Who's Winning the AI Race? With David Mattin

0:00If you've been considering futures trading, now might be the time to take a closer look. The futures market has seen increased activity recently, and Plus 500 Futures offers a straightforward entry point. The platform provides access to major instruments, including the S &P 500, NASDAQ, Bitcoin, natural gas, and other key markets across equity indices, energy, metals, forex, and crypto. Their interface is designed for accessibility. You can monitor and execute trades from your phone with a$100 minimum deposit. Once your account is open, potential trades can be executed in two clicks. For those who prefer to practice first, Plus500 offers an unlimited demo account with full charting and analytical tools.

0:45No risk involves while you familiarize yourself with the platform. The company has been operating in the trading space for over 20 years. Download the Plus500 app. Trading in futures involves the risk of loss. It is not suitable for everyone. Not all applicants will qualify. Hey, everyone. If you like this podcast, go behind the paywall to get privileged access to the smartest minds in finance. Visit realvision.com slash RVpod and use the promo code podcast10. That's podcast10 to get 10 % off our essential membership for the first year. Join the Real Vision community and learn how to become a better investor.

1:24And now to the top analysis of today's markets.

1:40Who's winning the AI race? Hi, everyone. Welcome to the Real Vision Daily Briefing. With me today is David Matten, founder of New World Same Humans. Hi, David. It's great to see you. Maggie, hi. How are you? It's great to be back. I'm feeling better because it's Friday, but we got a lot to cover. And we want to talk a little bit about AI. We're more than halfway through the U.S. session, and it looks like it's going to be another losing week for U.S. equities. They've been contending, of course, with rising yields on treasuries and some volatility there, concerns about China. But underneath it all has been this sort of ongoing worry about tech valuations, about valuations in general, but especially about tech valuations, which soared this year on the back of a lot of enthusiasm around AI.

2:28And a lot of people wondering whether that's run its course or if we're still in the early innings. So we wanted to sort of take a step back and just get a view of the landscape, because you really live at this intersection of what's happening with technology and these massive transformational changes. And I know you're watching the AI space closely. So where are we in terms of what companies are making the most progress when it comes to positioning themselves for this massive opportunity? Kind of give us an update of what's going on. Yeah, I mean, I think fundamentally where we're at is just a scramble among the big technology players to lead this space and to own it.

3:12And we can all see that Microsoft took an early lead. I mean, Microsoft are partnering with OpenAI. OpenAI have the slam dunk most exciting large language model right now and the slam dunk kind of product right now, which is ChatGPT. And Microsoft are their partners. Really interesting news this week, though, coming out of Alphabet. I mean, earlier in the year, there was a lot of intrigue around what was going on inside Alphabet because it took its two big famous AI divisions. One is Google Brain, one is DeepMind, and it just merged them. It just kind of threw them together instantly. And the news coming out of Alphabet now is that it is about to launch a set of large language models called Gemini, and that they're going to be kind of the largest, most cutting edge, most multimodal large language models out there.

4:08So not only will they produce text outputs, but they'll produce images, they'll produce graphics, they'll allow you to control software by voice command. And what this is clearly is just the big play to go head to head with OpenAI and Microsoft to be the leaders in this space. And it's just going to be fascinating to watch that play out. I mean, it's going to be a case study for the ages inside business schools as to how Google lost the lead when it comes to this incredible generative AI revolution. Because don't forget that the research paper that kicked this whole thing off, that invented the transformer model that underpins generative AI, that research paper came out of Google.

4:49That was Google researchers. Attention is all you need. It's the iconic research paper that kicked this whole moment off. And somehow they lost that momentum. They lost that lead to Microsoft and OpenAI, and they are going full out to get it back. And it's just going to be a fascinating few months, because that's how quickly it moves, watching that play out. In the meantime, you have Meta launching its big new large language model, LLAMA 2, and partnering with, guess who, Microsoft to open source that, to bring that to market, to allow developers to build on top of that. They report for free. There are some reports that it's going to be a little more complex than that, but they're essentially saying build on top of our large language model for free.

5:36If you want to build on top of the OpenAI large language model, which is GPT-4, that costs you money. they're saying you can do it for free. We're partnering with Microsoft to bring you that. I think there must have been some very awkward conversations around the table between Microsoft and OpenAI when they launched this big partnership with Meta. The whole thing is very weird bedfellows on that front, isn't it? But is there a sense, first of all, super interesting, super interesting to hear that about Google. But their hope for people who know Google and who've been following it, Apple, I should say, but this really goes back to the Google days.

6:13They have that whole Project X. I had a fascinating conversation with a guy who ran it one time, came in and gave us all a briefing. That's what they do, like the Hyperloop, half the stuff that Elon said, pretty much all of this stuff out there. A lot of it comes originally from the researchers at Google, right? Because they have that like, break it, break it, break it. And then let's see how far it can go, that culture of sort of trying to see if they could get it to fail. And sometimes it goes really far. And then they just say, oh, there's not a market. We're not going to bring it. And then they let it go out into the universe.

6:46So it sounds like that's maybe what happened in that case. Can they, before we talk about Zuck, can they catch up? I mean, it seems like with, what's the barometer for success for catching up? So they've got the money. They also have a lot of data, don't they? Yeah, I mean, that's, yes, they can catch up. I think, you know, they have every chance of catching up and taking the lead because look, they have a whole, they have the world's best data set, you know, 8 billion and rising Google searches a day. And don't forget, you know, every single YouTube video in existence. They've trained Gemini, this new multimodal large language model on a whole ton of YouTube videos.

7:37So it's going to be, you know, bets are it's going to be text to video as well. You will be able to type a text prompt and generate a video or type a text prompt and generate music for, you know, this kind of video. You know, I'm creating a marketing video for a badger farm. I want music for that. Give me some music. And it will be able to do that because it's trained on like a billion YouTube videos. So they have all the data. And then don't forget, they have instant deployment across number one, search, obviously, and number two, you know, a suite of tools, Google documents, the whole lot. They can just throw it into those tools used by billions of people kind of instantly if they want.

8:16I mean, OK, Microsoft have Microsoft Office. You know, they can throw the open AI tools into Microsoft Office and they've kind of been experimenting with that. But yeah, Google can definitely take the lead back if they play this right. But you're absolutely right. They kind of have a history of incredible moonshot technologies, incredible R &D that they struggle to sort of commercialize. I mean, you saw it with self-drive as well, which has been a thorny, a more difficult problem to solve than we thought that we are now getting there. That's a whole other conversation. But yeah, they took an early lead there.

8:49And it's felt that that early lead has dissipated. When we talk self-drive now, we don't just talk about Waymo.

8:59It's going to be deeply interesting. But like you say, incestuous is the right word. It's becoming deeply overlapping and deeply incestuous. And it's just a powerful insight into the way the evolution of these technologies stresses the nature of the competitive market-driven system. These huge organizations are on paper in intense, deep, perpetual market-driven war with one another, but they are also now deeply reliant on and intertwined with one another. If you're an alien looking at this from space, it's a very strange system we're running to push our technology forward. Do they hate each other or do they love each other?

9:45It's sort of a bit of both. Do they need each other? Yeah, they need each other. That may be the answer to those. The answer may be yes, yes, but they need each other, so they've got to. So it's interesting because I think it was just today, Microsoft CEO Satya Nadella, who has, to be fair, always been focused on AI. I mean, every time I interviewed him, we had to talk about Cortana. You know, like he was always thinking about this, which is probably why they were able to seize on the commercial. But he's saying AI is as big a tidal wave as the Internet, referencing a memo that Bill Gates wrote back in 1995.

10:21So there is a sense that this is going to eclipse that. Can they build a moat around these business models? So we're looking at we didn't even talk about Apple, but we can. So Facebook's kind of figuring its way out. it sounds like. Can these businesses build a moat around it? Or based on what you just said about kind of being in bed with each other and having to rely on each other, is it going to be harder than it was in the internet? And a lot of people were unhappy with the way that they were able to own the internet, by the way. So there's that recognition as well. Are those business models, those revenues going to be able to be moated because of their advantage of AI?

10:56Or is it not going to be that simple this time around? It's certainly not simple. it's certainly not simple but I think there is something of a moat to be built around having the largest most sophisticated most powerful models out there you know whether they're whether they're text to text like chat GPT or text to video or text to image whatever it is you know there's going to be a position to be built around having the cutting edge most sophisticated most powerful multimodal model. And only a handful of huge technology companies have the kind of resources and the compute power that you need to train that.

11:40And if you do that and you build a lead, you can build a moat there. But there will be an incredible long tail of much smaller, much more domain-specific models that less resourced organizations can build, that startups can build. And there'll be a huge amount of play and innovation there too. I mean, you don't always need, you know, you don't always, you don't always need the Ferrari. Sometimes you just, you know, it makes no difference whether you're, it's sometimes it makes no difference whether you're, you know, whether you've got a swimming pool, the size of an ocean or a swimming pool, the size of your back garden, you know, if you just want to go swimming.

12:16I like how you were struggling to find another car brand other than Ferrari, David, which might give us a little peek into your fancy lifestyle. I'll name off any of the very pedestrian cars I have. Yeah, I mean, there's going to be a lot of play in the long tail, but I don't think it's true that there's no possible moat. I think you can build a lead here and build a moat around that, and a lot of people are going to want to build on top of the most powerful, most sophisticated model. And that model becomes essentially a form of infrastructure. It becomes a sort of something akin, as I've said so often before, to electricity, to a new form of energy.

12:57If you're the supplier of intelligence to a million other developers building on top of your model, and that's what we're seeing with GPT-4 right now, there's a whole ton of startups that are just, they're building a wrapper. I know we're using that word a lot these days and it's questionable. They're building a wrapper around GPT-4 to make it easier to use, to make it do something specific. If you're the technology at the heart of that, right now it's GPT-4, but it doesn't have to stay that way. That's a powerful place to be. Hey, everyone. We're going to take a quick break right now to hear a word from our partners.

13:34We'll be right back with more of the day's top analysis on the Real Vision Daily Briefing. Have you ever wanted to trade Bitcoin but haven't dared try? With Plus500 Futures, you can trade crypto without the hassle of opening a wallet. With just a few clicks, you can register and start practicing with their free and unlimited demo. See a trading opportunity? You'll be able to trade it in just two clicks. Feel ready? You can move to real money with as little as$100 once your account is approved. And the great thing is that in addition to crypto, Plus500 gives you access to a wide range of instruments.

14:06S &P 500, NASDAQ, gas, and much more. explore equity indices energy metals forex and beyond with a simple and intuitive platform you could trade anytime anywhere experience the fast accessible futures trading you've been waiting for with plus 500 with over 20 years of experience plus 500 is your gateway to the markets visit us.plus500.com to learn more trading and futures involves the risk of loss and is not suitable for not all applicants will qualify plus 500 it's trading with a plus

14:42which is contrary to some of the other ideas out there that we can't moat this. So that's super interesting. And of course, raises so many other questions that we're not gonna get to today. But I know that you're diving into it on a regular basis, on the shows that we do on the platform, because there are regulatory issues, there's societal issues, there's military, geopolitical implications from that sort of that concentration of power. I wonder, so that's the sort of development side of that. On the other side, it's kind of the receiving the football side of it, which is adoption. And that's the part that's sort of, I think, touching a lot of us, certainly from a worker's point of view.

15:27We had a fantastic discussion on our platform between Jordi Visser and Sultan Megji. And they talked about the decisions facing executives right now. Let's have a listen to that. At the end of the day, the most effective companies, the companies that are going to make up the most, you know, the highest revenue businesses, the largest market cap, you know, these mega cap companies, they're not going to be companies that restrict AI. They're going to be companies that give agency to the people in their offices to go off and be creative and be explore and find ways to make their, you know, make their jobs better.

16:01And the companies that understand that AI is going to be the fastest accelerator to that kind of value creation are the ones that I get interested in, that I want to invest in. And the ones that don't are the ones I'm going to avoid, frankly. And I think as we think about building, whether it's investment pieces moving forward, companies to look at, or what the composition of the S &P 500 looks like a few years from now, I think this becomes a fundamental pivot point for a lot of these companies. You either embrace it and make it part of your daily life or you start looking, you know, start looking elsewhere because you're going to end up losing all your best people and therefore you're going to lose the ability to be innovative and drive value.

16:47I mean, the amount of knowledge that that guy has is incredible. You should listen to his bio. For those of you, I highly recommend it. I mean, he's touched so many parts of tech. That full interview is on the platform. as are some really fascinating conversations with the executives in charge of digital and Web3 brand build out at Adidas, Starbucks, LaCoste. We have this whole series going. Really, really interesting about how these companies are sort of forward positioning themselves. Check them all out because I think they have real valuation implications. As you know, we've closed the site to new members if you're watching on YouTube or listening on YouTube, but there is a wait list to get on it.

17:29just scan the QR code and it'll lead you down that path. So David, I thought what they were saying was very interesting. So again, we're thinking about how it's going to impact us as workers, but these executives are figuring out what they need to spend, how quickly they need to spend, how they're going to deploy it. And Jordi and Sultan were talking about the fact that possibly as much as 40 % of the current S &P 500 are going to potentially swap out or disappear if companies don't embrace this technology? I mean, that's how important they feel like it's going to be to survival of companies. I mean, that sounds like a really big number, but is it that important to get that right?

18:10And how quickly, what should we be listening to? Should we be zoning in on this and earnings calls to figure out what companies are doing with AI to determine whether they're going to remain competitive? Yes, I think so. I mean, I think it would be worrying if an organization wasn't thinking deeply about that. I mean, fascinating to look at, for example, McKinsey just a couple of days ago announcing their specialist large language model, essentially their specialist AI that they call Lilly. They've trained it on, you know, hundreds of thousands of internal documents and videos and presentations.

18:46And now this is essentially an AI companion that every single McKinsey consultant will have access to 24-7, whenever they like to ask questions, to give them guidance, to help generate work for them that they can then edit. It's just a constant virtual companion for every single consultant inside McKinsey. Now, we'll get to see how that plays out and how and whether it supercharges the productivity of McKinsey consultants. Yeah, efficacy matters, though, right? Because everyone's talking about it on conference calls. I think this is what's difficult for investors. Everyone's saying my AI strategy because they know if they don't, well, they know it'll pop their stock.

19:24And if they don't talk about it, it'll push it down. Trying to determine who's doing it right is a totally different story, though, isn't it? It's yeah. And there's no easy answers to that right now because we're in such an experimental phase. I think you have to be you have to be looking closely at what they're talking about, the credibility of the people they're bringing in to lead these AI strategies. You know, that's hugely important. And then we just we have to watch it play out. It's very hard to conceptualize right now the impact of AI of this AI moment on on the real economy and on sort of billions of knowledge workers.

19:58But I think it's going to have a profound effect. And some of them will be deeply surprising. I mean, you know, across the life sciences, what it's going to do to drug discovery, to health care, you know, to the way we access health care. It's go, as Raoul says, it's going to be hugely deflationary. It's going to change productivity. It's going to change the nature of productivity and the nature of it's going to change headcount inside organizations. it's just too early to say how you get that right and what the best uses end up being. Yeah, because there's a lot of disruptive change maybe before the payoff.

20:38By the way, the title of that conversation they had was From Scarcity to Abundance. And they talked about energy too in that with Sultan and Jordi. So super, super interesting and big implications. So in order for this technology to work, though, we need chips, right? NVIDIA reports earnings next week. It's going to be monster for the market because so many people, even people who are convinced equities are going much lower, are scared to short NVIDIA. Like everyone is just very, very nervous about that juggernaut because everything you think, every time people think they're not going to deliver, they do, partly because of what you're talking about.

21:14But you're closely watching something that's going to be really important to all the chipmakers, and that's the latest moves from the Biden administration. It's been a really busy week, so some of this kind of maybe fell to the back burner, especially when we had natural disasters hogging up all the headlines, rightly so. But what do we need to be paying attention to in terms of what's happening on that front? Yeah, I mean, I think we're seeing the most incredible moves from the Biden administration when it comes to the AI race and when it comes to the compute race. I mean, you know, we're talking about who's winning the AI race.

21:48We're in a race to own more advanced artificial intelligence. We're in a race ultimately towards super intelligence. And who has the lead? I mean, at the moment, I think most people would agree, you know, it's between the US. If we're talking geopolitically, it's between the US and China. China are a step behind, but only a step. They have a slight edge on the amount of research papers coming out. They have an incredible domestic kind of talent when it comes to AI. But look, the most advanced models are coming out of the US right now. The US has control of compute power, and we'll get to that in a moment how crucial that is.

22:29And a lot of Chinese talent comes over to the US. what we're seeing now from the Biden administration. I mean, a few weeks, a few couple of weeks ago, essentially, Biden signed an order that bans the investment in domestic Chinese advanced chip and AI chip industries and bans the sale of those kinds of chips to China. So, you know, most startlingly, the NVIDIA A100 GPU, that's the chip that trains these huge large language models like GPT-4. The sale of that is banned now to China, and he's weighing in the balance further moves in that direction, further moves to ban the investment or sale of advanced chip technology and AI technology to China.

23:15He's persuaded other big players, I mean, the Dutch huge chip technology manufacturer, ASML, and he's persuaded Taiwan as well, probably a bit easier to persuade them to go along with this. That is an incredible moment. That amounts to steps taken to just cripple China's advance in the AI race, to cripple it essentially technologically. And then what you're talking about is slowing it down dramatically in terms of science, in terms of innovation, in terms of economic growth. That is a huge geopolitical moment. I mean, there are people out there saying this is as significant a moment geopolitically as the war in Ukraine.

24:00And I have to say, I completely agree. It's just an incredible indicator of the way compute power has become a fundamental, perhaps the most fundamental geopolitical issue, how it's going to reshape the kind of world order, the geopolitical order, and how, you know, points of tension like Taiwan are going to arise out of that. We're going to take another quick break to hear a word from our partners. We'll be right back with more of the day's top analysis on the Real Vision daily briefing.

24:35And you guys had a terrific Twitter spaces on that, which you can still find talking about some of the sort of intersection of geopolitics and AI right now. Ed saying most people don't know how deep Tesla is in real world AI. We didn't touch on that when we were talking about some of the sort of arms race with the mega cabs. How are you thinking about Tesla in relation to AI? I mean, he's absolutely right. I mean, it's often said, Tesla, they're not a car company. They're an AI company. And more specifically, and I think this is the part that isn't perhaps understood quite so well. They're a machine vision company.

25:16Tesla has incredible machine vision technology. And you've got 160 ,000 Teslas driving around now with their kind of full self drive technology. It's not quite full self drive, but it's approaching it, collecting huge amounts of data from the open road, from open streets that they're using to refine and refine and refine what is essentially the world's leading machine vision AI. And then at the same time, they're developing the Optimus robot, this humanoid robot that they want to be in, like every factory and fulfillment center around the world. And it will run on exactly the same machine vision technology.