Narrator / Host (1:56)

And now there's all kinds of problems. Notice that this bishop on C6. The bishop can easily fall victim to what we call an overload tactic.

Gregory Warner (2:06)

And this game was intense. I remember watching it on TV at the time, looking at Kasparov running his hand through his hair, getting up, walking around the room, coming back to the board, concentrating on each and every move he made while the machine is just this glowing screen, very methodical, making its moves quickly and decisively.

Narrator / Host (2:29)

He looks disgusted. In fact, he looks just like he can't believe what's going on right now. Is an unhappy camper.

Gregory Warner (2:35)

And in the final match, our human champion fell into a trap.

Yoshua Bengio (2:41)

And whoa.

Narrator / Host (2:42)

Kasparov has resigned in an absolutely stunning, stunning 19 mover. Kasparov has just simply stormed away. Machine didn't beat man, but trounced him. Newsweek magazine called it the brain's last stand.

Liv Barry (3:05)

The game of chess Supposedly, a true test of human intellect will never be the same again.

Gregory Warner (3:11)

Now, for a while, it did feel like we were witnessing a moment of profound change.

Narrator / Host (3:16)

Call it a blow against humanity. The victory seemed to raise all those old fears of superhuman machines crushing the human spirit.

Gregory Warner (3:31)

But pretty quickly. While it was a very impressive feat that the programmers at IBM had pulled off, it wasn't actually transformative. I mean, it didn't really change. Chess in the years since, chess has only become more popular, both in the amount of chess people are playing, in the viewership of chess competitions human to human, and in the world outside of chess. This did not usher in an age of competition between humans and true thinking machines. And while some of us looked at this and shrugged or perhaps breathed a sigh of relief, there were those who looked at this moment and thought we were playing the wrong game.

Karen Howe (4:08)

Although I was fascinated by these daily chess programs that they could do that, I was also slightly disappointed by them. Because Deep Blue, even though it was a pinnacle of AI at the time, it did not seem intelligent that way.

Yoshua Bengio (4:19)

That Deep Blue beat Kasparov was nothing like how human chess players do it.

Karen Howe (4:27)

I was actually more impressed with Kasparov's mind than I was with the machine, because he was this brute of a machine. All it can do is play chess, but it couldn't even play tic tac toe. And then Kasparov can play chess, but also can do all the other things that humans can do. So I thought, you know, doesn't that speak to the wonderfulness of the human mind?

Gregory Warner (4:47)

For today, the games and the gamers that brought us the AI revolution and the small band of contrarian scientists determined to make an AI that didn't think like a machine. Okay, Amy Mills, Gregory Warner, let's talk about games.

Liv Barry (5:09)

I actually think one of the interesting ways of following the path of AI is, funny enough, through games.

Narrator / Host (5:15)

So I originally got into this connection between games and the history of AI through Livburi.

Liv Barry (5:19)

My name is Liv Barry, and I used to be a professional poker player for a long time. My original background is actually in physics.

Narrator / Host (5:27)

She herself is very good at games. She's famous for being a world champion poker player.

Gregory Warner (5:31)

We met her in Episode one, right?

Narrator / Host (5:33)

Yeah. She spent a good chunk of the last decade publicly advocating for AI safety. And that's in part because she's not just a game player, but she's a game theorist.

Liv Barry (5:43)

Being able to solve a game is being able to understand a particular environment where you have different objectives, different scoring metrics. Perhaps the environment can be changing and Being able to adapt in order to be the best at that goal. So as artificial intelligence gets better at game theory, essentially in increasingly complex and increasingly real life environments, then they are getting better at navigating the world. And that's really the trajectory we're seeing. When you hear people talk about AGI, this idea of artificial general intelligence, what they're really saying is an agent that is as good as a human generally is at navigating all of the different things in this world that we live in.

Gregory Warner (6:29)

So, on the one hand, games are this very practical kind of benchmark, right? Like just a way for computer scientists to test how their system is doing. But as the games get more and more complex, they are getting closer to that holy grail of a thinking machine.

Narrator / Host (6:46)

Yeah, that's the idea. And it goes back all the way to the 1950s to the first generation of AI researchers.

Liv Barry (6:53)

Some of the earliest computer programs were built to try and play basic games.

Narrator / Host (7:00)

Tomorrow, a preview of the future as it begins to take shape in the laboratories of the world. One of the earliest ones came from the uk. It was a system that played tic tac toe, but because it was the uk, they called it knots and crosses. In his spare time, engineer D.W. davis built an automatic noughts and crosses machine that thinks for itself by its own effort. It selects from the 6045 alternatives, the one that always wins.

Liv Barry (7:26)

And then over the rest of the 20th century and into the 21st century, we saw games of increasingly more complexity be defeated by computers to the point of superhuman level.

Narrator / Host (7:37)

Fast forward to the 1970s. You've got an AI system that can play checkers. That man isn't playing checkers against a computer, is he? Sure.

Geoffrey Hinton (7:45)

And it plays pretty well, sometimes even better than the men who designed them.

Narrator / Host (7:49)

After that comes backgammon. Then, in the 90s, you get the famous chess match.

Keech Hagee (7:53)

IBM's Deep Blue computer demolished the greatest chess player ever, Garry Kasparov, in the final and decisive game of their match.

Narrator / Host (8:02)

And then by the time you get to 2011, you've got an AI system that challenged humans to what was seen at the time as the most ambitious game yet. This is Jeopardy.

Gregory Warner (8:13)

The IBM challenge.

Narrator / Host (8:16)

And now here is the host of Jeopardy. Alex Trevette. So, IBM, they're back again, and they're doubling down on the same strategy that they used with Deep Blue to win in chess. They've poured millions of dollars and years of research into winning Jeopardy. Because as they see it, this is an even more complicated challenge for an artificial intelligence.

Geoffrey Hinton (8:42)

Language is an Area where from the.

Gregory Warner (8:44)

Very beginning of the computer era, people.

Geoffrey Hinton (8:47)

Kept expecting computers to do reasonably well.

Narrator / Host (8:49)

At they expected computers could talk. And so far, the computers have failed to deliver on this promise. On top of having to answer questions on any different subject. You know, history, pop culture, philosophy, this system will have to speak and understand language. A little over three years ago, the folks at IBM came to us with a proposal that they considered to be the next grand challenge in computing. And that was designing a computer system that could understand the complexities of natural language well enough to compete against Jeopardy's best players. Jeopardy. And IBM, they hype up this big three night showdown. So you are about to witness what may prove to be an historic competition, an exhibition where the AI system, Watson, is going to go head to head against two of the best players in Jeopardy history. Ladies and gentlemen, this is Watson. Let's play Jeopardy.

Gregory Warner (9:45)

Here we go.

Narrator / Host (9:51)

And at first, things start off a bit rocky for Watson. The human players, they're neck and neck. Stylish elegance or students who all graduated in the same year.

Gregory Warner (9:59)

Watson, what is chic?

Narrator / Host (10:01)

No, sorry. What is class? Class. You got it. But by the second night of this three night special, and anytime you feel the pain, hey, this guy. Refrain, don't carry the world upon your shoulders. Watson, who is Jude.

Geoffrey Hinton (10:17)

Yes.

Narrator / Host (10:17)

Watson. Starts crushing the humans, losing to him by 100th of a second. Watson, who is Michael Phelps. Yes. Black holes boundary from which matter cannot escape. Watson, what is event Horizon?

Jasmine Sun (10:32)

Yes.

Narrator / Host (10:32)

Watson, who is Grendel? Yes. Watson, what is the last Judgment?

Karen Howe (10:37)

Correct.

Yoshua Bengio (10:37)

What is London?

Karen Howe (10:38)

Correct.

Narrator / Host (10:38)

What is stick? Stick is right. And with that, you have add to your lead. You're at 5,000. And just like with Deep Blue beating the world's best chess player, there was this moment afterwards where it felt like we might really be witnessing some kind of milestone.

Sponsor / Advertisement Voice (11:00)

IBM says the technology could help speed up medical diagnosis and other challenging computing.

Narrator / Host (11:07)

IBM, they put out this documentary saying this is going to be transformational. All these things are now going to be possible.

Sponsor / Advertisement Voice (11:14)

Of course, this whole project is not ultimately about playing Jeopardy. It's about doing research and deep analytics and a natural language understanding. This is about taking the technology and applying it to solve problems people really care about. We're just so excited about all the things we can do with this.

Narrator / Host (11:35)

But again, just like with Deep Blue, the strategy was very good at winning a complicated game, but it failed to live up to the hype. It failed to lead to anything very useful outside of the world of that game.

Gregory Warner (11:49)

Okay, so you're saying that for decades, computer Scientists were testing their AI systems against the world of games under the theory that as the games got more complex and, and as the skills they had to program in became more interesting, that somehow those skills would translate into the real world. And yet none of these AI systems that can play these games can make the jump into real life.

Narrator / Host (12:16)

Yeah, they can't make that jump that Liv Barry was talking about.

Gregory Warner (12:19)

Their intelligence does not transcend into the real world.

Narrator / Host (12:23)

Exactly.

Gregory Warner (12:24)

And what is the theory for why?

Narrator / Host (12:26)

Well, it comes down to the strategy, to the way that these AI systems were built to win in these different games, which relies on a massive amount of engineering. Deep Blue, for example, was programmed essentially hand coded with all the rules of chess, with the millions and millions of possible chess positions they might encounter coming from all the world's best chess books. Or the same with Watson. It's programmed on all these encyclopedias, and then during the game, they're essentially just running algorithms to try and retrieve the possible right answer or the possible right move as fast as they can. And do you remember what were you thinking back in 1997 when Garry Kasparov gets beat by IBM's Deep Blue and there's all this excitement about what's going to happen next?

Yoshua Bengio (13:12)

Nothing much, because we knew that it was just brute force search, which is a classical AI technique that is very unlike human intelligence.

Narrator / Host (13:21)

But it turns out that all of this time that all the attention was being paid to Watson in Deep Blue, there were some AI researchers like Yoshua Bengio.

Yoshua Bengio (13:30)

That way that Deep Blue beat Casper up was nothing like how human chess players do it.

Narrator / Host (13:38)

And this small band of AI researcher outsiders who were essentially shaking their heads, saying that these AI systems, they are not doing what anyone would consider thinking, and they thought that they had a better way. Would you describe yourself as a bit of a contrarian?

Geoffrey Hinton (13:58)

I'm tempted to disagree with that, but I think you might be right.

Narrator / Host (14:03)

And one of them that would turn out to be the most consequential was a guy named Geoffrey Hinton. Well, let's just get into it. First off, can you just introduce yourself? You know, what's your name and what title do you go by these days?

Geoffrey Hinton (14:14)

My name is Geoffrey Hinton. I've been doing research on neural networks since 1972. That's a little over 50 years. And for a long time this was regarded as crazy. And more recently it's turned out it works much better than symbolic AI.

Gregory Warner (14:33)

This is the same Geoffrey Hinton who quit his job at Google in 2023, very publicly to warn the world about the existential risk of AI.

Narrator / Host (14:41)

Yes, it is. He is now a Nobel Prize winner for his work on artificial intelligence. Many call him the godfather of AI. But before he quit his job at Google, before he even had his job at Google, for much of Hinton's career, his ideas, his strategies, his approach was resoundingly rejected by almost all of his peers.

Geoffrey Hinton (15:05)

I had a very smart student, wanted to do graduate work with me, and one of the other professors in my department told him, oh, don't work with Hinton. That'll be the end of your career. It's a dead end.

Narrator / Host (15:16)

I'm sorry to laugh. I just know what happens at the end of the story makes it ridiculous. But what did that feel like at the time?

Geoffrey Hinton (15:23)

My view is you shouldn't give up on an idea that goes against the grain until you understand why it's wrong.

Keech Hagee (15:30)

So Hinton, he initially started as kind of ostracized by the AI community because he was working on the approach that most people thought was a dud.

Narrator / Host (15:42)

Unsurprisingly, Hinton's name and his backstory came up in almost every conversation that I had for the series, including with the author Karen Howe.

Keech Hagee (15:50)

He just felt very strongly about it, in part because he originally started studying AI, not because he wanted to recreate human intelligence, but actually because he wanted to understand human intelligence better. So he was interested in it from the perspective of if we successfully create intelligent systems and computers that will enable us to better understand our own intelligence.

Narrator / Host (16:13)

And one of the things that's so cool about Hinton is that he always studied artificial intelligence alongside brain science because he had this deep seated belief that the two were intrinsically linked.

Keech Hagee (16:25)

And so he was coming from a more neuroscience background, and he strongly felt that if we can create software that mimics the processing power of the brain, surely we will be able to get to some kind of intelligent system.

Gregory Warner (16:43)

So the idea is that the path to intelligence, real intelligence, to get the machine to think like a human brain, it needs to be structured like a brain. Like the same way we have a bunch of neurons in our brains all talking to each other. That's what they're going to kind of design for this computer.

Narrator / Host (16:58)

Yes. This is how you get the approach called neural networks or neural nets, which is basically that it's a AI system with all these different layers and layers of artificial neurons and they fire and they change in a way that kind of mimics or mirrors the way that neurons fire in our brains.

Geoffrey Hinton (17:16)

So my aim has always been to understand how the brain works. But in our attempts to understand how the brain works, we've developed this technology, which is amazing.

Yoshua Bengio (17:23)

I thought, wow, this is really cool. Why don't we take inspiration from human brains to figure out how to do AI?

Narrator / Host (17:32)

Yoshua Bengio is, like Hinton now, one of the most decorated and celebrated AI researchers in his field. He's won the Turing Award. He's actually the most cited living scientist on earth right now. But like Hinton, most of his career was full of rejection.

Yoshua Bengio (17:49)

My papers got rejected because they were about neural nets, and my students didn't want to work on neural nets because they were afraid they wouldn't get a job.

Narrator / Host (17:58)

And why were you so committed to this? Why not just follow the mainstream AI models?

Yoshua Bengio (18:04)

Some scientists, at least I, and many others I know, have an emotional relationship with ideas. You get really excited about something and you feel strongly that this is the path. If you want to be honest, you know, you can't be sure, but still you have this strong feeling. These emotions is what allowed us to go through the times when it was maybe difficult to work on these topics.

Narrator / Host (18:30)

Walk me through, like, the 80s into the 90s. What was it like to study this? Did it feel like it was fringe? Like, what language should we use to accurately describe what you were up to?

Geoffrey Hinton (18:42)

Fringe is quite good. There was a period even quite late on, Even in the 2000s, when people were saying things like, this paper's about neural networks. It shouldn't be submitted to a machine learning conference.

Narrator / Host (18:54)

Like, it's not even worth submitting.

Geoffrey Hinton (18:56)

We don't want that kind of stuff in machine learning. It's obvious nonsense, and machine learning shouldn't pay any attention to it.

Gregory Warner (19:02)

Here is what I'm not getting right. If AI, from its origins, from the beginning of the term in 1956, even earlier with Alan Turing, the whole idea of AI was to mimic human intelligence. The brain is our thinking organ, and this camp had a way to mimic the brain. So why was that idea? Why were they out in the cold?

Narrator / Host (19:25)

Part of the reason is almost a philosophical resistance to this idea among AI researchers, because from that 1956 summer program where that debate emerged between, you know, the symbolists who want to make expert systems and the connectionists who want to make these AI toddlers, AI babies, the expert system side just totally dominated, partly because the systems they made were just better at doing things that looked like intelligence.

Yoshua Bengio (19:57)

For decades, higher intelligence, as in, you know, what mathematicians do or physicists, or people who play chess and so on and win tournaments, that was considered the peak of Human intelligence, like an AI.

Narrator / Host (20:14)

System that can beat a chess master, an AI system that can, you know, stomp two nerds in jeopardy.

Gregory Warner (20:20)

That must be smart.

Narrator / Host (20:21)

That is what intelligence looks like to us. And Bengio and Hinton and their side, they're over there saying, no, no, no. Intelligence is a toddler. Intelligence is a four year old.

Yoshua Bengio (20:31)

Wait a second. We need to build the foundations firsts. And the foundations for human intelligence is the intelligence of a one year old. And you don't spoon feed a one year old with mathematical formulae. Right? You let him experience life, you show him things.

Narrator / Host (20:52)

But they also had a couple of very serious technical obstacles in their way as well.

Geoffrey Hinton (20:57)

For many, many years, most people claimed it was crazy, but for not completely unrealistic reasons.

Narrator / Host (21:04)

The biggest one was that for decades, critics of Hinton and this neural net approach, they would say, if your system is going to learn on its own, learn its own patterns, what are you going to do when it learns the wrong thing?

Geoffrey Hinton (21:19)

They said a big network of brain cells with random connection strengths in them will never learn to do anything interesting. If you try just tinkering with the connection strengths to make it behave better, you'll get stuck in what's called a local optimum.

Narrator / Host (21:32)

And the metaphor that Hinton uses for this problem is to imagine a hiker on a huge mountain range with a simple mission. Climb to the tallest peak. And the hiker has one rule, always go up. And this works fine all the way until they reach the top of a smaller mountain peak. And at that point, every direction that they can go is down.

Karen Howe (21:56)

Right?

Narrator / Host (21:57)

And so from the hiker's perspective, it thinks it's on top of the world, when in reality there's a much larger peak nearby.

Geoffrey Hinton (22:05)

It's like a mountain range where you get trapped on a peak and you can't get to the higher peaks because you have to go downhill to get to the higher peaks. And if you just try going uphill, you'll be trapped on this local peak and you'll never really get anywhere.

Gregory Warner (22:18)

So it's like the hiker has learned the terrain, it's figured out the mountain. It's actually been able to climb all this distance and. But it cannot figure out how to go back down the path in order to take the right trail to the even higher peak.

Narrator / Host (22:33)

Yes. To retrace its steps and to find a different solution, so to speak, to correct an error. And this was a huge problem. And one of the reasons that Hinton and Bengio are referred to as godfathers of AI, one of the reasons that they are legends in their field is because, despite all of the naysayers, they continued to go back to their labs. They kept doing their research, trying to solve problems like this, and finally they did. And it's a crazy story. They used this old, largely forgotten algorithmic system called back propagation. And they were able to give these neural nets a way to metaphorically retrace their steps, go back to where they started from and start climbing again. In a sense, the machine could now learn from its mistakes.

Gregory Warner (23:23)

Got it. So it's like back propagation is like a math way of saying, hey, go back and correct your error.

Narrator / Host (23:28)

Like, the network can now escape the small hills, find its way to scale up to the real mountains, so to speak. And this would become revolutionary in theory. But the trouble was, when they discovered it, they were still running into two other very persistent problems. One of them was that they just needed an insane amount of computing power.

Geoffrey Hinton (23:53)

One aspect is computers were small and slow relative to what they are now. If you tried using neural networks, you couldn't get them to do much.

Narrator / Host (24:01)

Right. You can imagine a digital brain firing with digital neurons, trying to not only learn patterns, but go back and learn from its mistakes like that. That's going to take a lot more computing power than you could get from, like, a 1980s or 1990s IBM. Right, right. And you also need just an insane amount of data for this AI system to be combing through and learning inside of and making mistakes and learning again. And therefore, for years, they continued to live on the fringes of AI research, to watch, as you know, IBM's Watson and deep Blue get all the money, get all the attention. And then came the year 2012, when they finally had their chance to completely flip this dynamic.

Jasmine Sun (24:49)

A huge breakthrough came at Geoff Hinton's lab in 2012, and that came in.

Narrator / Host (24:55)

The form of a game called imagenet, the imagenet Challenge. I talked about this with the writer Jasmine sun, who's working on a book about AI right now, which is this.

Unknown AI Expert / Commentator (25:06)

Grand challenge with a huge data set.

Narrator / Host (25:08)

Of images, and with the Wall Street.

Jasmine Sun (25:11)

Journal's Kichegi, this contest that had been running for many years.

Narrator / Host (25:15)

They explained to me that this was a simple game between the world's best AI systems in an AI versus AI challenge of essentially name that picture, look.

Jasmine Sun (25:26)

At a bunch of images, and have the computer describe what was in the images. That's a cat, that's a dog, et.

Unknown AI Expert / Commentator (25:32)

Cetera, tons and tons of photos. And can someone build a system that can, like, label and classify them as accurately as possible.

Narrator / Host (25:39)

It turns out that this is something that is really easy for humans to do, even children, but actually very difficult for machines.

Geoffrey Hinton (25:47)

Many of the things that we do effortlessly, like recognize objects or recognize the words when somebody's talking, are actually very difficult computational tasks that require huge amounts of computation. So even the people doing symbolic AI understood that things that appear very difficult, like playing chess, are actually much easier than things that appear very simple to us, and that a three year old child can do, like recognizing objects.

Narrator / Host (26:13)

So it was almost easier to make a chess champion than to make a two year old who could tell the difference between a banana and a ball.

Geoffrey Hinton (26:22)

Yes.

Narrator / Host (26:23)

Up to this point, even the best AI systems that entered into this competition, they were still making a lot of mistakes. They were often mislabeling one out of every five or one out of every four of the images that they tried to categorize. And that is because they were all expert systems, meaning that they relied a lot on hand encoding for all the insane amounts of patterns and textures and colors and shapes that they would need to know to identify an image.

Gregory Warner (26:52)

This is almost like trying to teach an AI that's just looking at pixels to tell the difference between, I don't know, a butterfly and a moth.

Narrator / Host (26:59)

Right. How would you mathematically give a AI system the patterns and the textures that it needs to understand the difference between, you know, a seal and a sea lion, you know, a teacup in a.

Gregory Warner (27:12)

Coffee mug, or even like distinguishing cat and dog? The classic one, Right. That's. That's even hard because both are pretty similar. I mean, if the AI is looking for shapes, it's looking for like two triangles. That would be the ears, a kind of blob. That's the face, fur, I guess you look for whiskers in the cat.

Narrator / Host (27:29)

You can see how trying to embed the rules, put the codes and the shapes and the math into a machine like this would be really hard. So in 2012, Hinton and two of his grad students, one of them, by the way, is Ilya Suskover, who would go on to be a founder at OpenAI and would help create ChatGPT. They joined this competition with their totally different approach, where they're going to let their AI learn and find patterns totally on its own. And one of the reasons that they're so confident that they can win is that the data problem that they had for years, this had largely been solved by the era of big data on.

Keech Hagee (28:11)

The Internet in 2012. What happens is during all this time, the connectionists, which have been Sort of in academic exile, have continued to make progress on their research. And there are a couple things that happen that assist them. One is that the Internet suddenly makes the aggregation of data far cheaper. And when you're trying to build data driven machine learning systems, you need a lot of data. And before collecting it from the analog world was just not as practical.

Narrator / Host (28:44)

And even in the early age of the Internet, I imagine dial up was a little bit tough, you know.

Keech Hagee (28:49)

Yeah. But the second thing that happens is that computer chips become a lot more powerful.

Narrator / Host (28:54)

And on top of that, they got an assist from the video gaming world. Can you tell me what is a gpu and how is it that video game players ended up being the unsung heroes here?

Unknown AI Expert / Commentator (29:05)

Yeah, a GPU is a piece of hardware, graphics processing unit that was originally used in gaming.

Narrator / Host (29:11)

Because it turns out that after years and years of many lonely late nights, as the stereotype goes, where all these gamers are playing all these different video games and they want really sweet graphics, so they want really smooth play. An industry of chip makers emerged that made these GPUs. The most important one being this little company called Nvidia.

Unknown AI Expert / Commentator (29:33)

Because when you build video games, you just need huge graphics. Like they have to move really fast, be really smooth. It's just like gaming just happens to require an insane amount of computing power compared to typing on word or whatever most of us are doing most of the time.

Narrator / Host (29:47)

And Hinton and his colleagues, they realized that those GPUs that make sweet graphics and games also packed a huge punch when you're trying to run a digital brain neural net that's trying to learn its own patterns and learn from its own mistakes.

Gregory Warner (30:03)

So now Hinton and his team, they've got their math, they got the back propagation algorithm, they got the data from the Internet and they've got the compute, thanks to those video gamer GPUs. And so now it's onto the game.

Narrator / Host (30:20)

And it's a real David and Goliath situation here because remember, this is Hinton and two grad students, they're at the University of Toronto. Their university doesn't even fund their experiment. They are going up against way bigger AI labs in China at universities like mit. But when the results come in, this.

Unknown AI Expert / Commentator (30:39)

Just like blows everything else out of the water with how suddenly accurate it is.

Jasmine Sun (30:44)

Jeff Hinton's technology was able to do this and win this contest, have the best performance that any tech had ever had.

Geoffrey Hinton (30:52)

They were just amazed.

Narrator / Host (30:54)

Hinton and his grad students, they don't just win, but they cut the error rates nearly in half.

Keech Hagee (30:59)

And people suddenly realize that maybe the connectionists were on to something all along.

Geoffrey Hinton (31:08)

And something happened, which doesn't often happen in science, which was that some of the best researchers in the field who had been vigorous opponents of neural nets, saying that stuff will never work on real images, they pretty much immediately switched their opinion. They said, this is amazing. We're going to start doing that.

Narrator / Host (31:25)

And even though there are no cameras, there's no press around like there was for Deep Blue in that chess match or Watson on Jeopardy. This is the AI that actually makes the leap from the world of games into the world beyond it.

Unknown AI Expert / Commentator (31:40)

I've talked to AI researchers who were sort of like, I remember being on hacker news in 2012, seeing Alexnet and then going, holy crap, if this can work for images, there's no limit to what an AI system can't do. And so that was when I realized I had to get into the field.

Narrator / Host (31:56)

People like Hinton and Bengio, after years in the cold, they're now essentially proved right. They're suddenly winning awards, being celebrated. There's a bidding war that breaks out. All the top tech companies are trying to hire Hinton and his grad students. Eventually, they end up at Google. Hinton, in his 60s, suddenly becomes a multimillionaire, something he told me he never expected would happen. What was that like?

Geoffrey Hinton (32:18)

Um, that was weird. We had no idea how valuable what we'd done was.

Gregory Warner (32:24)

Okay, so now finally, Hinton and his connectionists, they are no longer on the fringes. Their approach goes from being rejected to.

Narrator / Host (32:33)

Being embraced by Google. Yeah, yeah.

Gregory Warner (32:36)

And they themselves become wealthy Google employees. So what does Google actually do with this technology? How is it useful to them?

Narrator / Host (32:44)

Yeah, it was immediately useful for a number of things. Obviously Google image search or YouTube video recommendations. But it also is this seismic shift in the strategy for all of these other categories. So if you were working in facial recognition up until this point, you were using those expert systems, you were hand coding all these different complex algorithms. But now, after imagenet, you make the switch. You are letting your AI learn from patterns on its own. This is also true in the world of language translation and all these other different categories of automation. And yet it comes with a trade off, because, remember we talked about this before, if you go with the model of the AI toddler over the model of the AI expert, a trade off you have to make is mystery, essentially an understanding of exactly how it works. And for you, is this just something that you've always accepted that if you're going to Make AI in this way, then you just have to accept that you will never fully understand how they know, what they know, how they work.

Yoshua Bengio (33:55)

Well, we'd like to know, but the reality is that if you let go of that requirement, then you can get much more powerful systems.

Narrator / Host (34:05)

And for researchers like Hinton or Bengio, that is the trade off that will give you intelligence.

Jasmine Sun (34:11)

And one of the things that really fascinated me was that these early AI sort of neural net research was in many cases an attempt to understand the human brain.

Narrator / Host (34:23)

This came up when I was talking to Keech Hagee. She was saying that not understanding exactly how the AI is working, that isn't a flaw. It's actually better understood as a feature, not a bug.

Karen Howe (34:36)

Right.

Jasmine Sun (34:37)

They weren't trying to make some like robot that would do stuff for you. They were trying to actually understand ourselves. And we don't know how the human mind works. So it's a mirror in some ways.

Narrator / Host (34:51)

Right. And the idea is that if you're trying to create something that is truly intelligent, discovering that its interworkings are a mystery in some ways is a signal of success, that you're making progress.

Jasmine Sun (35:05)

Yes. Our own heads are a black box to us.

Gregory Warner (35:12)

After the short break, one man stares into the machine's black box and thinks he sees a way to build super intelligence.

Sponsor / Advertisement Voice (35:35)

The last invention is sponsored by Ground News. Ground News is one of the most helpful tools that I use to avoid the echo chambers and media bias online, especially when it comes to shining a light on our blind spots. So whether you're politically on the left or the right or somewhere in the center, the blind spot feature from Ground News highlights the stories that tend to be disproportionately covered by one side or the other. As an example, take these two stories about President Donald Trump. One which had low coverage among left leaning outlets.

Karen Howe (36:05)

It's a very important relationship we're going to get along good with.

Sponsor / Advertisement Voice (36:07)

China reported that Trump says US will accept 600,000 Chinese students as part of a trade deal.

Karen Howe (36:14)

I hear so many stories about we're.

Narrator / Host (36:16)

Not going to allow their students.

Karen Howe (36:17)

We're going to allow. It's very important.

Narrator / Host (36:19)

600,000 students.

Sponsor / Advertisement Voice (36:20)

And another largely uncovered by right leaning outlets.

Keech Hagee (36:23)

Trump's social media company is using crypto.com's Trump family.

Sponsor / Advertisement Voice (36:28)

Crypto empire expands with crypto.com partnership.

Geoffrey Hinton (36:32)

That's our transactional Trump family.

Gregory Warner (36:34)

Make some money when you can by.

Sponsor / Advertisement Voice (36:35)

Seeing which stories are amplified or ignored depending on the outlet. Ground News helps you step outside the filter bubbles that shape most people's news diets. Giving you a fuller picture of what's actually happening. I really think that if you like this podcast, you're going to like their mission. Go to groundnews.com invent to get 40% off the same unlimited access vantage plan that we use. You can even sign up to stay up to date with the biases in your coverage proactively with the weekly blind spot report delivered directly to your inbox. This is a great way to support them and the work that they do. Because Ground News is a subscriber supported platform. We appreciate what they're up to, we appreciate their support for this podcast. So go check them out and make sure to use our link ground news.com invent so they know we sent you.

Geoffrey Hinton (37:23)

Foreign.

Sponsor / Advertisement Voice (37:27)

This episode of the Last Invention is brought to you by fire, the foundation for individual rights and expression. There's a pattern that you can trace throughout history. In ancient Athens, Socrates was put to death for asking tough questions of the powerful. Centuries later, monarchs banned and burned books they considered dangerous. And in the last century, authoritarian governments shut down newspapers, censored broadcasts, even jailed their critics. The struggle was always the same. Who gets to decide what people can know. Today, that struggle is playing out in a new arena, and the risk now is subtler search results that quietly vanish, recommendation engines that steer us towards safe and comfortable answers, and AI filters that can suppress ideas before we ever even see them. That's where FIRE comes in. FIRE has spent decades defending free inquiry on our campuses, in the courts, and in our culture. And now, through a $1 million grant program in collaboration with the Cosmos Institute, they are supporting projects that keep free thought alive in the era of AI. Join us today@the fire.org the last invention. By supporting fire, you're protecting the future of free inquiry in America and ensuring that tomorrow's most important questions can still be asked. Once again, visit thefire.org thelastinvention and thanks.

Gregory Warner (38:53)

So the Imagenet victory, it was kind of a jailbreak for AI in general. AI was very quickly out in the world. It was in our phones, it was in our browsers. But these are just products, right?

Narrator / Host (39:06)

Yes.

Gregory Warner (39:07)

This is not Turing's vision of a thinking machine that might outthink humans. So where do we get to the next step of AGI?

Narrator / Host (39:17)

Yeah, the way that a lot of different tech insiders have explained it to me is that when Google looked at Hinton and his two grad students and the amazing achievement of ImageNet, they saw a way to make money. And they saw a way to increase the efficiency and usefulness of their products. But it would take another contrarian, another game for the industry to take the next step to seeing not just a way to make money, or not just a way to increase their market share in the technology field, but to see a way to make a digital supermind that might change the world forever. And that gamer, that guy is named Demis Hassabis.

Karen Howe (40:01)

More and more people are finally realizing leaders of companies, what I've always known for 30 plus years now, which is that AGI is the most important technology probably that's ever going to be invented. So to me, it's been obvious for many, many years that AI, if it was possible, and it seems that it is, it would transform everything.

Narrator / Host (40:24)

So who is Demis Hassabis and how is it that he comes to be this bridge between Hinton's work and this moment that we're in right now, where people think that we are seriously on the cusp of AI changing everything, for better or for worse?

Unknown AI Expert / Commentator (40:38)

So Demis was a child prodigy. He was a champion games player.

Narrator / Host (40:44)

Demis Hassabis, he grows up in England, and by all accounts, he was a child genius.

Unknown AI Expert / Commentator (40:49)

By age 4, he was playing chess competitively against adults.

Narrator / Host (40:53)

By age 13, he's already representing England in these international chess championships. But he's not just good at chess, he's good at almost every single game that he plays. And he starts entering into these Pentomind championships. Are you familiar with Pentomind?

Gregory Warner (41:09)

It's like the Olympics of mind sports, basically, right?

Karen Howe (41:12)

Yeah.

Narrator / Host (41:12)

Some people describe it as like a decathlon of the mind. It's been called the biggest gathering of anorax ever, experts in chess, backgammon, bridge, and other more obscure mind games. You play all these games ranging from chess to go to bridge to poker, all at the same time. And Demis, he started entering into these world championship matches the Pentamind world champion, and of course he dominates.

Unknown AI Expert / Commentator (41:40)

I think it's quite notable that he was a Pentamind champion, that is that he's not just good at one game, but has some sort of like flexible, cognitive, metacognitive skill set that allows him to succeed across a whole range of different strategic activities, different environments, different rule sets.

Narrator / Host (41:56)

But all this also leaves him with a question.

Unknown AI Expert / Commentator (42:00)

I think what this makes him interested in is like, why is he so good at games? What makes some people good at games and some people less good at games?

Narrator / Host (42:07)

What is happening in my mind? What is my own intelligence? And how could I recreate a general intelligence like my own inside of a machine?

Unknown AI Expert / Commentator (42:20)

And what would it take to also build a computer system that could do that as well?

Narrator / Host (42:24)

And so this question inspires him to go to Cambridge University, where he studies computer science, and then to open his own gaming studio, where he's not only designing and building his own video games, but he's using the latest artificial intelligence technology to do so. But at the time, he's just really unimpressed with the quality of, of those AI systems.

Unknown AI Expert / Commentator (42:47)

These early forms of AI that he was working with in the early 2000s, the 90s, they just weren't advanced enough to do the kinds of things that he wanted. And he felt like the computers just weren't smart enough yet. So he closes down his game studio in his early 20s and says, no, I gotta get a PhD in neuroscience. Like, actually, I need to understand the brain better.

Narrator / Host (43:06)

And in classic Demis Hassabis form, he doesn't just get a PhD in neuroscience, but his thesis paper ended up being named among the top 10 scientific breakthroughs by Science magazine back in 2007.

Gregory Warner (43:20)

All right, this guy's pretty unstoppable, right?

Narrator / Host (43:23)

I think he's earned the name genius. But anyway, it's while he's studying to get his PhD that he meets another student, a guy named Shane Leg, who is also really into this idea of artificial intelligence and of building a true thinking machine. And the two of them decide to found this company together called DeepMind.

Unknown AI Expert / Commentator (43:43)

They're like, we're creating a research lab. We are going to pursue this crazy idea that nobody takes seriously, and we're going to get a bunch of researchers who believe in this vision to do it with us.

Narrator / Host (43:55)

And so in 2010, this boy genius and his co founder, armed with their education and their big dream, they head off to the world capital of ambitious tech startups, Silicon Valley. And they start going around telling these different investors that they're not just going to make a new tech product, that they want to make real AGI that will transform the economy, transform healthcare, that will supercharge humanity into this age of abundance. But not even Demis Hassabis can overcome the fact that at this time, almost no one in Silicon Valley thinks that AGI is going to be possible anytime in the near future.

Gregory Warner (44:42)

Well, not only that, it's like Kevin Roos from the Times was telling you, it was considered sort of embarrassing for a company to talk about AGI, right?

Narrator / Host (44:49)

And so at first, according to Demis Asabas, they had a really hard time hiring people. They had a hard time scraping together enough money to really get their company off the ground until finally they landed their first big investor, Peter Thiel. You know, one of the standard ways people think about technology is that if it happens, it's great. If it doesn't happen, not a big deal. I think little could be further from the truth. Our entire civilization, our entire culture is predicated on accelerating technological change. Peter Thiel, as you know, he has embraced the nickname the contrarian of Silicon.

Gregory Warner (45:25)

Valley for some ways that have made.

Narrator / Host (45:26)

Him quite controversial, but, yes, controversial in some eyes, beloved in others. But what's so interesting about his initial investment, and I believe it's been reported that it was about $2.5 million, is that Thiel doesn't necessarily buy into the fact that AGI is going to be utterly transformative. The thing that he's most interested in is funding and backing tech projects that are promising that technology can actually deliver a far better world. And so his investment, it's less a deep seated belief that Dimmus is going to pull this off, and it's more like a vote of confidence in a kind of tech entrepreneur dream.

Gregory Warner (46:13)

Is it like all the money was going to dating apps and attention economy, kind of grabbing sort of projects, and he's like, no, these folks want to build a supermind, let's give them some money, right?

Narrator / Host (46:23)

Like, okay, maybe they can build a super mine, maybe they can't build a supermind, but this is the direction that Silicon Valley should be going in.

Gregory Warner (46:31)

So Peter Thiel's in, they've got some money. What do they do with it?

Narrator / Host (46:35)

Well, right away they start trying to build these different AI models. They start doing AI experiments, but they don't really make any, you know, breakthroughs to speak of until the Imagenet competition in 2012. When they see what Hinton and his grad students were able to do, they look at that and they think, there it is. There is our path to making a true world changing AGI. And so, in true gamer fashion, they decide to build a groundbreaking AI that plays Atari.

Jasmine Sun (47:13)

It basically showed this AI agent teaching itself in real time how to play a vintage Atari game, and then became better at it than any human in the world.

Karen Howe (47:26)

So we started with probably the most iconic of the game consoles, the Atari 2600 from the 80s.

Narrator / Host (47:33)

Their idea was to take things even further than Hinton did. And with this AI system, they weren't going to give it any instructions, any data, any information at all.

Karen Howe (47:47)

So this is literally the first time the machine has ever seen this data stream, this pixel data stream. So it has no idea it's controlling the green Rocket at the bottom of the screen has no idea how to get points, no idea how it loses lives.

Narrator / Host (47:59)

Demis Asabis later, he did a presentation where he walked people through how this AI played the game Space Invaders. This is a game where you are a fighter plane and you're trying to defeat a bunch of aliens in spaceships that are shooting at you.

Karen Howe (48:13)

And you'll see it loses its three lives almost immediately. So it's just playing randomly at the moment. Then after overnight trainings on a single GPU machine on our servers, it's just playing the game some more. You come back in the morning, and now it's better than any human can play the game. So every single bullet hits something, it can't be killed anymore by the spectrum. Space Invaders. It's worked out that the mothership at the top of the screen going across now is worth the most points. It does these unbelievably accurate shots to get those points.

Narrator / Host (48:41)

In just a few hours, this AI learned on its own how to execute every single move in such an efficient way that it not only earned a perfect score, but it seemed to understand what was going to happen in the game before it happened.

Karen Howe (48:57)

It's built up such an accurate model of this world that it's in that if you watch the last space invader, they get faster as there's less of them. Watch the last bullet, it sort of predictably fires where it thinks the space invader will be in a few seconds time. It's perfectly modeled this very complex data stream. Now, of course, these are just games, but this could be anything. This could be climate data, this could be economics data, stock market data, anything that has temporal sequences of data, which is most things these days.

Narrator / Host (49:30)

Silly as it sounds, Atari ends up being the trigger that fires the opening shot in the AI race.

Jasmine Sun (49:40)

That was a huge breakthrough. And a video of that kind of was circulated on private planes of billionaires. And it was what prompted Google to very quickly buy up DeepMind.

Narrator / Host (49:55)

It starts this bidding war that eventually leads to to Google purchasing DeepMind and hiring Demis, just as they had hinted before. Only this time they're no longer saying, we want you to work on our products. They are putting their seal of approval, they're putting their money and their resources behind this previously wild pie in the sky idea of creating a true digital supermind.

Jasmine Sun (50:22)

And when Google sucked this really promising technology up inside the Borg of Google, that radicalized Elon Musk when that happened and made him convinced there had to be some alternative to sort of counter it.

Narrator / Host (50:38)

This is what would lead Elon Musk to do everything in his power first to stop Dimmus and his creation.

Geoffrey Hinton (50:46)

With artificial intelligence, we are summoning the demon. Mark my words, AI is far more dangerous than nukes.

Narrator / Host (50:53)

I think that's the single biggest existential.

Geoffrey Hinton (50:56)

Crisis that we face.

Narrator / Host (50:57)

And then ultimately to beat him to it.

Sponsor / Advertisement Voice (51:11)

Next time on the Last Invention, how the technologists who were most concerned about the risks of AGI began one after another to believe that the only way for it to be safe was to make sure that they were the ones who built it and built it fast. The Last Invention is produced by Longview Home for the curious and open minded. To support our work go to longviewinvestigations.com Special thanks this episode to Keech Hagee, Jasmine sun and Karen Howe. Links to their work can be found in our show notes. Thanks for listening and we'll see you soon. This episode is sponsored by Ground News, the app that helps you spot media bias and see a broader picture of the news shaping our world. World get 40% off their vantage plan at Ground News Invent this episode is sponsored by Fire Defending Free Thought in the Age of AI. You can learn more at thefire.org thelastinvention.