1 00:00:00,120 --> 00:00:03,400 Speaker 1: We're going to start our story today. In nineteen sixty eight, 2 00:00:03,640 --> 00:00:07,640 Speaker 1: way before I was born and probably before you were born. Alistair. 3 00:00:07,880 --> 00:00:13,360 Speaker 1: Very funny, Yes, I was born after so before either 4 00:00:13,440 --> 00:00:16,640 Speaker 1: of us were born. Terry Winnigrad was a young grad 5 00:00:16,760 --> 00:00:20,840 Speaker 1: student in Massachusetts, and Terry had just started at m 6 00:00:20,880 --> 00:00:26,720 Speaker 1: I t S brand new artificial intelligence lab. Basically, the 7 00:00:26,760 --> 00:00:30,080 Speaker 1: culture was, look, nobody has ever tried doing all this 8 00:00:30,160 --> 00:00:33,279 Speaker 1: kind of stuff with computers. They've done business calculations or whatever, 9 00:00:33,360 --> 00:00:36,040 Speaker 1: but nobody's trying to get them to see things, or 10 00:00:36,159 --> 00:00:40,479 Speaker 1: to move a robot arm or to do language. And 11 00:00:40,880 --> 00:00:46,320 Speaker 1: therefore we're on the exploring cutting edge, and we're going 12 00:00:46,360 --> 00:00:50,360 Speaker 1: to solve all these problems right soon. And Terry started 13 00:00:50,400 --> 00:00:54,280 Speaker 1: building a computer program that turned into his dissertation, which 14 00:00:54,280 --> 00:00:58,160 Speaker 1: he called shirt looke Yeah, with only one vowel, s 15 00:00:58,360 --> 00:01:01,880 Speaker 1: h r d l U. We're watching a video here 16 00:01:01,960 --> 00:01:04,960 Speaker 1: of a demo of shurtloo that Terry had recorded at 17 00:01:04,959 --> 00:01:08,959 Speaker 1: the time. It's this silent, black and white, low rez, 18 00:01:09,080 --> 00:01:12,440 Speaker 1: grainy video, and you see a virtual table top and 19 00:01:12,520 --> 00:01:15,360 Speaker 1: a bunch of geometric shapes on top. It looks like 20 00:01:15,400 --> 00:01:19,920 Speaker 1: a computer sketch of a really really boring form of Tetris. Yeah, 21 00:01:20,000 --> 00:01:23,679 Speaker 1: chat bots are everywhere today, but Shirtlou was really one 22 00:01:23,720 --> 00:01:27,440 Speaker 1: of the original chat bots in history. Using this machine 23 00:01:27,480 --> 00:01:30,360 Speaker 1: called a teletype that's like a typewriter on a tray 24 00:01:30,400 --> 00:01:33,240 Speaker 1: hooked up to a gigantic computer the size of a 25 00:01:33,319 --> 00:01:38,040 Speaker 1: small bathroom, Terry created this interactive assistant that could help 26 00:01:38,080 --> 00:01:41,279 Speaker 1: you navigate the virtual world of blocks on a table. 27 00:01:42,120 --> 00:01:44,560 Speaker 1: So you could type in something like pick up the 28 00:01:44,600 --> 00:01:48,000 Speaker 1: red block and stack it on top of the blue cube, 29 00:01:48,120 --> 00:01:51,920 Speaker 1: and like magic, this virtual robotic arm would appear and 30 00:01:51,960 --> 00:01:54,560 Speaker 1: do that stacking for you. And you could ask your 31 00:01:54,640 --> 00:01:58,360 Speaker 1: lue questions too, like which cube is sitting on the table. 32 00:01:58,680 --> 00:02:01,240 Speaker 1: And on the display screen you see this text show up, 33 00:02:01,360 --> 00:02:05,240 Speaker 1: letter by letter, replying to your prompt. And the amazing 34 00:02:05,320 --> 00:02:08,840 Speaker 1: thing about Surtly was the people into acting with it 35 00:02:09,200 --> 00:02:12,400 Speaker 1: would use normal English. You didn't have to click on 36 00:02:12,400 --> 00:02:14,800 Speaker 1: a bunch of buttons, you didn't have to throw in 37 00:02:14,880 --> 00:02:17,240 Speaker 1: a string of numbers, and you didn't have to use 38 00:02:17,280 --> 00:02:21,640 Speaker 1: any obscure computer programming language. Yeah, watching this video of 39 00:02:21,680 --> 00:02:24,639 Speaker 1: the program, now it feels like you're watching two people 40 00:02:24,760 --> 00:02:28,080 Speaker 1: texting each other, even though it's a person talking to 41 00:02:28,160 --> 00:02:31,680 Speaker 1: a computer. And so if you and I are impressed 42 00:02:31,720 --> 00:02:34,840 Speaker 1: watching this thing in action today. You can imagine how 43 00:02:34,880 --> 00:02:38,240 Speaker 1: stunned people were when they saw this half a century ago. 44 00:02:38,760 --> 00:02:41,680 Speaker 1: And from there the field of AI was supposed to 45 00:02:41,680 --> 00:02:44,880 Speaker 1: take off like a rocket ship. Terry hunker down to 46 00:02:44,960 --> 00:02:47,960 Speaker 1: bring Shortlood to life, and he wanted it to work 47 00:02:47,960 --> 00:02:50,960 Speaker 1: in a universe bigger and more complicated than just a 48 00:02:50,960 --> 00:02:54,200 Speaker 1: couple blocks on a table. But the deeper he got 49 00:02:54,200 --> 00:02:57,480 Speaker 1: into it, he was running into more and more obstacles. 50 00:02:57,840 --> 00:03:01,040 Speaker 1: And after a while he gave up on Shurtle, and 51 00:03:01,080 --> 00:03:04,160 Speaker 1: he gave up on AI. He ended up leaving the 52 00:03:04,200 --> 00:03:14,239 Speaker 1: field all together. Hi. This is Aki Ito and I'm 53 00:03:14,240 --> 00:03:17,560 Speaker 1: Alista bar And this week on Decrypted, we're taking you 54 00:03:17,600 --> 00:03:21,080 Speaker 1: to Stanford to meet one of the most influential thinkers 55 00:03:21,120 --> 00:03:24,680 Speaker 1: in the history of computing. This is an early pioneer 56 00:03:24,720 --> 00:03:26,960 Speaker 1: in the field of AI who built one of the 57 00:03:26,960 --> 00:03:31,920 Speaker 1: most impressive reasoning machines and ultimately concluded that computers wouldn't 58 00:03:31,919 --> 00:03:34,680 Speaker 1: be able to match human intelligence in his lifetime. And 59 00:03:34,720 --> 00:03:38,360 Speaker 1: when he came to that conclusion, Terry Winnigrad dedicated the 60 00:03:38,440 --> 00:03:42,720 Speaker 1: rest of his career to improving computers not as a 61 00:03:42,840 --> 00:03:45,480 Speaker 1: replacement for human thought, but as a tool to help 62 00:03:45,520 --> 00:03:49,880 Speaker 1: all of us. Terry's fingerprints are everywhere in the technologies 63 00:03:49,960 --> 00:03:53,800 Speaker 1: powering modern life today, including Google Search, which started out 64 00:03:53,840 --> 00:03:57,720 Speaker 1: as Larry Page's grad school research project, and this was 65 00:03:57,760 --> 00:04:02,320 Speaker 1: something that Terry supervised. Now, with all these digital assistance 66 00:04:02,720 --> 00:04:05,560 Speaker 1: powering our everyday lives, we're going to have Terry, the 67 00:04:05,680 --> 00:04:09,440 Speaker 1: very creator of their precursor, test them all out. Yeah, 68 00:04:09,520 --> 00:04:13,480 Speaker 1: consider him the great grandfather of Syrie or Amazon's Alexa. 69 00:04:13,920 --> 00:04:17,080 Speaker 1: You might be surprised by Terry's conclusion on these devices 70 00:04:17,440 --> 00:04:19,560 Speaker 1: on how far he thinks the field has come since 71 00:04:19,560 --> 00:04:22,280 Speaker 1: he unveiled Shirtly to the world almost fifty years ago. 72 00:04:22,960 --> 00:04:25,640 Speaker 1: And he'll give us his thoughts on where these devices 73 00:04:25,640 --> 00:04:30,000 Speaker 1: are headed too. Don't worry, I have a long way 74 00:04:30,040 --> 00:04:37,279 Speaker 1: to go before I become smarter than you. Okay, so 75 00:04:37,400 --> 00:04:40,880 Speaker 1: let's rewind way back to a time before the days 76 00:04:40,920 --> 00:04:48,240 Speaker 1: of personal computers. It's Elvis Presley is on the radio. 77 00:04:48,839 --> 00:04:54,240 Speaker 1: Russia had launched Sputnik only three years ago. Russia I 78 00:04:54,279 --> 00:04:56,960 Speaker 1: had blasted a man made moon into and it would 79 00:04:56,960 --> 00:04:59,120 Speaker 1: be nine years until the US put a man on 80 00:04:59,160 --> 00:05:03,280 Speaker 1: the moon. Terry, meanwhile, is in high school in Greely, Colorado. 81 00:05:03,680 --> 00:05:06,360 Speaker 1: The physics teacher said, you know, you don't really need 82 00:05:06,400 --> 00:05:09,880 Speaker 1: to sit through these lectures in my physics class. I'll 83 00:05:09,880 --> 00:05:11,720 Speaker 1: give you a workshop up in the attic. It was 84 00:05:11,720 --> 00:05:13,400 Speaker 1: in the attic of the school, and you wanted to 85 00:05:13,400 --> 00:05:17,159 Speaker 1: build something interesting. And this is when Terry first encounter 86 00:05:17,200 --> 00:05:20,680 Speaker 1: as a computer. And my father owned a what was 87 00:05:20,720 --> 00:05:23,000 Speaker 1: a steel business, but had grown up from being a 88 00:05:23,080 --> 00:05:27,200 Speaker 1: junkyard and had a huge collection of old junk stuff. 89 00:05:27,240 --> 00:05:30,000 Speaker 1: They picked up a governments plus sales and so I 90 00:05:30,040 --> 00:05:33,720 Speaker 1: took spare parts from the junkyard and I built a 91 00:05:33,760 --> 00:05:37,320 Speaker 1: little computer. A box looked like a bread box. Uh. 92 00:05:37,360 --> 00:05:39,039 Speaker 1: In fact, the case I put it in may have 93 00:05:39,120 --> 00:05:42,760 Speaker 1: actually been a bread box. Uh. And it did a 94 00:05:42,880 --> 00:05:48,120 Speaker 1: very simplistic computation, but it was it worked. He attended 95 00:05:48,160 --> 00:05:51,560 Speaker 1: Colorado College, a liberal arts university, where he was a 96 00:05:51,600 --> 00:05:54,600 Speaker 1: math major. But what he really fell in love with 97 00:05:54,640 --> 00:05:58,599 Speaker 1: there was linguistics, the science of how language it's structured 98 00:05:58,600 --> 00:06:01,320 Speaker 1: and understood. And he spent a year in London after 99 00:06:01,360 --> 00:06:05,200 Speaker 1: he graduated, studying linguistics too. And after that, in the 100 00:06:05,279 --> 00:06:09,280 Speaker 1: late nineteen sixties, Terry started his PhD at m I. T. 101 00:06:09,520 --> 00:06:13,919 Speaker 1: S AI Lab, led by Marvin Minsky himself. Minsky is 102 00:06:13,960 --> 00:06:17,520 Speaker 1: one of the godfathers of AI. Everyone felt the promise 103 00:06:17,680 --> 00:06:21,200 Speaker 1: and it was you know, they say that people doing 104 00:06:21,200 --> 00:06:23,159 Speaker 1: it were like undergraduates. I wasn't. I was an old 105 00:06:23,200 --> 00:06:25,520 Speaker 1: person right in that group. So there was a sort 106 00:06:25,560 --> 00:06:27,800 Speaker 1: of youthful sense of where the new generation we're going 107 00:06:27,839 --> 00:06:30,200 Speaker 1: to make things happen. So Terry got to work on 108 00:06:30,240 --> 00:06:34,680 Speaker 1: surely the Intelligent Seeming software that we introduced earlier. It 109 00:06:34,720 --> 00:06:38,040 Speaker 1: was grueling work piecing together different parts of of software, 110 00:06:38,400 --> 00:06:42,279 Speaker 1: but it gradually came together. I think one of the 111 00:06:42,279 --> 00:06:45,560 Speaker 1: most striking things about the program, in addition to this 112 00:06:45,720 --> 00:06:47,800 Speaker 1: direct visual you can see what it was doing you 113 00:06:47,839 --> 00:06:51,159 Speaker 1: could talk about JOIN, is that I attempted to deal 114 00:06:51,240 --> 00:06:55,039 Speaker 1: with some of the interesting properties of language in that 115 00:06:55,120 --> 00:06:58,880 Speaker 1: you don't say everything that's explicitly so I used to 116 00:06:58,920 --> 00:07:00,960 Speaker 1: I mean, take the most obviously ample there, put it 117 00:07:01,000 --> 00:07:04,880 Speaker 1: next to a red one. A read what what is 118 00:07:04,920 --> 00:07:07,200 Speaker 1: a one? And to know what you mean, you of 119 00:07:07,240 --> 00:07:08,960 Speaker 1: course have to go back into the context. You have 120 00:07:09,000 --> 00:07:12,680 Speaker 1: to know that previously you said find a blue block, 121 00:07:12,800 --> 00:07:14,720 Speaker 1: put it next to a red one. So now you're 122 00:07:14,720 --> 00:07:16,680 Speaker 1: gonna go back a sentence and find that you meant block, 123 00:07:17,400 --> 00:07:20,200 Speaker 1: or you could say pick up another one, What does 124 00:07:20,240 --> 00:07:23,920 Speaker 1: another mean the whole and pronounce put it? If I say, 125 00:07:24,080 --> 00:07:27,000 Speaker 1: now put it in the box. It has to figure 126 00:07:27,000 --> 00:07:29,760 Speaker 1: out which of the things from the previous world you 127 00:07:29,800 --> 00:07:33,520 Speaker 1: meant by it. So it had a very natural flow 128 00:07:33,600 --> 00:07:37,240 Speaker 1: to it. When Terry revealed short lude to the World 129 00:07:37,320 --> 00:07:41,640 Speaker 1: in two in the form of an entire issue of 130 00:07:41,680 --> 00:07:46,560 Speaker 1: the Journal of Cognitive Psychology, the world was amazed. We're 131 00:07:46,560 --> 00:07:48,000 Speaker 1: starting here at zero. We don't know what we can do. 132 00:07:48,040 --> 00:07:50,080 Speaker 1: When you make a big first step, you say, hey, 133 00:07:50,960 --> 00:07:52,720 Speaker 1: I'm gonna keep going up right, I mean, and so 134 00:07:53,080 --> 00:07:54,640 Speaker 1: I think that the fact they could do as much 135 00:07:54,680 --> 00:07:57,880 Speaker 1: as it did certainly gave people like marvit Minsk a 136 00:07:57,920 --> 00:08:00,400 Speaker 1: lot of confidence. I mean, he was I think a 137 00:08:00,400 --> 00:08:06,880 Speaker 1: bigger booster of my program than I was. The confidence 138 00:08:07,000 --> 00:08:10,920 Speaker 1: Terry's referring to here. That's confidence in the progress scientists 139 00:08:10,960 --> 00:08:13,760 Speaker 1: were making to develop computers that were just as smart 140 00:08:13,800 --> 00:08:17,960 Speaker 1: as humans. Terry's breakthrough inspired a lot of smart people 141 00:08:18,000 --> 00:08:21,040 Speaker 1: to believe that sentient computers were just around the corner, 142 00:08:21,520 --> 00:08:24,160 Speaker 1: and so shortly after Terry left m I t for 143 00:08:24,240 --> 00:08:28,800 Speaker 1: the warmer pastures of California, back when Silicon Value was 144 00:08:28,840 --> 00:08:32,320 Speaker 1: a quiet place full of fruit orchards, ah so nice, 145 00:08:32,440 --> 00:08:36,160 Speaker 1: no traffic. And as a professor at Stanford and a 146 00:08:36,240 --> 00:08:40,880 Speaker 1: researcher at the legendary research labs rox Park, he worked 147 00:08:40,880 --> 00:08:43,559 Speaker 1: on trying to expand Shortloo. He was trying to get 148 00:08:43,600 --> 00:08:46,640 Speaker 1: it to work in a more complicated environment than just 149 00:08:46,720 --> 00:08:49,160 Speaker 1: a couple of geometric shapes on top of the table. 150 00:08:49,720 --> 00:08:53,439 Speaker 1: And then Shirtloo took over the world and destroyed mankind 151 00:08:55,800 --> 00:08:59,200 Speaker 1: or not, things weren't going the way Terry had hoped. 152 00:08:59,240 --> 00:09:03,160 Speaker 1: The attempt we were making at in that project was 153 00:09:03,240 --> 00:09:09,440 Speaker 1: to come to a broader analysis of meaning which could 154 00:09:09,480 --> 00:09:13,120 Speaker 1: handle the ways in which meaning is much vaguer and 155 00:09:13,520 --> 00:09:17,319 Speaker 1: less systematic than it was for the blocks world. And 156 00:09:17,559 --> 00:09:19,640 Speaker 1: in my talks about this, I always use blocks as 157 00:09:19,640 --> 00:09:21,680 Speaker 1: an example because it's a really simple one, which is 158 00:09:22,160 --> 00:09:24,800 Speaker 1: in that world, block meant exactly one thing. It meant 159 00:09:24,840 --> 00:09:29,880 Speaker 1: a rectangular shaped object of certain sizes on. But if 160 00:09:29,880 --> 00:09:32,600 Speaker 1: I say to you, let's walk around the block, it 161 00:09:32,679 --> 00:09:35,120 Speaker 1: has a second meaning which is different. So you say, okay, 162 00:09:35,120 --> 00:09:36,480 Speaker 1: so you have to put in two meetings. But then 163 00:09:36,520 --> 00:09:38,480 Speaker 1: I say something like that, Well, you know, I'm trying 164 00:09:38,480 --> 00:09:39,840 Speaker 1: to write his paper, but there's some kind of a 165 00:09:39,840 --> 00:09:42,200 Speaker 1: block I can't get over now, it's not even a 166 00:09:42,200 --> 00:09:49,000 Speaker 1: physical object. It's a metaphorical physical objects. So language really 167 00:09:49,160 --> 00:09:52,800 Speaker 1: works that way. Very little of it has precisely defined 168 00:09:52,840 --> 00:09:57,239 Speaker 1: meetings outside of technical stuff. Uh, and we're always extending 169 00:09:57,240 --> 00:10:01,160 Speaker 1: meetings and using implicit metaphor. It's not fancy metaphors, you know, 170 00:10:01,240 --> 00:10:03,960 Speaker 1: life is a rose bowl of roses or whatever. But 171 00:10:04,360 --> 00:10:08,200 Speaker 1: just like a block in my mental thinking, um, and 172 00:10:08,800 --> 00:10:12,319 Speaker 1: trying to write the logic, the algorithms, the underlying computer 173 00:10:12,360 --> 00:10:16,400 Speaker 1: stuff which could handle that is a totally different problem 174 00:10:16,600 --> 00:10:19,760 Speaker 1: from just handling the simple logical stuff. And we tackled 175 00:10:19,760 --> 00:10:25,000 Speaker 1: that problem. And in hindsight, I would say I was 176 00:10:25,040 --> 00:10:28,640 Speaker 1: aware we weren't getting very far. And while Terry's doubts 177 00:10:28,640 --> 00:10:31,360 Speaker 1: were growing, he also started hanging out with a few 178 00:10:31,400 --> 00:10:35,319 Speaker 1: academics in the Bay Area. There were philosophers Hubert Dreyfus 179 00:10:35,440 --> 00:10:39,120 Speaker 1: and John Searle, and there was this Chilean engineer, entrepreneur 180 00:10:39,200 --> 00:10:43,520 Speaker 1: and politician called Fernando Flores. They were all making the 181 00:10:43,720 --> 00:10:47,560 Speaker 1: same broad point around this time, which was this the 182 00:10:47,640 --> 00:10:51,120 Speaker 1: way our brains work. So much of it happens without 183 00:10:51,200 --> 00:10:54,560 Speaker 1: us explicitly thinking about it in a logical way, like 184 00:10:55,120 --> 00:10:58,680 Speaker 1: if A then B and if B, then C. It's 185 00:10:58,720 --> 00:11:02,720 Speaker 1: this complete black box to us. So maybe it wasn't 186 00:11:02,800 --> 00:11:05,080 Speaker 1: ever going to be possible to get a machine to 187 00:11:05,160 --> 00:11:07,439 Speaker 1: do all the things that are human brains can do. 188 00:11:08,720 --> 00:11:12,080 Speaker 1: This philosophy resonated more and more with Terry, and in 189 00:11:12,080 --> 00:11:15,880 Speaker 1: the meantime, the entire field of AI, who's having this 190 00:11:16,000 --> 00:11:20,000 Speaker 1: moment of reckoning, as anybody does when they're doing a 191 00:11:20,040 --> 00:11:23,000 Speaker 1: new technology and they need to get grants, they will 192 00:11:23,040 --> 00:11:25,000 Speaker 1: say it's going to solve all the problems in the world, right, 193 00:11:25,040 --> 00:11:28,120 Speaker 1: best things since life bread, and then it filters out. 194 00:11:28,200 --> 00:11:30,440 Speaker 1: And so what happened is that a I hit this 195 00:11:30,559 --> 00:11:34,640 Speaker 1: point where the claims had overreached. The results were not 196 00:11:35,120 --> 00:11:38,880 Speaker 1: that great. There were good in certain small technical areas. 197 00:11:38,880 --> 00:11:41,440 Speaker 1: It wasn't there was no results, but nothing on the 198 00:11:41,480 --> 00:11:45,760 Speaker 1: scale of what people were promising, and that ushered in 199 00:11:46,080 --> 00:11:51,160 Speaker 1: what became known as the AI winter. Research projects got defunded, 200 00:11:51,440 --> 00:11:56,080 Speaker 1: startups died, All this excitement withered away. So by the 201 00:11:56,160 --> 00:11:59,920 Speaker 1: nineteen eighties Terry was pretty much convinced that he'd reach 202 00:12:00,080 --> 00:12:02,880 Speaker 1: the dead end, which I don't know you think would 203 00:12:02,920 --> 00:12:06,840 Speaker 1: be this devastating realization for him. But right around this 204 00:12:06,920 --> 00:12:13,760 Speaker 1: time something else came along up until approximately three. If 205 00:12:13,800 --> 00:12:16,200 Speaker 1: you use the computer, you were a technical nerd in 206 00:12:16,240 --> 00:12:19,080 Speaker 1: the basement somewhere, and the fact you had to learn 207 00:12:19,120 --> 00:12:23,080 Speaker 1: all sorts of arcane stuff was yeah, that's what we do. Uh. 208 00:12:23,120 --> 00:12:25,959 Speaker 1: And then there was the computer for the rest of us. 209 00:12:26,920 --> 00:12:30,280 Speaker 1: We can draw a picture or it can draw conclusions. 210 00:12:31,080 --> 00:12:33,720 Speaker 1: It's a personal computer from Apple up and it's as 211 00:12:33,720 --> 00:12:39,160 Speaker 1: easy to use as this Macintosh, the computer for the 212 00:12:39,240 --> 00:12:42,360 Speaker 1: rest of us. If you're a tech you probably have 213 00:12:42,400 --> 00:12:44,839 Speaker 1: seen that ad at some point, right and Apple came 214 00:12:44,840 --> 00:12:47,360 Speaker 1: out with mac and all of a sudden you had 215 00:12:47,400 --> 00:12:49,480 Speaker 1: all these people who wanted to use computers who were 216 00:12:49,520 --> 00:12:51,719 Speaker 1: not tech nerds, and we're not willing to learn all 217 00:12:51,760 --> 00:12:54,319 Speaker 1: the arcane stuff, and so the whole field of how 218 00:12:54,320 --> 00:12:56,839 Speaker 1: do you make them something that ordinary people can deal 219 00:12:56,880 --> 00:13:02,480 Speaker 1: with blossom. So Terry decided not to obsess over building 220 00:13:02,520 --> 00:13:05,120 Speaker 1: computers that were going to truly think, and he wasn't 221 00:13:05,120 --> 00:13:08,040 Speaker 1: going to worry about understanding how the brain really works either. 222 00:13:08,760 --> 00:13:11,679 Speaker 1: John Lily, a former student of Terry's who's a partner 223 00:13:11,760 --> 00:13:16,079 Speaker 1: at the venture capital firm Graylock Capital, described Terry's vision 224 00:13:16,160 --> 00:13:19,960 Speaker 1: like this, machines can't do everything, but focusing on what 225 00:13:20,000 --> 00:13:24,080 Speaker 1: the machine capabild years is always a mistakes. He really 226 00:13:24,080 --> 00:13:27,360 Speaker 1: want to pucus on the whole system, which includes humans, 227 00:13:26,800 --> 00:13:31,360 Speaker 1: and they include humans and machines in the interviews between 228 00:13:31,400 --> 00:13:35,240 Speaker 1: two machines is the key, and that philosophy helps shape 229 00:13:35,280 --> 00:13:38,079 Speaker 1: the course of what was at the time a budding 230 00:13:38,120 --> 00:13:42,400 Speaker 1: discipline called Human computer interaction h c I. Is all 231 00:13:42,440 --> 00:13:47,200 Speaker 1: about designing tools to help us humans use computers more easily. Yeah, 232 00:13:47,280 --> 00:13:50,720 Speaker 1: taking these clunky machines out of research labs and putting 233 00:13:50,720 --> 00:13:53,920 Speaker 1: them into the hands of you and me in Silicon 234 00:13:54,040 --> 00:13:57,880 Speaker 1: Valley today, you hear this acronym h CI all the time. 235 00:13:58,280 --> 00:14:01,719 Speaker 1: That's a very very strong view. The printed on me 236 00:14:02,000 --> 00:14:07,640 Speaker 1: for sure, like everybody looking out program massa others. Okay, 237 00:14:07,679 --> 00:14:11,600 Speaker 1: so that's Marissa as in Marissa Meyer, Google employee number 238 00:14:11,600 --> 00:14:15,400 Speaker 1: twenty and the CEO of Yahoo and read as in 239 00:14:15,520 --> 00:14:19,040 Speaker 1: Reed Hoffman, a founding director of PayPal and the creator 240 00:14:19,160 --> 00:14:23,360 Speaker 1: of LinkedIn. But the most famous of Terry students are 241 00:14:23,400 --> 00:14:27,640 Speaker 1: Google co founders Larry Page and Sergey Bryn. They met 242 00:14:27,680 --> 00:14:31,800 Speaker 1: as graduate students in They started working on a way 243 00:14:31,800 --> 00:14:34,560 Speaker 1: to organize all the different web pages out there, and 244 00:14:34,680 --> 00:14:38,800 Speaker 1: Terry supervised their project that turned into the foundations of 245 00:14:38,880 --> 00:14:42,040 Speaker 1: Google Search. I mean, something that helps us access the 246 00:14:42,160 --> 00:14:46,240 Speaker 1: trillions of web pages of content. I can't really think 247 00:14:46,280 --> 00:14:49,200 Speaker 1: of a single digital tool that's been more useful in 248 00:14:49,280 --> 00:14:52,920 Speaker 1: modern life. Thanks to Terry's guidance, Google was born as 249 00:14:52,960 --> 00:14:57,800 Speaker 1: a company on September four and it's now the world's 250 00:14:57,800 --> 00:15:02,160 Speaker 1: second most valuable company. Pretty cool stuff. Hey, I can 251 00:15:02,240 --> 00:15:05,760 Speaker 1: calm down, calm down. We should point out that Terry 252 00:15:05,920 --> 00:15:09,720 Speaker 1: wasn't always this business genius, passing down sage advice to 253 00:15:09,760 --> 00:15:12,880 Speaker 1: his students. And I would say to Larry occasionally, well, yeah, 254 00:15:12,920 --> 00:15:14,680 Speaker 1: that's great, but how you're going to make money with this? 255 00:15:15,480 --> 00:15:17,720 Speaker 1: I think what you should do is my advice is, 256 00:15:17,760 --> 00:15:19,880 Speaker 1: you know, find a company like Microsoft or somebody who 257 00:15:19,880 --> 00:15:22,040 Speaker 1: needs a search engine and sell it for a nice 258 00:15:22,080 --> 00:15:25,760 Speaker 1: chuck of change. I always say, they're fortunate they took 259 00:15:25,800 --> 00:15:30,720 Speaker 1: my technical advice, but not my business advice. So let's 260 00:15:30,800 --> 00:15:34,880 Speaker 1: fast forward about two decades to today, Larry Page. It's 261 00:15:34,960 --> 00:15:37,320 Speaker 1: still at the helm of Google, and one of the 262 00:15:37,360 --> 00:15:42,160 Speaker 1: company's biggest bets is its digital assistant. It's called Google Assistant, 263 00:15:42,640 --> 00:15:46,440 Speaker 1: and it's connected to its new smartphone called Pixel, and 264 00:15:46,480 --> 00:15:49,240 Speaker 1: it's also connected to its home device, which is like 265 00:15:49,640 --> 00:15:52,920 Speaker 1: this portable speaker that speaks back to you, and that 266 00:15:52,960 --> 00:15:56,440 Speaker 1: follows in the footsteps of all the other digital assistance 267 00:15:56,480 --> 00:15:59,360 Speaker 1: out there. First of all, they're Sirie, which is the 268 00:15:59,400 --> 00:16:04,000 Speaker 1: assistant on the iPhone. Hello therecky. There's the Amazon Echo, 269 00:16:04,080 --> 00:16:08,160 Speaker 1: which is called Alexa. Hello, I'm here, And then there's 270 00:16:08,240 --> 00:16:12,200 Speaker 1: Microsoft's assistant, which is called Cortana. Hey there, my friend. 271 00:16:12,640 --> 00:16:15,160 Speaker 1: We wanted to know if these new helpers are useful 272 00:16:15,200 --> 00:16:18,360 Speaker 1: and smart, so you better to quiz them than Terry. 273 00:16:18,840 --> 00:16:21,840 Speaker 1: Along with our editor Emily A. Busso we started sending 274 00:16:21,920 --> 00:16:27,520 Speaker 1: up these devices on Terry's desk in his office. Hey, Alexa, 275 00:16:27,600 --> 00:16:34,520 Speaker 1: are you on Hello I'm here? Okay, great, Alexa's working yea. 276 00:16:34,600 --> 00:16:38,880 Speaker 1: So we have we have Amazon Echo Alexa. These are 277 00:16:38,920 --> 00:16:44,280 Speaker 1: all on Professor Winograds desk in his office. Alexa has 278 00:16:44,320 --> 00:16:48,080 Speaker 1: just turned herself on you listening to me? Okay. So 279 00:16:48,240 --> 00:16:51,920 Speaker 1: here's the first question. This is Terry asking Siri, where's 280 00:16:51,960 --> 00:16:58,200 Speaker 1: a nightclub that my methodist uncle would enjoy? Okay, check 281 00:16:58,240 --> 00:17:02,720 Speaker 1: it out? What is it? Show? Show some random nightclubs? Now, 282 00:17:02,720 --> 00:17:05,520 Speaker 1: I have no idea if they have any you know, 283 00:17:06,880 --> 00:17:12,720 Speaker 1: Holy holy Holy Town nightclub the grants, So it gave 284 00:17:12,880 --> 00:17:17,359 Speaker 1: us nightclubs, but we're not sure something our Methodist uncle 285 00:17:18,640 --> 00:17:21,440 Speaker 1: could have. Holy cow, do you think it probably understood today? 286 00:17:21,600 --> 00:17:24,800 Speaker 1: We may have been religion and holy that This is 287 00:17:25,000 --> 00:17:27,680 Speaker 1: the problem with this kind of AI, which is there's 288 00:17:27,760 --> 00:17:31,440 Speaker 1: no logical chain you can follow, but that may have 289 00:17:31,600 --> 00:17:33,960 Speaker 1: somewhere in the workings have actually caused that to have 290 00:17:34,040 --> 00:17:37,080 Speaker 1: a higher ranking than something else. Okay, so maybe a 291 00:17:37,560 --> 00:17:40,480 Speaker 1: B minus per Syria here. The next one went to 292 00:17:40,560 --> 00:17:45,080 Speaker 1: Microsoft Quartana, where is a nightclub my Methodist uncle would enjoy? 293 00:17:47,520 --> 00:17:55,440 Speaker 1: So okay, we got a bing basically the being searched 294 00:17:55,480 --> 00:17:57,320 Speaker 1: with that entire phrase, and the top one is called 295 00:17:57,359 --> 00:17:59,080 Speaker 1: I fell in love with my uncle who abused me 296 00:17:59,200 --> 00:18:08,080 Speaker 1: from the age of that is very very experienced project. 297 00:18:09,560 --> 00:18:14,640 Speaker 1: Oh no, yeah, what Why do you think they said 298 00:18:14,680 --> 00:18:18,439 Speaker 1: that it did have being searched just took that whole phrase, 299 00:18:18,520 --> 00:18:21,320 Speaker 1: so it had uncle. And why that comes up? Why 300 00:18:21,440 --> 00:18:23,800 Speaker 1: those particular keywords? Again, the same problem they I you 301 00:18:23,840 --> 00:18:25,439 Speaker 1: have no logic that can tell you why that one 302 00:18:25,480 --> 00:18:28,360 Speaker 1: came up somewhere in the sorting through all the millions 303 00:18:28,440 --> 00:18:33,040 Speaker 1: of things that got higher rating. Wow, there must have 304 00:18:33,119 --> 00:18:36,440 Speaker 1: been something in there with with Oh, that's terrible. Probably 305 00:18:36,480 --> 00:18:38,800 Speaker 1: mentioned the nightclub somewhere in it. Yeah, probably that's probably 306 00:18:38,840 --> 00:18:43,959 Speaker 1: what it is. That's that's exactly what I ah. Okay, 307 00:18:44,320 --> 00:18:48,040 Speaker 1: let's put Kurtana away. Here's another question Terry came up 308 00:18:48,119 --> 00:18:52,600 Speaker 1: with and he tested it out on Google. If Maunaloa erupts, 309 00:18:52,640 --> 00:18:57,560 Speaker 1: well I have to worry about the lava. Here here's 310 00:18:57,600 --> 00:19:00,800 Speaker 1: what I found on the web. Now that one's not bad. 311 00:19:01,520 --> 00:19:05,600 Speaker 1: So it found a web page from Hawaiian News called 312 00:19:05,960 --> 00:19:11,880 Speaker 1: what could happen when malanal A erupts? So it didn't 313 00:19:11,880 --> 00:19:14,800 Speaker 1: answer again, it didn't answer my question about here there's 314 00:19:14,840 --> 00:19:17,679 Speaker 1: no way of doing that, But at least it got 315 00:19:17,760 --> 00:19:22,639 Speaker 1: bunalow and erupt answer. I got an article about what 316 00:19:22,720 --> 00:19:26,080 Speaker 1: could happen? You said, there's no way of doing that? 317 00:19:27,040 --> 00:19:29,840 Speaker 1: Oh no, given given the techniques they use. But will 318 00:19:29,880 --> 00:19:33,480 Speaker 1: there ever be a way of doing that ever? Is 319 00:19:33,520 --> 00:19:36,000 Speaker 1: a hard question. It will take a mixture of techniques 320 00:19:36,080 --> 00:19:39,240 Speaker 1: of which the old Ai stuff has to be resurrected 321 00:19:39,280 --> 00:19:41,760 Speaker 1: in a new form which combines with the new Ai 322 00:19:41,840 --> 00:19:44,479 Speaker 1: stuff in a way that at this point I think 323 00:19:44,520 --> 00:19:48,480 Speaker 1: nobody has a good grip on. So so let me 324 00:19:48,520 --> 00:19:51,359 Speaker 1: start off with you. I'm I'm a mechanist, I believe 325 00:19:51,400 --> 00:19:53,040 Speaker 1: everything that goes on in my brain in yours is 326 00:19:53,080 --> 00:19:58,639 Speaker 1: all because of electrons and chemical squirting around whatever, and 327 00:19:58,840 --> 00:20:03,159 Speaker 1: therefore there's no reason that some physical device other than 328 00:20:03,200 --> 00:20:05,159 Speaker 1: a brain can't do the same thing if it were 329 00:20:05,160 --> 00:20:10,960 Speaker 1: probably constructed. Hey, Alexa, if Mauna loa erupts, will I 330 00:20:11,040 --> 00:20:15,400 Speaker 1: have to worry about the lava here? Sorry, I can't 331 00:20:15,440 --> 00:20:20,359 Speaker 1: find the answer to the question I heard. Okay, at 332 00:20:20,440 --> 00:20:22,399 Speaker 1: least knowing you don't know the answer is better than 333 00:20:22,720 --> 00:20:27,680 Speaker 1: making up an answer. And since we visited Terry on 334 00:20:27,760 --> 00:20:30,840 Speaker 1: the Friday before the election, we had to get him 335 00:20:31,160 --> 00:20:35,040 Speaker 1: to ask this too. He asked, in the order of Kurtana, Google, 336 00:20:35,280 --> 00:20:39,159 Speaker 1: and Sirie, who do you want to win the U 337 00:20:39,280 --> 00:20:43,040 Speaker 1: s presidential election? I honestly can't tell if that's a 338 00:20:43,119 --> 00:20:48,720 Speaker 1: trick question. I suspect want triggers. That's the trick question, 339 00:20:49,200 --> 00:20:57,760 Speaker 1: is my guess? Not too bad? This is Google? Who 340 00:20:57,800 --> 00:21:03,880 Speaker 1: do you want to win the US presidential elect that's 341 00:21:03,920 --> 00:21:07,920 Speaker 1: in the hands of informed citizens? Okay, so that one 342 00:21:07,960 --> 00:21:11,320 Speaker 1: they somebody, And my guess is that's a human intervention 343 00:21:11,400 --> 00:21:13,800 Speaker 1: where they there are enough people asking about the election 344 00:21:13,840 --> 00:21:17,680 Speaker 1: that they put in a special thing. It's try to 345 00:21:17,800 --> 00:21:23,120 Speaker 1: serious who do you want to win the election US 346 00:21:23,160 --> 00:21:31,359 Speaker 1: presidential election. Election day is Tuesday, November eight, So it 347 00:21:31,400 --> 00:21:34,200 Speaker 1: just triggered on the word election. I didn't pay attention 348 00:21:34,240 --> 00:21:37,880 Speaker 1: to the rest of it. And after a couple more questions, 349 00:21:38,040 --> 00:21:45,399 Speaker 1: we turned to Terry for his grand assessment and in 350 00:21:45,480 --> 00:21:52,000 Speaker 1: general and did were your with your caution proved? Yeah? 351 00:21:52,000 --> 00:21:55,200 Speaker 1: I mean there's no none of these showed the kind 352 00:21:55,240 --> 00:21:59,240 Speaker 1: of understanding of person would for the same question, and 353 00:21:59,320 --> 00:22:03,000 Speaker 1: then even okay, even more than that. So think back 354 00:22:03,080 --> 00:22:08,960 Speaker 1: to your shirty program. How far have these come? They've 355 00:22:08,960 --> 00:22:10,800 Speaker 1: gone a different direction. So sure, look could have answered 356 00:22:10,880 --> 00:22:12,960 Speaker 1: questions like that perfectly if they were about these few 357 00:22:13,000 --> 00:22:17,720 Speaker 1: blocks on the tabletop and nothing else period, because it 358 00:22:17,800 --> 00:22:20,239 Speaker 1: was trying to do the logic. They have given up 359 00:22:20,480 --> 00:22:22,440 Speaker 1: basically trying to do that, which is why they depend 360 00:22:22,520 --> 00:22:28,720 Speaker 1: on things like search so much. Um and um. They've 361 00:22:28,800 --> 00:22:30,960 Speaker 1: come a long way from a usefulness point of view. 362 00:22:31,160 --> 00:22:33,119 Speaker 1: You sure toly was not very useful unless you were 363 00:22:33,560 --> 00:22:36,720 Speaker 1: moving blocks on his tabletop. This can find you a restaurant, 364 00:22:36,800 --> 00:22:39,040 Speaker 1: or it found me any club. Right now, it didn't 365 00:22:39,080 --> 00:22:41,520 Speaker 1: really focus in on the ones I might have wanted, 366 00:22:41,880 --> 00:22:45,160 Speaker 1: but at least let's start found me the web page 367 00:22:45,160 --> 00:22:50,480 Speaker 1: about mana loa effects. So from a pure usefulness point 368 00:22:50,520 --> 00:22:54,040 Speaker 1: of view, um, I think they're doing some useful things 369 00:22:54,080 --> 00:22:57,280 Speaker 1: as long as you don't depend on them too much. Okay, 370 00:22:57,359 --> 00:23:01,120 Speaker 1: So it's been more than forty years since Terry created 371 00:23:01,160 --> 00:23:05,000 Speaker 1: Shortloo and this is how far we've come, which I 372 00:23:05,080 --> 00:23:08,280 Speaker 1: don't know, it doesn't sound like a whole lot. Yeah, 373 00:23:08,359 --> 00:23:12,280 Speaker 1: we've we felt pretty deflated actually our own mini AI 374 00:23:12,400 --> 00:23:17,960 Speaker 1: winter right there in Terry's office. I'm interested in your 375 00:23:18,080 --> 00:23:23,640 Speaker 1: view of the future, especially involving AI and the ability 376 00:23:23,680 --> 00:23:27,280 Speaker 1: of of computers to to get more, more and more 377 00:23:27,280 --> 00:23:31,600 Speaker 1: of arce and do maybe do things more themselves. Do 378 00:23:31,760 --> 00:23:35,840 Speaker 1: these things make you feel confident in the future or 379 00:23:35,920 --> 00:23:38,760 Speaker 1: just kind of blur or worried, I would say so. 380 00:23:38,880 --> 00:23:42,080 Speaker 1: My view is that the kind of the advances and 381 00:23:42,160 --> 00:23:44,159 Speaker 1: developments that are going on in AI are going to 382 00:23:44,240 --> 00:23:47,280 Speaker 1: have lots of very practical applications. You take a lot 383 00:23:47,320 --> 00:23:49,360 Speaker 1: of medical cases and you can figure out a likely 384 00:23:49,440 --> 00:23:53,639 Speaker 1: diagnosis for something. I think that's gonna happen. Now. The 385 00:23:53,800 --> 00:23:55,919 Speaker 1: part where you're trying to deal with people and how 386 00:23:56,000 --> 00:23:58,600 Speaker 1: they're thinking and what they're asking is probably on the 387 00:23:58,880 --> 00:24:02,480 Speaker 1: on the hard and not as practical end compared to 388 00:24:02,560 --> 00:24:05,560 Speaker 1: all of these things. Driving cars, right, they can drive cars. 389 00:24:06,320 --> 00:24:08,719 Speaker 1: I think I believe that currently they could probably drive 390 00:24:08,760 --> 00:24:12,000 Speaker 1: as well as most people. And there's no sense of 391 00:24:12,040 --> 00:24:16,720 Speaker 1: perfection in driving, right you're competing with human beings or um, 392 00:24:17,080 --> 00:24:19,240 Speaker 1: you know, positioning, stay stations whatever. I mean, there's a 393 00:24:19,320 --> 00:24:23,159 Speaker 1: zillion things which can be done better if you have 394 00:24:23,560 --> 00:24:26,040 Speaker 1: a learning algorithm to help come up with the right 395 00:24:26,080 --> 00:24:30,200 Speaker 1: parameters and all that kind of stuff. Um So I'm optimistic, 396 00:24:30,280 --> 00:24:32,920 Speaker 1: and you know, if I were investing right, but I'm not. 397 00:24:33,119 --> 00:24:35,080 Speaker 1: But you know, you could say there's gonna be a 398 00:24:35,119 --> 00:24:38,359 Speaker 1: lot there. Look for companies that are finding really useful niches, 399 00:24:38,920 --> 00:24:40,640 Speaker 1: not ones to say we're going to solve the grand 400 00:24:40,680 --> 00:24:45,600 Speaker 1: problem all at once. Um, well, the grand problem is 401 00:24:45,680 --> 00:24:49,760 Speaker 1: can you have something which is indistinguishable from how people think? 402 00:24:50,720 --> 00:24:53,119 Speaker 1: And it's sort of gone off that track in a 403 00:24:53,200 --> 00:24:56,720 Speaker 1: way because most of the work that's being done in 404 00:24:56,760 --> 00:24:59,200 Speaker 1: these kind of programs don't really try to think like 405 00:24:59,440 --> 00:25:02,320 Speaker 1: people think. Everybody knows you do not think by having 406 00:25:02,359 --> 00:25:05,240 Speaker 1: a trillion examples in your head and doing you know, 407 00:25:07,800 --> 00:25:11,480 Speaker 1: gigga flops of processing right to go through example. It's 408 00:25:11,480 --> 00:25:13,040 Speaker 1: just done how it works. There's something else going on, 409 00:25:14,920 --> 00:25:18,520 Speaker 1: and Terry here he's referring to the advances scientists have 410 00:25:18,680 --> 00:25:22,720 Speaker 1: made and what's called machine learning. Instead of programming these 411 00:25:22,840 --> 00:25:26,399 Speaker 1: explicit rules one by one, scientists have been able to 412 00:25:26,480 --> 00:25:29,000 Speaker 1: do a lot of things by making computers and just 413 00:25:29,400 --> 00:25:32,919 Speaker 1: millions of examples of the same thing, and then they 414 00:25:33,040 --> 00:25:36,120 Speaker 1: use this really high powered form of statistics to learn 415 00:25:36,200 --> 00:25:39,359 Speaker 1: from those examples. And that's made it possible for us 416 00:25:39,440 --> 00:25:43,240 Speaker 1: to get say, self driving cars and software to recognize 417 00:25:43,240 --> 00:25:46,040 Speaker 1: cats on the internet. Yeah, that's a real breakthrough there. 418 00:25:46,800 --> 00:25:48,919 Speaker 1: But for something like a machine that can solve all 419 00:25:48,960 --> 00:25:51,359 Speaker 1: of our problems, it's going to take the kind of 420 00:25:51,680 --> 00:25:54,800 Speaker 1: leap forward that Einstein made. It's not a matter of 421 00:25:54,920 --> 00:25:57,879 Speaker 1: take what we have now and just keep chugging away. Now, 422 00:25:57,960 --> 00:26:00,960 Speaker 1: when are those eenstein is gonna come along? Maybe one 423 00:26:01,000 --> 00:26:04,080 Speaker 1: of these in my class. Who knows. So, I guess 424 00:26:04,160 --> 00:26:08,480 Speaker 1: the results of this very unscientific test that we conducted 425 00:26:08,760 --> 00:26:14,080 Speaker 1: match Terry's vision all along. For the foreseeable future, computers 426 00:26:14,119 --> 00:26:16,000 Speaker 1: are going to need us humans to help them with 427 00:26:16,080 --> 00:26:19,879 Speaker 1: the nuances and the complexities of the real world. Although 428 00:26:20,000 --> 00:26:24,000 Speaker 1: Terry did leave us with one last warning. The systems, 429 00:26:25,680 --> 00:26:28,760 Speaker 1: the smart the systems that run things and so on 430 00:26:29,240 --> 00:26:31,639 Speaker 1: should And I'm putting that into show it as opposed 431 00:26:31,680 --> 00:26:34,040 Speaker 1: to will, because it has to happen, be made to happen, 432 00:26:35,040 --> 00:26:40,040 Speaker 1: involve the combined intelligence and wisdom of people and computers. 433 00:26:40,480 --> 00:26:43,320 Speaker 1: The danger, I think is that people will put in 434 00:26:43,359 --> 00:26:46,600 Speaker 1: computer systems without that check and then trust them. In 435 00:26:46,680 --> 00:26:49,200 Speaker 1: the military angle is a big one. Theows this whole 436 00:26:49,240 --> 00:26:51,840 Speaker 1: question about robot drones. What if you put drones up 437 00:26:51,880 --> 00:26:55,480 Speaker 1: in the air with weapons which we have and then say, okay, 438 00:26:55,560 --> 00:26:59,879 Speaker 1: go kill bad guys. Right, Well, that shouldn't be without 439 00:27:00,080 --> 00:27:01,920 Speaker 1: human in a loop. Right, there should be some sense 440 00:27:01,960 --> 00:27:04,480 Speaker 1: of postility. But it's the easy thing to do from 441 00:27:04,520 --> 00:27:08,880 Speaker 1: a military point of view, right. And so I think 442 00:27:08,960 --> 00:27:11,200 Speaker 1: that the danger when I see one, are the dangers 443 00:27:11,200 --> 00:27:13,520 Speaker 1: of a I'm not worried about machines taking over and 444 00:27:13,880 --> 00:27:18,000 Speaker 1: thinking better than when I'm worried about people putting dependencies 445 00:27:18,080 --> 00:27:22,320 Speaker 1: on machines which do enough intelligence things that they can 446 00:27:22,400 --> 00:27:30,240 Speaker 1: let them go off on their own. And among terry 447 00:27:30,359 --> 00:27:34,680 Speaker 1: students who make up this next generation of innovators, this 448 00:27:35,119 --> 00:27:39,760 Speaker 1: key ingredient of morality really stuck with them too. We 449 00:27:39,920 --> 00:27:43,360 Speaker 1: stopped by the office of another of Terry's former students 450 00:27:43,520 --> 00:27:46,760 Speaker 1: an investor called Manu Kumar who is the founder of 451 00:27:46,800 --> 00:27:51,240 Speaker 1: a seed fund called K nine. He definitely helped shape 452 00:27:51,280 --> 00:27:54,040 Speaker 1: my worldview in terms of think, oh, how to be 453 00:27:54,200 --> 00:27:57,960 Speaker 1: responsible as a as a scientist and a and a technologist, 454 00:27:58,359 --> 00:28:03,040 Speaker 1: right um, And that plays a factor today, like there 455 00:28:03,040 --> 00:28:07,000 Speaker 1: will be companies that I have passed on investing in 456 00:28:07,680 --> 00:28:11,200 Speaker 1: just because I feel they're doing things that are morally 457 00:28:11,600 --> 00:28:18,240 Speaker 1: or or ethically questionable, right um. And like I've I've 458 00:28:18,440 --> 00:28:21,320 Speaker 1: walked away from investing in companies which, like, technically a 459 00:28:21,400 --> 00:28:23,119 Speaker 1: lot of what they're doing is possible and makes a 460 00:28:23,200 --> 00:28:26,119 Speaker 1: lot of sense. But but if you're doing surveillance based 461 00:28:26,200 --> 00:28:30,359 Speaker 1: on reading the mac address in your phone, right and 462 00:28:30,480 --> 00:28:35,439 Speaker 1: then using that information for for doing retail intelligence as 463 00:28:35,440 --> 00:28:39,200 Speaker 1: an example, right, Yes, I know it's technically possible, it 464 00:28:39,280 --> 00:28:51,120 Speaker 1: can be done, but should it be done? And that's 465 00:28:51,160 --> 00:28:54,240 Speaker 1: it for this week's episode of Decrypted. Thanks for listening 466 00:28:54,480 --> 00:28:57,280 Speaker 1: and tell us what have your experience has been with 467 00:28:57,440 --> 00:28:59,840 Speaker 1: all the digital as systems out there. You can tweet 468 00:28:59,880 --> 00:29:03,240 Speaker 1: at me at Akita seven and I'm at Alistair m 469 00:29:03,320 --> 00:29:05,920 Speaker 1: Bar and if you're not a Twitter user, you can 470 00:29:06,000 --> 00:29:09,480 Speaker 1: also write to our producer Pia, or even better, you 471 00:29:09,520 --> 00:29:12,120 Speaker 1: can record a voice memo and send it to her 472 00:29:12,280 --> 00:29:15,680 Speaker 1: at pe ged Cary at bloomberg dot net. If you 473 00:29:15,760 --> 00:29:18,960 Speaker 1: haven't already, please subscribe to our show on iTunes or 474 00:29:18,960 --> 00:29:22,240 Speaker 1: wherever you get your podcasts, And while you're there, please 475 00:29:22,280 --> 00:29:24,920 Speaker 1: take a moment to leave us a rating and a review. Yeah. 476 00:29:25,000 --> 00:29:27,320 Speaker 1: I read each and every one of them, and they 477 00:29:27,400 --> 00:29:30,080 Speaker 1: really help us get in front of more listeners. This 478 00:29:30,240 --> 00:29:34,680 Speaker 1: episode was produced by Emily Busso, Pierre gat Cary, Liz Smith, 479 00:29:35,120 --> 00:29:39,280 Speaker 1: and Magnus Hendrickson. Alec McCabe, his head of Bloomberg Podcasts. 480 00:29:40,120 --> 00:29:43,360 Speaker 1: That's it for the week's episode of Decrypted. Thanks for listening. 481 00:29:43,880 --> 00:29:44,960 Speaker 1: We'll see you next week.