it. We're going to have AI systems that basically are going to plan their actions and actions can include sequences of words that you tell someone. But they're going to plan the sequence of actions or words so as to optimize a series of objectives that we set them. OK, so what objective is does this answer the question I just asked?
16:23Okay, another objective might be, were you attacking to a 13 year old, make that answer understandable by a 17 year old? Another objective might be, I asked you to answer a question about the world, so be factual. Or it's a question about yesterday's political event, can you kind of be compatible with everything you've read in the press this morning? Things like that, right? You'll have those systems that have a series of objectives and their output, their answer, by construction is going to have to satisfy those objectives. And some of those objectives would be hardwired to make those systems safe.
16:52Like if it's a domestic robot that can cook dinner and can wield kitchen knife in its arm, there's going to be a term in there that says stop moving your arm when there is people around because you might hurt them. So that's going to be an objective that the system cannot violate because by construction is going to have to satisfy them. So that's the way to build safe AI system. You make them produce answers that by construction have to satisfy objectives and you designed those objectives so that their actions are safe. Now, how precisely to do this is not a completely self -question, but you try it, you deploy it at a small scale, you see what the effect is, and you correct it when it doesn't work.
17:25And you fix it progressively, not like if you get it wrong, it's going to destroy humanity. How do you determine who's able to set the objectives? Because that could be right or wrong, depending on who sets them. That's true. So there's going to have to be a process by which we are people to do this, so some vetting process. The same way that there's a vetting process for people to take care of your health or cut your hair or fix your plumbing or your car, right? So there's some vetting process, certainly some testing and market deployment procedure with regulating agencies for things that are potentially dangerous, probably not for all applications, but for many applications, certainly in healthcare, transportation and things like that.
18:03And then let's take the example of intelligent assistants. Let's imagine a future where everyone can talk to their intelligent assistant. That system will have pretty close to human -level intelligence, probably more accumulated knowledge that most students, they could translate in any language and give you a quick summary of yesterday's newspaper and things like that, explain mathematical concepts to you, things like that. So people are probably going to use this almost exclusively in the future for their interaction with the digital world. We're not going to go to Google or Wikipedia, you're just going to talk to your assistant.
18:35And the only way to do this properly is for the basic infrastructure for those assistants, there would be so preventive, so much will ride on those systems that I don't think anyone will accept that those assistants be behind the eventual horizon in a private company. They wouldn't see that the infrastructure is open. They wouldn't see it also that the vetting process by which those systems are trained be something maybe like Wikipedia. We tend to trust Wikipedia, sometimes with a grain of salt, but we tend to trust Wikipedia, because there is a vetting process so that whenever an article is modified, some editor can check on it and then the changes are accepted or not, things like that.
19:10So you can imagine that the sort of common repository of all human knowledge that will be our assistants will be constructed through some sort of cross -sourcing process, perhaps similar to Wikipedia, where you're going to have a bunch of people training those systems and fine -tuning them so that whatever they, and so they produce are correct. It's so funny you say about that kind of the benefits there of the open approach over the closed approach because that's where I've been stuck Which is like where does value accrue is it to the closed model or the open model and then we had the leaked internal Mammoth day from the Google employee who said we're not ahead open AI or not ahead There's this third being which is actually far more significant and we haven't taken notice of and summarized There was triggered by Lama which is the model that was put together by my colleagues at fair which was the code was open source The model, sadly, was distributed only for research and non -commercial purpose.
20:01And the reason for that is basically complicated legal issues of what's the status of the data that the system had been trained on and things like that. It's not a lack of desire from meta to open source. It's more a kind of complex legal issues that go beyond my. Why does open win against a more controlled tight net, well -funded open AI for other large corporate with a big balance sheet and a very rigorous but streamlined team. It's very simple, it's because no outfit as powerful as the MAB has a monopoly on good ideas. If you do it in the open you recruit the entire world's intelligence to contribute to things and having ideas and ideas that you might as have thought about which an outfit was for 100 people as no chance of thinking about or even a large company with 50 ,000 employees may not want to devote any resources to because they may not think it's useful in the long -term or they have more urgency to take care of.
20:55So you give it away and then you have tons and tons of people, some of whom are undergraduate students or people, you know, in their parents' basement. So coming up with amazing ideas that you would never have thought about or willing to spend the time to crunch down the seven billion weight alarm hour so that it runs on a Mac or on a laptop. I think that's why open -source projects succeed particularly when they concern and basic infrastructure. So if you think about it, the early days of the internet, there was a battle between Microsoft and San Microsystems to provide the basic infrastructure for the internet, operating system, the web server, things like that, right?
21:29So on San Microsystems, it was Solaris and whatever web server and Java on the Microsoft side, I was Windows with IIT or whatever, you know, an ASP which was their kind of server and client side protocol. Both of them lost. In fact, San Microsystem pretty much went bankrupt and was sold for parts to Oracle. One was Linux and Apache, which is completely open -source, and you might ask why. The entire internet and the entire tech industry runs on Linux, right? And your phone probably runs on Linux too if you're Android. So that's three -quarter of the phones in the world. So the reason for this is that it's just a much better way of gathering competence and talent around a common project, even if it's not motivated necessarily by profit.
22:09Yeah, I agree, and I love this. What with matter? My question in David Marcus' question was how does matter win then? It's been the case that META in the past has opened towards pretty much everybody everything that it's ever produced in terms of basic infrastructure, right? So you have you know react for the framework for web and mobile apps. You have PyTorch. PyTorch is not even owned by META anymore. The ownership was transferred to the Linux Foundation Because it's so essential to the AI R &D infrastructure nowadays. Chat GPD was developed on PyTorch. OK. All open AI runs on PyTorch. The entire world, in fact, runs on PyTorch except Google.
22:47Because they had no one to eat, right? But it goes beyond that, right? Meta open sources, it's hardware, server, backplane design, so that hardware manufacturers can build to its specifications. And pretty much everything aside from legal issues that are sometimes due to recent laws or court decisions, pretty much everything has been open source. It is not because other people can use your technology that you can't exploit it to the same extent, right? Who can use Smart NLP systems for Transformation or content moderation on Facebook other than Facebook. It doesn't matter if other people have access to the same technology It's I totally agree with you and this can lead to my next question Which you actually tweaked about which comes to the size of data modes and size of data availability Is it simply a case that the largest model wins and how do you think about value in small?
23:35models as well. Yeah, so it's not the case. This is really what Nama has demonstrated and really kind of shown people. They've demonstrated that you don't need those models to be very large to work really well. And I think it caused a bit of a epiphany for a lot of people realizing, oh, okay, maybe you need 1000 GPUs running for a couple of weeks to train it the way system. In fact, that number is going down to two because people are figuring out how to do this more efficiently. But once it's pre -trained, you can use it for all kinds of stuff and you can fine tune it really easily. And then at the end you can run it on your laptop, right?
24:06That's going to be amazing or maybe on a desktop machine with a GPU in it or a couple GPUs So I think it opened the minds of people to the fact that there is like enormous opportunities that really weren't Thought to be possible before and I think it's going to make even more progress because if we go towards the design of AI systems Perhaps along the lines of what I described with objectives and planning. I think those systems could actually be even smaller to some extent and how would they be even smaller? Because for them to work, you have to train them on gigantic amounts of data. Way more data than any humans has ever been trained on.
24:37So the amount of data alumni is trained on, for example, is something like 1 .4 trillion tokens, which is a quarter of the internet, or something else, something absolutely enormous. It would take someone reading eight hours a day at almost the about 22 ,000 years to read through that. So obviously, those systems can accumulate a lot of knowledge from text, but they don't do it the same way humans do it, because we don't need that much time to be that smart and to learn that much. So obviously we are much more efficient or braze a much more efficient than those models at learning things. Like, how is it that a teenager can learn to drive a car in about 20 hours of practice?
Read the full transcript
25:11We still don't have level five start -roving cars. So obviously we're missing something really big. And what we're missing, I think, is abilities for AI systems to learn how the world works via observation mostly. And then this ability to plan so as to satisfy objectives. and then beyond that the ability to set so objectives in the satisfaction of a bigger one. Okay, that's parachycal planning. And we do this, humans do this, so many more do this, so mixed up. Every animal's mammal and birds is capable of some level of planning, or to reverse it with an element. Basically you don't do planning, or a very simple form of it.
25:42Yeah, and you mentioned the efficiency that can come from actually smaller models than expected, and how actually size of models is in everything. The other thing we spoke about open and closed, The other thing I've been thinking about, everyone's been thinking about my ventfube many, leading AI experts and they say, the value will accrue to the incumbents. Startups, they don't have the data, they don't have the models, it'll accrue to the incumbents. Is that right? Will the value accrue to the incumbents? Or do you believe that given what you just said about size not being everything in terms of models, it could be startups as well?
26:14So it depends on which scenario you believe in. So the scenario I think will happen and then so if you're waiting for is the scenario you were described earlier, you have some sort of open platform for base LLMs. So base LLMs basically would be seen as a basic infrastructure, TCP IP Linux Apache, essentially, completely open. And then there would be an ecosystem of companies building stuff on top of it, which for vertical applications for specific things, right, to specialize those systems for particular application, to offer support, to make it customized for enterprise applications, for personal things.
26:44There'll be like a whole economy around this, which will create jobs, by the way, not make them disappear. So this is the scenario that I believe will happen. And the reason I think it will happen is because there is essentially a need to use essentially millions of contributions for making those systems factual and correct, et cetera, so Wikipedia style. So I think the proprietary approaches will actually fall behind. So that's one point, okay. The second point is you can ask yourself the question, how is it that the companies that were best positioned to produce something that ChagyPT, namely Google and Meta, didn't?
27:15Why is it open AI? The small ad -shade was 400 people. And the answer is, it's not because Google or Meta did not have the competence of the technology. It's just that they didn't have the pressure to produce completely new products that had a lot of risk attached to them. And we know where the risks are because a few weeks before Chagypt, my colleagues at Fair, produced a large -range model called Galactica, which was experimental system. So galactic was an art -transition or a train to train on the entirety of the scientific literature And it was basically designed to help scientists write papers So you would start writing a paragraph or something like that to describe the top -you -go -paragraph and then galactic I would basically complete the paragraph and it wouldn't be factually correct It would have to fix it But it would like you would ask it to build a table of result and it would just put the light take commands to build the thing and populated with the known results on the literature, that the topic that you're working on, or you would type a chemical formula for something, it would turn it into an actual name for it, very useful for scientists.
28:16As soon as the demo was put out, it was murdered by the social network Twitter sphere. People said, oh, this is going to destroy scientific publication because now any random person can write a narrative -tively sounding scientific paper that is nonsense. And there was so much vitriol thrown at the system that the people at Meta who built it couldn't take it They took down the demo because they said we can't sleep at night So here is an example of a very useful system was it's under could have been extremely useful particularly for writers or scientific papers Who are non native English speakers that basically was destroyed by AI donors people who just did not think about the risk Benefit analysis the risk of Flooding the literature with non -sense is ridiculous because scientific publications are vetted and things like that There was not a significant danger.
29:03And then, Chagypti came two weeks later, and was welcome as the second coming of the Messiah. Why does that tell you? And then, a few months later, Google came out with Bard, and in the demo Bard made a tiny minor factual mistake about some astronomical fact, and Google's stock went down by 8%. Now, what that tells you is that when something is produced by a large company that has a reputation, particularly a reputation to defend, they can put out things that's too nonsense, but it's okay for a small company. So that's the landscape of what happens now, which is why I think is a bit of a Product which is said the companies that have the best technology basically can't have difficulties Preeding it out because of those legal issues and public image.
29:44Do you not also think there's just cool business model Challenge that which is it's the classic innovators dilemma like why didn't Google do this because it would have killed that Absolute cash cow of Google ads the cost to service a query versus the cost of this is so significantly different and you'd be killing your cool cash cow with this, with unknown upside, versus retaining what is a great business. You know the choice. There's no question that you could take a while, but there is no question that people interact mostly with the digital world using AI assistants, and they may run into your augmented reality glasses, okay, or something of that type.
30:19Like in the Spike John's movie, Her, that's not a bad depiction of what the way things could develop. And so if you take the assumption, the reactive assumption, this is gonna happen, you'll have to build it as quickly as you can. And it might cannibalize your new feedback with them or whatever or because Google you're search engine. But you have to do it. As meta has been known to make those choices in the past, like the move to mobile, for example, and the move to short form video, for example, which obviously TikTok has been very successful at. meta has entered that business in kind of a big way, despite the fact that the amount of revenue derived from it is lower than traditional new seed, because it's hard to put ads in video it was basically, you mentioned the job creation element now.
30:56I do just want to touch on the job site because it is the classic AI Doomer that we're all going to be unemployed and we're going to have Universal basically income and optimistic world. You said about job creation that we don't hear about job creation through AI. How do you see what jobs will be created through this new ecosystem and what that world of employment could look like? So a hundred years ago or maybe a hundred and twenty years ago, most people in most of the world worked in the fields in the food production, pretty much the majority of the population. Today in developed countries is between one and two percent and that has caused the migration of people into the cities and the development of a service business, same thing 20 years ago or 20, 30 years ago there was a big movement towards automation of manufacturing and a lot of manufacturing jobs disappeared in developed countries but they were replaced by other things.
31:45So 20 years ago if we would have thought that you could make a living with a podcast. I just didn't think I could five years ago and I'm surprised is everyone else. Right, a lot of jobs appear 30 years ago. There was no such thing as web designer and now I have engineers in the world basically do this, right? The number of economists that I've talked to which is pretty large where I asked that question. We tell me we're gonna run out of jobs because we're all gonna be replaced by I think is exactly zero. No economics believes this. No economics believes we're gonna run out of jobs because no economics believes that we're gonna run out the problems to solve or requirement for human creativity and human communication and stuff like that.
32:22This is gonna create as many jobs as it makes disappear. And those jobs, by the way, are gonna be more productive. So overall technology makes people more productive. In other words, for the same amount of hours worked, you produce more wealth, okay? But every technological revolution unless it's accompanied by political changes and social changes generally profit as well number of people at least temporarily, right? That happened in the industrial revolution in the late 19th century where a few people became extra rich and a lot of people were exploited. And then society changed and there were like social programs and income tax and high tax for richer people and stuff like that which the US says backpedal on this but not Europe.
33:00So there is a question of how you distribute the wealth if you want, okay? How do you organize society so everyone profits from it? But that's a political question because of the technology question. It's not you, it's not caused by AI, it's just caused by technological evolution, right? It's not a recent phenomenon. This is so unfair of me to ask. But what do those jobs look like? Are they creative oriented? But what does that actually mean? Sorry, I know that's a really hard question, but I'm just trying to understand how we actually spend all time in my children, which I don't have by the way.
33:27And what do they do? Like sculpt or paint? I don't know. I don't know. That's a good question. But it's not because I don't know that it won't happen. You look at how many people exercise their creative juices today, right, with all the tools that are available that weren't available 10, 20, or 30 years ago. Like 3D artists or something like this, you know, game designers, all kinds of things. So they're two types of jobs that they have a bright future of creative jobs, whether they are scientific, technical, educational or artistic. ACI has to do with communication, right? And communication of human emotions, which is intrinsically human.
33:59So that's why I'm categorical. And then the other one is personal services, so where you need actual people to interact with you. The only thing that I worry about is that the speed of transition, like when you look at past industrial revolution, when you put even the introduction of PCs into kind of working environments, these were multi -daccate introductions. Blondney, what AI feels like in some industries today, we use it as the media company and it's cutting our employ, like the speed of transition is much more compressed in this timeline, which will lead to short -term significant high unemployment.
34:32Do you concede that or do you not concede? So this is something I used to be really worried about, that the speed of progress of technology was going to leave a certain number of people behind who cannot be basically retrained, fast and not for big or maybe they are too old to retrain themselves for the new world. I was worried about this. And then I talked to a virtual economist and they say, oh, not really, because the speed at which a technology disseminates in the economy is actually limited by how fast people can learn to use it. A good person to talk to about this is Eric Binyovson at Stanford.
35:04And when he said this is that, when a new technology is introduced, let's say the PC, with graphical user interface, the mouse, etc. In the mid -90s, how long did it take to have a measurable effect on productivity, which is the amount of wealth produced by per hour worked? Typically it's 15 -20 years, and the reason is that's what it takes for people to learn to use that new technology. But do you buy that here? People are pretty good at prompts. Social media content managers are using prompts very efficiently to produce content plans, to create content ideas in under half an hour after watching a couple of TikToks.
35:36Yeah, but what is going to be the effect of this on first of all on measurable productivity? Second of all on the job market. Is it going to make people lose their job like right away? And no, it's going to take a while. It's going to take 10 to 15 years, possibly more. It depends when you start counting, right? Because the AI revolution made it start 10 years ago. So if you start counting then, then it might only take another 10 years. But I don't think you want to underestimate the degree of conservativeness of the business world, right? Things tend to change not that quickly. But if it's that easy to learn, like people will run it and then invent new professions out of it or become more productive themselves.
36:11Why do you think we love the doom, Jan? I love your approach and mindsets and I agree with it, but why do you think we are like neti -sized? Oh, we're all going to be unemployed in the doom. Because I think we're hardwired to pay attention to things that occur or may occur that could be dangerous to us. Because it means that there's something about the world that we don't completely understand and we have to pay attention to it and be careful about it. For example, take a young child, five months old, and show us an audio to this more child of a little car that is sitting on the platform, and then you push the car off the platform, and instead of following the car appears to float in the air, the five -month -old will barely pay attention to it.
36:46But if you show this to a ten -month -old, the ten -month -old will look at it with huge eyes and stare at it for a long time, watering was going on, because in the meantime, babies around the age of between six and nine months, run it by gravity, they run that objects and are not supported as opposed to fall. And so the mental model is that an object is not supported to fall. And they see this object, they appears to float in the air. And they say, like, this can be. Like, there's something I don't understand about the world. I need to look at this and investigate. OK, so we're hardwired for this because that's the way we learn our internal mental model of the world that allows us to predict what's going to happen, allows us to plan.
37:19That's what makes us smart. That's the basis of intelligence, the ability to predict. And so we naturally pay attention to stuff that is surprising or dangerous, or both, which is why you see a treasure piece of new clickbait at the bottom of some website and you have to convince yourself not to click on it. Can I ask you a couple of direct questions I'm just too interested and we can take them out if needed. What did you say to Jeff when you heard that he was obviously making moves that he did? I'm sure you had a conversation with him. What did you say to him? We haven't spoken yet actually. We're going to speak to get each other's opinion on it.
37:53I don't think he knows my opinion on this because I don't think he follows what I post on Twitter or whatever, even though he is on Twitter itself. So I think we have a discussion to have. I've had this discussion before with Yotra Benjou, but not with Jeff. And to me, the fact that he left Google is not particularly a surprise. The fact that he leaves Google to be able to speak his mind, I think is not surprising. So I have a very different D -Lat meta, which is that I say whatever I want. OK, I'm not under the type control of communications department or anything. I just say what I think. How did you get that deal, Jan?
38:25But no, seriously, many of my friends are not met in very high positions as with mutual friends. They don't have that deal. There is a particularly sweet spot because I have a quite a bit of a following people who trust me or believe me or want to hear what I have to say even if they don't trust me at all. And at the same time, I'm not an officer. So it's not that I see it's like I can't say because of illegal issues, you know, with an inch blah blah blah, right? and vice president, but I'm just below the level where you have to be really careful and control your message. And I think there is a cost -benefit trade -off here of AI is such a complicated, fast evolving issue that you basically need someone to be able to speak freely.
39:07And I think Jeff didn't feel like he had that option at Google, maybe for various reasons. So I just don't know why he might have wanted to leave, but I don't agree with him at all with the at the whole for the quality of human extinction or whatever. Have you ever felt your role at Massa as impede your ability to be impartial? I don't believe so. No. There are some things that I would post on social media that are popping up the work of my colleagues as I'm obviously biased about this because I know about the work and they are friends and or colleagues and I think it's interesting probably because I followed the part of the victim.
39:39So yeah, for this kind of stuff I might be biased, take this with a grain of salt, you don't have to believe me, things like that. But it's given me a vision also of pathiques are built, where the problems are. For example, there is a narrative, a very common narrative, that AI is the culprit for a lot of the bad side effects of social networks in the past. And in fact, it's completely backwards. AI is the solution to those problems. The video you backpedal 12 years ago was something. Even before I joined the meta, where meta started experimenting with a new speed. And the new speed was an algorithm that would pick which piece of news to show to everyone and originally it was decided by how friends are you with a person making the post and things like that, right?
40:19How many interactions you have with a person. Eventually, a bit of machine learning was put into it shortly before I, I don't matter. It was very simple. It was something like logistic regression, something like the simplest method you can imagine with a lot of engineering behind it and a lot of hacks by hand and special cases, but basically it was something like logistic regression. Some big vector that describes for what you click on, like how much time you spent on on a particular piece of content and blah blah blah, and then it would decide to give a rating to everything. That was deployed, and people ended up spending more time on Facebook, but then also it created problems that were critically identified, like information bubbles in the context of political discourse.
40:54And the fact that what I was talking about earlier, that people tend to click on things that is more outrageous, right? So it caused the appearance of cheekbait companies that big zero just like farms of teenagers in Montenegro was someplace making false use to get people to click on them and get money from the ads that you show them. Okay, so then this was realized there were like big groups at Facebook at the time studying the, where the effect of those things are, and this was corrected. So that's the way you make those some work, right? You try that most, most scale, you see what the effect is, if there is bad side effect, you correct it, and then you compare two systems, and then sometimes something unexpected occurs and you have to back that whole and completely change the way you do things.
41:31That's what happened in 2017. After the presidential election, the American presidential election in 2016, the main new CDR goes on as completely changed so that there was no clickbaits anymore. There was no like news outlets that could like push their content that was propaganda basically, you know, much more effort to take down false accounts and attempts to corrupt the democratic system and stuff like that. You correct it. And then what the progress of AI over the last few years basically allowed systems to be deployed to do things like take down hate speech, relatively reliably, in hundreds of different languages, which was basically possible to do before.
42:04He mentioned Krat to Elon Musk's adult with Tucker Carlson. The trouble with AI is you can't release and then correct. Unlike all prior technological developments, one's release is too powerful to be able to bring back into the box. It cannot be amended in that way. Is that not true? That's not true. That's completely false. It makes an assumption which Elon and some other people may have become convinced of by reading the box terms books, super intelligence or reading some of Eliezer Yutkowski's writing. So this is predicated on an assumption that is just false, which is the existence of a hard takeoff.
42:40All right, so the fact that the minute you turn on a super intelligent AI system is going to take over the world and is going to escape your control and it's going to refine itself to be even more intelligent and the world will be destroyed. And that's just ridiculous. It's just completely ridiculous because there is no process in the real world that is exponential for very long. Although systems will have to recruit all the resources in the world, they will have to be given live -it -less power agency like why would we do this? And what's more, they would have to rebuild so that they have a desire to take over.
43:10Systems are not going to take over just because they are intelligent. Even within the human species it is not the most intelligent among us that want to dominate others. So here's Desire and many other leaders desire to prevent any further development and to regulate intensely right now and stop all progression is BS basically. It's obscurantism. It's like people who wanted to stop the printing press on the diffusion of printed books because if people could read the Bible for themselves they wouldn't have to talk to priests anymore and then would have their own idea about religion. And this is exactly what happened.
43:42People read the Bible for themselves and that created the Protestant movement in Europe and that created two other years of religious conflicts. But you also brought to us the Enlightenment, science, rationalism, philosophy, ideas of democracy and then the French and American revolutions. And then you can go through this with the Ottoman Empire, which for reasons of being able to control their population forbid the use of the printing press, it starts to use the use of decline. They were dominating science in the Middle Ages, which is why every story this guy has an Arabic name. I love this. I'm going to do a quick fire round with you now.
44:13So I say a short statement in your immediate thoughts and then we'll rock and roll. Does that sound okay? That's good. Which regions most need to change their modus operandi when it comes to the practice of scientific research and incentive mechanisms. Pretty much every region I'm afraid, but for different reasons. If you start with China, so China has a bit of a epidemic of bad science. There are a lot of very smart people in China, a lot of very good researchers, a lot of very good work coming out of China, particularly in AI, particularly in computer vision, but a lot of absolutely terrible work that has to be retracted a few months later is being published.
44:47And it's probably because of the incentive mechanisms in the academic system in China, so this is probably too fixed there. I can move to Europe. So in Europe, there are good things. The education system for like undergraduates, education in Europe is great because it's party free. So that allows talented people to go to the schools even if they're not rich, right? Which is not the case in the US, for example. So that's good for Europe. European engineers, scientists, are great. A top based in the world. But then what are the opportunities for people who want to go into science and research? And there are most European countries actually don't have systems that really encourage this and motivate the most talented people and students to go into science.
45:26And so some of them go to North American, like me, 34 or 35 years ago. There are opportunities now that are really good in research labs like Fair in Paris or Google also as labs in Paris. Actually, if I bother, it works at Google in Paris and various other outfits. So that gives opportunities for people who really want to be productive and don't seek that they can in the public research and academic system in France and the rest of Europe. The only European countries that rival the US in terms of the quality of job for an academic or a scientist is Switzerland. What do you think they do to rival that?
45:59What is the badder incentive mechanism structure that gives them that ability? Two things. They pay people better. Second thing is they give them resources for research. They can get extra resources for grants and stuff like that, but they're good. And then they also attract some of the best students in the world. So you get the ideal accommodation that you only get in the top 30 universities in North American. So we've got China, we've got Europe. What about the US? What could they do differently or improve? There's a lot of the US does in terms of research, which is to a large extent a bit of a partial explanation for the success of the tech industry in the US.
46:34Partly because the US devotes a significant amount of resources to fundamental research, to NSF and NIH and various other outfits, probably more than Europe. University pays the faculty pretty well. Now this comes with a downside. The downside is that studying in the US is expensive and it's a trade -off, right? So can you do one without the other? Switzerland figured out how to take Academy pretty well while actually offering free education to their students. So there is a way to do it. Canada also figured out a pretty good trade -off as well. So in other things the US does right, but one thing that the US system or like the US does right also is the willingness to take risk and invest on ideas that seem a little crazy, but basically the sort of vibrant started seeing in the Silicon Valley and other places in the US, in New York, and in the Boston area, is leading the world.
47:23Now, you start seeing a similar thing in Europe now. There's been a enormous growth, for example, of tech startups in Paris, a Paris area in France, more generally, continental Europe, more wager, and in the UK as well. And so I think that's a good thing, but it's still more difficult to have access to investment money in Europe than it is in That's why I'm here Jan, I'm happy to provide. I'm into a penultimate one for you. When you think about what you'd most like someone listening to take away, what would it be? When they hear this, what do you want them to take away as the number one thing?
47:55AI is going to bring a new renaissance for humanity. A new form of enlightenment if you want, because AI is going to amplify everybody's intelligence. Every one of us will have a staff of people who are smarter than us and know most things about, you know, most things and most topics. So it's going to empower every one of us. It's going to make us more creative because we're able to produce text art, music, videos without necessarily having all the technical skills that are currently required for doing those things and so exercise our creative uses. So that's the positive side. There are risks.
48:28There's no question, but it's not like those risks. Don't believe the people who tell you that those risks are inevitable or that they will inevitably need to catastrophe. This is just not true. This was thought in 1920 who would have thought that a mere 50 years later you could cross the Atlantic in a few hours in complete safety, you know, at near the speed of sound. And would people seriously want to ban aviation or call for regulation of jet engines before jet engines existed? I mean that's kind of insane. So I'm not against regulation. There should be regulation of AI products, particularly the ones that involve making critical decisions for people.
49:05But regulating or slowing down research is complete nonsense. It's just love scurrentism. Who's incumbent team do you most respect? It's a my when you look at Amazon Facebook, Google, in terms of their approach and talent internally outside of matter, obviously? So this is changing a lot. And the reason it's changing is because a lot of people are leaving large companies and large labs. And the reason they're doing this is said until recently a lot of AI research was really exploratory and now there is a path towards commercialization for a lot of things. And so people think that they are better off just leaving large companies and doing this on their own, doing a startup and things like that.
49:42So you see a relatively large motion of a private research engineer, a few scientists, basically leaving those labs to do startups. And that's across the board, right? So you look at the original paper from Google about BERT or Transformers, right? They think that revolutionized NLP. All of them have left. They're all in startups. Some of the people who produce Lama, the open source slams from Mita. So the key people have left already to do startups. I saw that company's... Yeah, one called Mistral. Yeah, there is insane amount of money in days. So we're proud to them. But I'm sad that they have to have a Mita, I was just thinking, but he's exactly in teaming me like, yeah, they're good.
50:22Yeah, yeah. But I think in terms of the so basic competence and the people who are going to push the science forward because what we need now is not to work on applications of L &M, there's a lot of people who are capable of doing this and they're going to do a good job. What we need to do, people like me, who are really working on research, is coming up with new concepts that will allow us to get machines that basically have common sense, have an experience of the real world, have basically human -evalid intelligence. And in my opinion, the outfits that are best positioned for this are fair from one side.
50:52And the new Deep Mind, now, which is Deep Mind Periscuba Brain. Yeah. There are a lot of people there who are interested in that question and I think there are probably the best together with Meta, fair, the best position to really have an impact on this something all of us have been working on for quite a while. Yeah, and if we do this again in 10 years time, where is Jan in 10 years time in 2033? Well, I'm 63, 10 years old now. Well, 12 years old now, I'll be Jeff in terms of age, okay? And I'm excited like a teenager now because I see the opportunity of the next step in AI and the opportunity perhaps to get to the goal that I and myself, so that I imagined for myself when I started working on AI many years ago, which of course I was very naive about at the time of understanding intelligence, first of all, it's a scientific question.
51:38What is intelligence? What is human intelligence? And one good way, as an engineer, a good way to understand intelligence is to build a widget that actually reduces it, right, to some extent. So I'm excited about this right now. I'll find the substrate, the landscape, the location, the position where I can make the best contributions to this and currently that just happens to be to be fair. At Meta, I keep a foot in academia because I think it's a very complimentary and also important the projects of different types that you do in academia and industry that are complimentary. So I like the combination of the two.
52:11As long as my brain keeps working that I see I can contribute and that I've given the means to contribute, I keep working. And then at some point where my brain will turn into white sauce or for totally out of it or something and I'll stop. Yeah, I want to say personally. Thank you so much and speak for many. I'm sure when I say we've learned so much from you in terms of your public Speaking and discourse and willing to speak. I think fewer willing to speak as openly as you have been so Thank you for educating so many of us and thank you so much for joining me today and thank you so much for having me.
52:41Yeah, this was fun I mean what an awesome show if you want to see the full video of course you can check it out on YouTube by searching for 20 TVC, but before we leave each day, are you building enough conviction to outpace other investors in this changing ecosystem? Tegas helps VCs get under the hood of their investment quicker. Map of all being markets with deeper research, monitor the financial health of portfolio companies and even source new deals. Tegas is trusted by top firms like Red Point and Spark Capital for everything from expert calls and transcripts to quickly surfacing insights across filings, comps and banjo marking.
53:18Find out why a majority of the top venture funds are using Teague as on a daily basis to differentiate, consolidate and accelerate their research, head to teagueas .com for slash growth for your free trial. And speaking of tools we cannot live without there, Secure Frame is the leading all -in -one platform for automated security and privacy compliance. Secure Frame simplifies and streamlines the process of getting and staying compliant to the most rigorous global privacy and security standards, like SOC2 and others. and secure frames industry leading compliance automation platform paired with their in -house compliance status and for more auditors helps you get all it ready in weeks not months so you can close more deals faster automate your security and privacy compliance with secure frame scheduler demo today at secure frame dot com that really is a must and finally are you tired of using the same old tools and platforms in an attempt to manage your global team will ask to why use 16 different tools when you could just use deal.
54:16Deal is the only global first HR tool with unmatched automation, customizations and integration, making it the only platform for building contracts, visas, background checks, payroll equipment, benefits and beyond. Are you ready to see how deal can scale with your global and local teams? Visit deal .com forward slash 20 VC and book your demo today. That's deal .com forward slash 20 VC to book your demo. Now stay tuned because on Wednesday we have a mad at stability .ai. Join us in the horse. He's such an incredible show coming.
From the publisher
Yann LeCun is VP & Chief AI Scientist at Meta and Silver Professor at NYU affiliated with the Courant Institute of Mathematical Sciences & the Center for Data Science. He was the founding Director of FAIR and of the NYU Center for Data Science. After a postdoc in Toronto he joined AT&T Bell Labs in 1988, and AT&T Labs in 1996 as Head of Image Processing Research. He joined NYU as a professor in 2003 and Meta/Facebook in 2013. He is the recipient of the 2018 ACM Turing Award for "conceptual and engineering breakthroughs that have made deep neural networks a critical component of computing". Huge thanks to David Marcus for helping to make this happen.
In Today's Episode with Yann LeCun:
1.) The Road to AI OG:
- How did Yann first hear about machine learning and make his foray into the world of AI?
- For 10 years plus, machine learning was in the shadows, how did Yan not get discouraged when the world did not appreciate the power of AI and ML?
- What does Yann know now that he wishes he had known when he started his career in machine learning?
2.) The Next Five Years of AI: Hope or Horror:
- Why does Yann believe it is nonsense that AI is dangerous?
- Why does Yann think it is crazy to assume that AI will even want to dominate humans?
- Why does Yann believe digital assistants will rule the world?
- If digital assistants do rule the world, what interface wins? Search? Chat? What happens to Google when digital assistants rule the world?
3.) Will Anyone Have Jobs in a World of AI:
- From speaking to many economists, why does Yann state "no economist thinks AI will replace jobs"?
- What jobs does Yann expect to be created in the next generation of the AI economy?
- What jobs does Yann believe are under more immediate threat/impact?
- Why does Yann expect the speed of transition to be much slower than people anticipate?
- Why does Yann believe Elon Musk is wrong to ask for the pausing of AI developments?
4.) Open or Closed: Who Wins:
- Why does Yann know that the open model will beat the closed model?
- Why is it superior for knowledge gathering and idea generation?
- What are some core historical precedents that have proved this to be true?
- What did Yann make of the leaked Google Memo last week?
5.) Startup vs Incumbent: Who Wins:
- Who does Yann believe will win the next 5 years of AI; startups or incumbents?
- How important are large models to winning in the next 12 months?
- In what ways does regulation and legal stop incumbents? How has he seen this at Meta?
- Has his role at Meta ever stopped him from being impartial? How does Yan deal with that?
