WEBVTT - Selects: A List Of Games You Would Surely Lose to a Computer 0:00:00.760 --> 0:00:03.840 Hi everyone, it's Josh and for this week's select, I've 0:00:03.920 --> 0:00:08.479 chosen our twenty eighteen episode Some Games You would Surely 0:00:08.560 --> 0:00:13.160 Lose to a Computer. It's a philosophical discussion about AI 0:00:13.280 --> 0:00:16.880 that's disguised as an episode on computer games. Honestly, we 0:00:16.920 --> 0:00:19.040 didn't plan it to be like that. It just turned 0:00:19.040 --> 0:00:21.960 out that way. We're pretty happy that it did. And 0:00:22.000 --> 0:00:24.480 in light of the recent advances with machine learning like 0:00:24.800 --> 0:00:27.520 chat GPT, a few of the things we say seemed 0:00:27.680 --> 0:00:30.479 naively quaint now. Plus it has a doll up of 0:00:30.480 --> 0:00:34.080 our tech stuff colleague Jonathan stricklanet and so that's a bonus. 0:00:34.320 --> 0:00:36.720 I hope you enjoy. 0:00:39.479 --> 0:00:45.240 Welcome to Stuff you should know, a production of iHeartRadio. 0:00:49.240 --> 0:00:52.440 Hey, and welcome to the podcast. I'm Josh Clark. There's 0:00:52.520 --> 0:00:56.840 Charles w Chuck Bryant, there's Jerry over there. I'm just 0:00:56.880 --> 0:00:58.960 gonna come out and tell everybody making fun of me 0:00:59.080 --> 0:01:04.559 for some weird reasons, vaguely weird ways. But I'm all right, 0:01:05.040 --> 0:01:07.840 So check out his story for you. Okay, I'm going 0:01:07.920 --> 0:01:11.880 to take us back to the seventeen seventies and they'll 0:01:12.080 --> 0:01:17.520 swing in town of Vienna, not Virginia, not Vanna Georgia, 0:01:18.120 --> 0:01:20.800 which you know, that's how they pronounce it, right, Yanna, 0:01:21.280 --> 0:01:24.440 Vienna sausages, right, Vienna Austria. 0:01:24.480 --> 0:01:25.240 Have you ever been there? 0:01:26.080 --> 0:01:30.000 Vienna, Austria. No, been to Brussels. That was pretty close. 0:01:30.720 --> 0:01:32.520 Vienna's lovely, I'm sure. 0:01:32.880 --> 0:01:34.560 I think it's a lot like Brussels. 0:01:35.319 --> 0:01:38.800 Very clean, lovely town. I just remember it being very clean. 0:01:39.240 --> 0:01:44.040 Yeah, very clean, gorgeous architecture, weird little angled side streets. 0:01:44.040 --> 0:01:48.600 They're very narrow, very pretty town. So we're in Vienna 0:01:48.680 --> 0:01:52.840 and there is a dude skulking about going to the 0:01:52.960 --> 0:01:58.680 royal Palace in Vienna. His name is Wolfgang von Kempelin, 0:01:59.600 --> 0:02:02.560 and he's he's an inventor, he's an engineer. He's a 0:02:02.600 --> 0:02:05.320 pretty sharp dude. And he's got with him what would 0:02:05.320 --> 0:02:08.480 come to be known as the Turk, but he called 0:02:08.520 --> 0:02:13.480 it the mechanical Turk or the automaton chess player, and 0:02:13.480 --> 0:02:17.560 that's what it was. It was a wooden figure that 0:02:17.760 --> 0:02:21.959 moved mechanically, seated at a cabinet, and on top of 0:02:22.000 --> 0:02:25.360 the cabinet was a chessboard. And when he brought it 0:02:25.400 --> 0:02:30.000 out to show to the royal court, he would it 0:02:30.040 --> 0:02:33.919 was cool kind of but nothing they hadn't seen before, 0:02:34.000 --> 0:02:38.000 because automata was kind of a hip thing by then. 0:02:38.760 --> 0:02:45.400 Yeah, people loved building these engineering these automata machines to 0:02:45.440 --> 0:02:48.359 do various things. And people are just knocked out by 0:02:48.480 --> 0:02:51.160 the fact that, you know, you hide these gears and 0:02:51.240 --> 0:02:54.960 levers behind wood or a cloth, and it looks as 0:02:54.960 --> 0:02:57.640 though there's a real well not real, but you know. 0:02:57.639 --> 0:02:59.919 What I mean that it's like a real machine. 0:03:00.600 --> 0:03:02.880 Yeah, but not. They weren't fooled anything like is that 0:03:02.960 --> 0:03:06.760 a real man? It was, but it was for their time. 0:03:06.880 --> 0:03:09.920 It was so advanced looking that it's like us seeing 0:03:10.639 --> 0:03:12.720 x Makina in the movie theater. 0:03:12.960 --> 0:03:15.760 Sure does that make sense? Yeah, no, it does make sense. 0:03:15.800 --> 0:03:19.400 But imagine seeing like X makinan being like I've seen 0:03:19.440 --> 0:03:22.160 this before. This isn't anything special, okay. 0:03:22.440 --> 0:03:24.520 Yeah, And this thing, to be clear, looked like a 0:03:25.520 --> 0:03:30.799 is it Zoltar or Zultan from Big Zoltar Zultan? 0:03:31.360 --> 0:03:32.680 I don't know. It's one of those two. 0:03:32.720 --> 0:03:34.560 One of those two. Like this, this guy's wearing a 0:03:34.600 --> 0:03:39.440 turban and it's in a glass case like a bust like, 0:03:39.480 --> 0:03:41.040 you know, like a chest up thing. 0:03:41.160 --> 0:03:44.080 Yeah, he's seated at this cabinet, so there's no need 0:03:44.120 --> 0:03:45.520 for legs or anything like that. 0:03:45.680 --> 0:03:45.960 Yeah. 0:03:46.040 --> 0:03:49.320 But the thing, this is what was amazing about the Turk. 0:03:50.240 --> 0:03:53.160 He could play chess, and he could play chess really well. 0:03:54.000 --> 0:03:57.240 So yeah, he was like an automaton and he moved 0:03:57.280 --> 0:04:00.880 all herky jerky or whatever. But he could play you 0:04:00.960 --> 0:04:07.120 in chess, which was a huge, huge advance at the time. Like, 0:04:07.160 --> 0:04:10.240 this is something that wouldn't come up again until the 0:04:10.320 --> 0:04:13.720 nineteen nineties, more than two hundred years later. This thing, 0:04:13.960 --> 0:04:17.080 this automaton could play a human being in chess and 0:04:17.200 --> 0:04:17.839 beat them. 0:04:18.120 --> 0:04:20.520 Well. Yeah, and it looked like when the game started, 0:04:20.560 --> 0:04:22.960 it would look down at the chessboard and like cock 0:04:23.080 --> 0:04:25.240 his head, like what should my first move be? 0:04:25.520 --> 0:04:25.760 Right? 0:04:26.040 --> 0:04:28.520 And if people I love this part. If people tried 0:04:28.560 --> 0:04:32.080 to cheat. Apparently Napoleon tried to cheat this thing because 0:04:32.120 --> 0:04:35.479 this guy he debuted at the VI and East Court. 0:04:35.960 --> 0:04:37.880 But then it, you know, it went on a world tour. 0:04:38.200 --> 0:04:41.000 Yeah, and he was even it was taken over by 0:04:41.040 --> 0:04:43.080 a successor to the guy who toured with it. 0:04:43.120 --> 0:04:44.920 Even further, people went nuts for this stuff. 0:04:44.920 --> 0:04:46.839 They did. They loved it because they were like, this 0:04:46.920 --> 0:04:49.840 is crazy, I can't believe what I'm seeing. Most people, though, 0:04:49.880 --> 0:04:51.640 were not taken in by it, they're like, there's some 0:04:51.720 --> 0:04:54.880 trick here. Sure, but von Kempelin and the guy who 0:04:54.920 --> 0:04:58.200 came after him, I don't remember his name, they would 0:04:58.240 --> 0:05:00.960 demonstrate you could open this happened and you could see 0:05:01.000 --> 0:05:04.960 all the workings of the mechanical turk inside. 0:05:05.600 --> 0:05:08.280 Right, So what I was saying is if this thing's 0:05:08.560 --> 0:05:13.400 since a cheater like Napoleon supposedly did it, would you 0:05:13.440 --> 0:05:15.880 know Napoleon would move a piece out of turner illegally 0:05:15.960 --> 0:05:19.040 or something. This dude, the turk Turk one eighty two 0:05:19.080 --> 0:05:21.320 would pick up the chess piece, move it back as 0:05:21.400 --> 0:05:23.760 if to say like, no, no, Napoleon, let' see what 0:05:23.800 --> 0:05:26.120 you're doing. And then if the person attempted to move 0:05:26.120 --> 0:05:28.039 it again, I don't know how many times on you 0:05:28.080 --> 0:05:32.039 two or three times, eventually it would just go ah 0:05:32.080 --> 0:05:34.800 and wipe his hand across the board and knock off 0:05:34.800 --> 0:05:37.839 all the pieces. Game over, right, Which is pretty great. Yeah, 0:05:37.880 --> 0:05:39.520 it's a nice little feature. 0:05:39.440 --> 0:05:43.240 Yeah it is. But it even showed even more that 0:05:43.320 --> 0:05:47.400 this thing was thinking for itself. Yeah, that's the key here, right. Sure, 0:05:47.520 --> 0:05:50.760 chess had been for a very long time viewed as 0:05:51.760 --> 0:05:55.400 only something that a human would be capable of, because 0:05:55.440 --> 0:05:58.000 it took a human intellect, and there was actually a 0:05:58.040 --> 0:06:03.760 guy in English engineer, I think he was a mechanical engineer. 0:06:03.960 --> 0:06:08.240 His name was Robert Willis. He said that chess was 0:06:08.279 --> 0:06:12.159 in the province of intellect alone. So the idea that 0:06:12.200 --> 0:06:16.240 there was this automaton playing chess blew people away. But 0:06:16.480 --> 0:06:19.440 again people figured out, like, okay, there's something going on here. 0:06:19.680 --> 0:06:23.960 We think that von Kempelin is controlling this thing remotely somehow, 0:06:24.279 --> 0:06:27.279 maybe using magnets or whatever. Other people hit upon the 0:06:27.320 --> 0:06:30.960 idea that there was a small person inside the cabinet 0:06:31.240 --> 0:06:34.480 who would hide when the cab when the workings were shown, 0:06:34.520 --> 0:06:37.120 when the cabinet was open to show the workings, and 0:06:37.160 --> 0:06:39.320 then when the cabinet was closed again and the mechanical 0:06:39.320 --> 0:06:42.000 turk started playing, the person that crawled back out and 0:06:42.160 --> 0:06:44.800 was actually controlling it. This seems to be the case 0:06:45.080 --> 0:06:47.960 that there was a person controlling it, but the idea 0:06:48.120 --> 0:06:53.120 that it was it was a machine that could think 0:06:53.520 --> 0:06:57.839 and beat humans in chess had like kind of unsettling implications. 0:06:57.960 --> 0:07:04.120 Yeah, this author Philip thick Ness, great name, British author 0:07:04.720 --> 0:07:09.359 for sure, Philip Thickness. Yeah, he said, and you know, 0:07:09.400 --> 0:07:12.720 people like you said, all those more complicated explanations. In 0:07:12.720 --> 0:07:16.200 this article you sent, Astuteley points out that he followed 0:07:16.200 --> 0:07:19.040 Occam's razor and basically said, he's got a little kid 0:07:19.040 --> 0:07:21.640 in there. He's got a little a little Bobby Fisher 0:07:21.640 --> 0:07:25.040 in there that's really good at chess, and that's what's 0:07:25.040 --> 0:07:28.600 going on. And other people speculated that other, you know, 0:07:28.720 --> 0:07:31.600 little people might be in there, just adults who would 0:07:31.600 --> 0:07:34.040 fit in there. But then you know, there's the explanation 0:07:34.080 --> 0:07:36.640 that he would open it up and shine a candle 0:07:36.640 --> 0:07:41.160 around and say, you know, nothing to see here everyone, 0:07:42.160 --> 0:07:46.040 So what should we reveal the real deal? 0:07:46.520 --> 0:07:48.480 Sure, I think I did already. 0:07:49.000 --> 0:07:51.160 Well I don't think you spelled it out as. 0:07:51.160 --> 0:07:52.160 Oh, we'll spell it out. 0:07:52.320 --> 0:07:54.920 There was a little person in there, Yeah, not just 0:07:54.960 --> 0:07:58.000 one little person, but they would travel around and recruit people. 0:07:58.040 --> 0:08:00.200 I guess people would get tired of being in there. 0:08:00.320 --> 0:08:02.480 Or they'd forget about them and they'd starve and have 0:08:02.560 --> 0:08:03.440 to replace them. 0:08:03.640 --> 0:08:05.680 But it really was a trick. There was a little 0:08:05.720 --> 0:08:08.120 person in there. They did the same thing as like 0:08:08.160 --> 0:08:10.600 the magic acts, you know, when they saw a person 0:08:10.640 --> 0:08:13.880 in half. It's that the lady just gets into a 0:08:13.880 --> 0:08:16.520 tiny little ball right in one section of that box. 0:08:16.680 --> 0:08:18.640 But my thing is this, like, this is not a 0:08:18.720 --> 0:08:20.880 satisfying explanation to me, Chuck. 0:08:20.760 --> 0:08:21.560 I think it's great. 0:08:21.840 --> 0:08:25.080 How did the person keep up with the board above? 0:08:25.800 --> 0:08:28.240 Well, I mean some I don't know if they ever 0:08:28.360 --> 0:08:29.880 proved exactly how it was going. 0:08:30.040 --> 0:08:30.680 That's what I'm saying. 0:08:31.080 --> 0:08:34.960 Oh, okay, whether or not I think they that the 0:08:35.080 --> 0:08:39.200 zoltar or I'm sorry, the turk was just hollowed out 0:08:39.320 --> 0:08:42.240 and you you would just put your arms through the 0:08:42.440 --> 0:08:47.520 arm hole turk, you would become the turk and the 0:08:47.600 --> 0:08:48.320 churk would fuse. 0:08:48.520 --> 0:08:50.839 That's what some people thought. I think that's what Egar 0:08:50.880 --> 0:08:52.439 Allen Poe thought too. 0:08:52.160 --> 0:08:54.040 He Kidney right. 0:08:54.800 --> 0:08:57.720 Other people thought that there was the the the little 0:08:57.760 --> 0:09:01.480 person was underneath the in the cabin operating the trick 0:09:01.520 --> 0:09:02.679 with levers and stuff like that. 0:09:02.720 --> 0:09:04.679 Well, there could have been a mirror or something, you know, 0:09:04.880 --> 0:09:06.960 I guess, like a telescopic mirror. 0:09:07.040 --> 0:09:09.000 That's what's getting me is how would they keep up 0:09:09.040 --> 0:09:09.720 with the game? 0:09:10.000 --> 0:09:10.200 Right? 0:09:10.559 --> 0:09:12.480 You could keep track of the game, but how could 0:09:12.480 --> 0:09:14.200 you see where the other person moved? You would know 0:09:14.200 --> 0:09:15.960 where you moved, but you wouldn't be able to see 0:09:15.960 --> 0:09:17.880 where the other person moved. That's what I don't get. 0:09:17.840 --> 0:09:19.280 Just mirrors smoking mirrors. 0:09:19.320 --> 0:09:21.959 Maybe so, But the point is is it was a fake. 0:09:22.000 --> 0:09:24.640 It was a fraud, but it raise some really big 0:09:24.760 --> 0:09:28.800 questions about the idea of a machine beating a person 0:09:28.840 --> 0:09:30.480 at something like chess. 0:09:30.800 --> 0:09:34.880 Yeah, and it really peaked the mind of one Charles 0:09:34.960 --> 0:09:38.719 Babbage was he was a kid or young at least 0:09:38.760 --> 0:09:40.960 at the time when he saw the Turk in person, 0:09:41.920 --> 0:09:44.680 and a few years afterward he began work on something 0:09:44.720 --> 0:09:48.559 called the Difference Engine, which was a machine that he 0:09:48.640 --> 0:09:54.679 designed to calculate mathematics automatically. So some point to this 0:09:54.840 --> 0:09:58.480 is kind of maybe the beginnings of humans trying to 0:09:58.559 --> 0:09:59.240 create AI. 0:10:00.320 --> 0:10:04.439 Well yeah, with Babbage's differential machine or difference machine. 0:10:04.200 --> 0:10:06.720 Yeah, difference engine. But at the very least, what this 0:10:06.920 --> 0:10:10.160 is is the first that I know of example of 0:10:10.720 --> 0:10:14.840 man versus machine, even though it was really man versus 0:10:14.880 --> 0:10:16.640 man because it was a man in the machine. 0:10:16.760 --> 0:10:20.600 Right. It was a fraud, Yeah, but it sparked that idea, 0:10:20.760 --> 0:10:25.040 It definitely did. And that's something that like chess in particular, 0:10:25.120 --> 0:10:28.480 has always been like this idea of like, if you 0:10:28.480 --> 0:10:30.760 can teach a machine to play chess, you have really 0:10:30.800 --> 0:10:35.000 achieved a milestone. And there's been you know, plenty of programs, 0:10:35.960 --> 0:10:39.880 most notably Deep Blue, which we'll talk about. But there's 0:10:40.040 --> 0:10:45.720 there's been this idea that like part of AI is chess, 0:10:45.800 --> 0:10:49.520 teaching it to play chess. But they, the people who 0:10:49.559 --> 0:10:53.400 develop AI, never set out to make a chess playing AI, 0:10:53.960 --> 0:10:56.320 just to make a machine that can play chess. That's 0:10:56.360 --> 0:10:59.400 not the point. Chess has always been this way to 0:10:59.520 --> 0:11:02.440 demons straight the progress of artificial intelligence. 0:11:02.520 --> 0:11:06.400 Yeah, because it's a complex game that you can't just 0:11:06.520 --> 0:11:09.120 program it, like it almost has to learn. 0:11:09.960 --> 0:11:12.600 Well, it depends on how you come at it at first, right, 0:11:12.679 --> 0:11:16.920 So initially they did try to program it. Okay, there's 0:11:17.040 --> 0:11:23.280 this from basically nineteen fifty to the about the mid 0:11:23.640 --> 0:11:26.880 like about say nineteen fifty to twenty ten, sixty years right, 0:11:27.440 --> 0:11:31.080 That is how they approached AI and chess is you 0:11:31.240 --> 0:11:35.320 figured out how to break chess down and explain it 0:11:35.360 --> 0:11:38.840 to a computer. Now, what if you could, ideally you 0:11:38.880 --> 0:11:43.480 would have this computer or this AI, this artificial intelligence 0:11:44.679 --> 0:11:48.559 be able to think about the outcome of every possible 0:11:49.360 --> 0:11:52.320 or every possible outcome of a move before making it. Right, 0:11:52.800 --> 0:11:55.559 that's just not possible. Still today we don't have computers 0:11:55.559 --> 0:11:57.680 that can do that. Right, So what you have to 0:11:57.679 --> 0:12:00.520 do is figure out how to create shortcuts for the machine, 0:12:00.840 --> 0:12:03.760 give it best practices, that kind of thing. Yeah, And 0:12:03.960 --> 0:12:06.600 that was actually laid out in nineteen fifty by a 0:12:06.640 --> 0:12:09.840 guy named Claude Shannon who's a father of information theory, 0:12:10.320 --> 0:12:12.920 and he wrote a paper with a pretty on the 0:12:12.960 --> 0:12:17.160 nose title called Programming a Computer for Playing Chess. And 0:12:17.200 --> 0:12:18.720 you have to say it like that when you say 0:12:18.760 --> 0:12:19.079 the name. 0:12:19.280 --> 0:12:21.520 Yeah, it's got a question mark at the end, right, But. 0:12:21.520 --> 0:12:26.320 He laid out two big things. One is creating a 0:12:26.760 --> 0:12:31.320 function of the different moves, and then another one is 0:12:31.360 --> 0:12:35.200 called a mini max. And if those were the two 0:12:35.240 --> 0:12:39.800 things that Shannon laid out, and they established about fifty 0:12:39.880 --> 0:12:42.840 or sixty years of development in teaching an AI to 0:12:42.840 --> 0:12:43.439 play chess. 0:12:43.840 --> 0:12:47.040 Yeah. So this evaluation function is just sort of the base, 0:12:48.160 --> 0:12:50.240 the very basis of it all kind of where it starts, 0:12:50.280 --> 0:12:52.679 which is you a kind of give a number to 0:12:52.840 --> 0:12:57.120 create a numerical evaluation based on the state of the 0:12:57.120 --> 0:13:01.280 board at that moment, right, and assign a real number 0:13:01.320 --> 0:13:06.000 evaluation to it. So the highest number that you would 0:13:06.520 --> 0:13:11.920 shoot for is obviously getting checkmate, getting a king and checkmate. 0:13:11.720 --> 0:13:14.880 Right, right, So what you've just done now is by 0:13:14.920 --> 0:13:18.200 assigning a number to a state like the pieces on 0:13:18.240 --> 0:13:21.800 a board. What you what you've done is to say, 0:13:22.080 --> 0:13:24.800 like shoot for this number. Right, the higher the number, 0:13:24.880 --> 0:13:26.800 like you're going to give this aither rule. Now, the 0:13:26.880 --> 0:13:30.320 higher the number, the more desirable that this move that 0:13:30.360 --> 0:13:34.240 could lead to that higher number. Function. Evaluation function is 0:13:34.360 --> 0:13:35.320 what you want to do. 0:13:35.440 --> 0:13:38.080 Right, like capture the night or capture the queen. Capture 0:13:38.080 --> 0:13:40.199 the queen would have a higher evaluation number. 0:13:40.040 --> 0:13:43.200 Right exactly. So that's the function. Then there's another one 0:13:43.280 --> 0:13:46.040 called the mini max. Yeah, this is pretty great where 0:13:46.040 --> 0:13:48.760 you want to minimize the maximum And this is another 0:13:48.760 --> 0:13:50.839 shortcut that they taught computers. 0:13:50.440 --> 0:13:52.640 The maximum loss that is right, Yeah. 0:13:52.600 --> 0:13:55.120 So what they what they taught computers to do is 0:13:55.400 --> 0:13:58.560 so you no computer can look through an entire game 0:13:58.640 --> 0:14:03.880 every possible outcome, but there are computers that can look 0:14:04.160 --> 0:14:08.120 pretty far down the line at every possible outcome. And 0:14:09.360 --> 0:14:13.080 what you can say is, Okay, you want to find 0:14:13.160 --> 0:14:16.640 the evaluation function that is the worst case scenario, the 0:14:16.640 --> 0:14:22.200 maximum loss, and then find the move that will minimize 0:14:22.240 --> 0:14:24.120 the possibility for that outcome. 0:14:24.600 --> 0:14:26.760 Yeah bye, And this is you're only limited by your 0:14:26.760 --> 0:14:29.680 programming power. But by looking not only at the state 0:14:29.720 --> 0:14:31.560 of the board right now, but if I make this 0:14:31.680 --> 0:14:35.480 move and I move the PAWD to this spot, what 0:14:35.600 --> 0:14:38.640 are the next like three moves possibly that could happen 0:14:38.680 --> 0:14:41.720 as a result of this move. And you're only limited, 0:14:41.760 --> 0:14:45.200 like I said, by programming power. So obviously, the more 0:14:45.440 --> 0:14:48.000 juice you have, the more moves ahead that you can 0:14:48.000 --> 0:14:48.920 look exactly. 0:14:49.240 --> 0:14:52.240 And then they just shy away from ones with the 0:14:52.280 --> 0:14:55.400 higher function number exactly or lower function number, depending on 0:14:55.440 --> 0:14:59.280 how you've programmed it. But they're making these decisions based 0:14:59.320 --> 0:15:02.280 on these rules. And then there's other things you can do, 0:15:02.360 --> 0:15:07.880 like little shortcuts to say, if a decision tree leads 0:15:07.920 --> 0:15:13.640 to the other players King being in checkmate, don't even 0:15:13.680 --> 0:15:16.920 think about that move any further, don't evaluate any longer, 0:15:17.000 --> 0:15:18.840 just abandon it because we would never want to make 0:15:18.880 --> 0:15:21.200 that move. So there's all these shortcuts you can do. 0:15:21.360 --> 0:15:24.000 And that's what they did to teach computers. That's what 0:15:24.400 --> 0:15:27.440 Deep Blue did when it beat Gary Kasparov in nineteen 0:15:27.480 --> 0:15:30.720 ninety seven. It was this huge, massive computer that knew 0:15:30.840 --> 0:15:35.000 a lot of chess, a lot about chess. It had 0:15:35.040 --> 0:15:38.800 a lot of rules, a lot of incredibly intricate programming 0:15:38.800 --> 0:15:43.080 that was extremely sharp, and it actually won. It became 0:15:43.160 --> 0:15:47.360 the first computer to beat an actual human chess grand 0:15:47.400 --> 0:15:50.680 master in like regulation match play. 0:15:51.000 --> 0:15:54.200 Yeah. I mean, and I don't think Kasparov gets enough 0:15:54.240 --> 0:15:58.000 credit for being willing to do this, because it was 0:15:58.080 --> 0:16:01.000 a big deal for him to lose. It was in 0:16:01.040 --> 0:16:04.800 this community and the AI community. It sent shock waves, 0:16:04.960 --> 0:16:09.080 and everyone that was alive remembers, even if you didn't 0:16:09.080 --> 0:16:12.320 know anything about either, one remembers Deep Blue being all 0:16:12.320 --> 0:16:14.480 over the news. It was a really big deal. And 0:16:14.560 --> 0:16:16.960 Kasparov put his name on the line and lost. 0:16:17.200 --> 0:16:19.960 Yeah, And I was wondering, Chuck, how how like you 0:16:20.000 --> 0:16:21.960 would get somebody to do that. 0:16:22.480 --> 0:16:24.920 I'm sure the Mountain of catch. 0:16:25.080 --> 0:16:26.840 I guess that would probably be part of it. But 0:16:26.880 --> 0:16:29.760 I mean, shit, I don't know. 0:16:29.840 --> 0:16:31.320 I bet, I bet that's out there. We just I 0:16:31.360 --> 0:16:32.160 just didn't look it up. 0:16:32.280 --> 0:16:36.240 So that's possible. It's also possible that they said, look, man, 0:16:36.360 --> 0:16:38.720 like this is chess we're talking about or whatever. But really, 0:16:38.760 --> 0:16:43.080 what you're doing is helping advance artificial intelligence. 0:16:42.560 --> 0:16:46.480 Right, because we're not really trying ultimately to win chess games. 0:16:46.720 --> 0:16:47.800 We're trying to cure cancer. 0:16:47.840 --> 0:16:50.440 I mean, yeah, we're going to take your title because 0:16:50.480 --> 0:16:52.840 we're going to beat you, or our machine's going to 0:16:52.880 --> 0:16:55.320 beat you. But even still, you're going to be helping 0:16:55.400 --> 0:16:58.840 with cancer. Think of the cancer Casparov. That's probably what 0:16:58.840 --> 0:16:59.160 they said. 0:16:59.200 --> 0:17:00.000 Should we take a break? 0:17:00.160 --> 0:17:00.360 Yeah? 0:17:00.840 --> 0:17:03.680 Uh wait, well should we tease our special guests first? 0:17:04.440 --> 0:17:06.280 Is he okay? I can smell him. 0:17:06.280 --> 0:17:08.159 I don't think we even said we're gonna have a 0:17:08.160 --> 0:17:12.480 special guest later in the episode, Mister Jonathan Strickland of 0:17:12.560 --> 0:17:13.040 Tech Stuff. 0:17:13.320 --> 0:17:13.560 Nice. 0:17:13.600 --> 0:17:15.960 It's just been a long time since, like years since 0:17:15.960 --> 0:17:16.720 we had Stricken. 0:17:16.920 --> 0:17:18.960 The last time we had Stricken was like two thousand 0:17:19.000 --> 0:17:21.440 and nine with the Necronomicon episode. 0:17:21.480 --> 0:17:24.359 What is what? Where's he been besides sitting in between 0:17:24.400 --> 0:17:24.880 us every day? 0:17:24.960 --> 0:17:27.119 It's been a strickling drought, is what it's been. 0:17:27.200 --> 0:17:29.160 Yeah, so Strickland's coming later, but we're going to come 0:17:29.160 --> 0:17:31.919 back after this and talk a little bit more about 0:17:32.280 --> 0:17:53.439 a man versus machine. 0:17:55.480 --> 0:17:59.240 Okay, dude, So what we just described was how AI 0:17:59.720 --> 0:18:03.199 was to play things like chess or to think like 0:18:03.280 --> 0:18:05.480 you take something, you figure out how to break it 0:18:05.560 --> 0:18:10.080 down into little rules and things that a computer can 0:18:10.160 --> 0:18:13.000 think of, right, and then follow these kind of rules 0:18:13.040 --> 0:18:15.879 to make the best decision. That's how it used to be. 0:18:16.800 --> 0:18:20.840 The way that it's done now that everybody's doing now 0:18:21.280 --> 0:18:24.400 is where you are creating a machine that teaches itself. 0:18:24.560 --> 0:18:25.760 Yeah, that's the jam. 0:18:25.840 --> 0:18:29.159 That was the breakthrough. You may have noticed back in 0:18:29.200 --> 0:18:34.760 about ty thirteen twenty fourteen, all of a sudden, things 0:18:34.800 --> 0:18:39.520 like Siri and Alexa got way better at what they 0:18:39.560 --> 0:18:44.320 are doing. They got way less confused. Really, your navigation 0:18:44.440 --> 0:18:47.000 app got a lot better. And the reason why is 0:18:47.040 --> 0:18:50.199 because this type, this new type of AI, this new 0:18:50.240 --> 0:18:54.520 type of machine learning that can teach itself and learn 0:18:54.880 --> 0:18:57.760 on its own, just hit the scene and they just 0:18:57.800 --> 0:19:00.360 started exploding. And one of the things that they were 0:19:00.359 --> 0:19:02.159 first trained on was games. 0:19:02.440 --> 0:19:04.720 Yeah, and it makes sense. And if you thought chess 0:19:04.840 --> 0:19:08.800 was complicated and difficult, when it comes to these new 0:19:08.840 --> 0:19:13.600 AIS that they're teaching to teach themselves game strategy. They said, 0:19:13.720 --> 0:19:16.760 we might as well dive in to the Chinese strategic 0:19:16.840 --> 0:19:21.160 game Go because it has been called the most complex 0:19:21.240 --> 0:19:24.800 game ever devised by humans. Yeah, and this was actually 0:19:24.800 --> 0:19:30.280 that was actually a quote from Demi Hasabi, a neuroscientist 0:19:30.480 --> 0:19:35.840 and the founder of deep Mind, which was deep Mind. 0:19:35.840 --> 0:19:38.440 They were purchased by Google or were they always part 0:19:38.440 --> 0:19:38.879 of Google. 0:19:39.040 --> 0:19:41.280 I don't know if they were a spun off branch 0:19:41.359 --> 0:19:43.640 or where they were purchased, but it's one of Google's 0:19:43.720 --> 0:19:45.159 AI outfits. 0:19:45.280 --> 0:19:47.640 Well, they're one of the teams, yeah, that are designing 0:19:47.640 --> 0:19:50.439 these new programs. And to give you an idea of 0:19:50.440 --> 0:19:53.560 how complex Go is, it deals with a board with 0:19:54.040 --> 0:19:59.520 different stones and there are ten how do you even 0:19:59.560 --> 0:19:59.960 say that? 0:20:00.119 --> 0:20:02.720 Ten to one hundred and seventieth power, So. 0:20:02.720 --> 0:20:06.479 That means one hundred and seventy zeros and take that 0:20:06.560 --> 0:20:09.560 number and that's the number of possible configurations of a 0:20:09.560 --> 0:20:10.359 go board. 0:20:10.359 --> 0:20:14.680 Right, So, like you say, chess is very complex and complicated, 0:20:14.720 --> 0:20:17.760 and it's very difficult to master Go. And I've never 0:20:17.800 --> 0:20:21.160 played Go, of you no, So it's supposedly it's easy 0:20:21.200 --> 0:20:21.879 to learn. 0:20:21.960 --> 0:20:24.840 Right, but very complicated in its simplicity. 0:20:24.440 --> 0:20:28.360 Right, right, exactly, it's extremely difficult to master. And there 0:20:28.400 --> 0:20:30.960 was a guy in the late nineties and I'm guessing 0:20:31.080 --> 0:20:36.560 that he was saying this after Deep Blue beat Caspar 0:20:36.560 --> 0:20:40.919 ofv It was an astrophysicist from Princeton. He said that 0:20:41.040 --> 0:20:43.320 it would probably be one hundred years before a computer 0:20:43.400 --> 0:20:45.760 beats a human a go. To give you an idea 0:20:45.760 --> 0:20:50.040 of just how complex GO is that deeply would just 0:20:50.080 --> 0:20:52.600 be caspar OFV. And this guy's saying it'll still be 0:20:52.640 --> 0:20:55.399 one hundred years before anyone gets beat at GO by 0:20:55.400 --> 0:20:55.920 a computer. 0:20:56.119 --> 0:20:58.399 And he was someone who knew about this stuff, who 0:20:58.440 --> 0:21:01.119 was an astrophysicist. He was just some schmoe at home 0:21:01.160 --> 0:21:02.600 and drunken as reclining. 0:21:03.280 --> 0:21:05.840 Is making asinine predictions. 0:21:07.040 --> 0:21:10.360 So and again we've said this before, but I want 0:21:10.359 --> 0:21:14.159 to reiterate the people that I think Alpha Go is 0:21:14.200 --> 0:21:16.600 the name of this program. The people that created this 0:21:16.680 --> 0:21:20.359 at Deep Mind, they wanted to stress that this is 0:21:21.080 --> 0:21:24.240 a problem solving program. We're just teaching it this game 0:21:24.280 --> 0:21:27.400 at first, just to make it learn and to see 0:21:27.400 --> 0:21:29.399 if it can get good at what it does. But 0:21:30.240 --> 0:21:33.000