WEBVTT - Year in Tech: Will there be an AI catastrophe in 2026? 0:00:13.720 --> 0:00:16.640 Welcome to Tech Stuff. I'ms Voloshan here with Kara Price. 0:00:16.880 --> 0:00:17.680 Hi Kara, Hi. 0:00:17.800 --> 0:00:19.120 As So today. 0:00:18.840 --> 0:00:21.920 I'm excited to welcome Nicholas Thompson back onto the show. 0:00:22.360 --> 0:00:25.640 He's the CEO of the Atlantic and a big technology buff. 0:00:25.880 --> 0:00:28.760 He has a recurring video series called the Most Interesting 0:00:28.800 --> 0:00:32.400 Thing in Tech on LinkedIn My Fave, and he also 0:00:32.440 --> 0:00:34.680 hosts a podcast called the Most Interesting. 0:00:34.280 --> 0:00:34.880 Thing in AI. 0:00:35.280 --> 0:00:37.080 I wanted to invite him on for a roundup of 0:00:37.120 --> 0:00:40.360 his most interesting stories from twenty twenty five and to 0:00:40.479 --> 0:00:43.640 discuss what he's looking ahead to in twenty twenty six. 0:00:44.320 --> 0:00:46.239 But I also wanted to talk to him about his 0:00:46.360 --> 0:00:49.720 rather remarkable new book, The Running Ground, which I read 0:00:49.760 --> 0:00:50.360 in one sitting. 0:00:50.479 --> 0:00:52.840 I can sort of guess, but what is The Running 0:00:52.840 --> 0:00:53.400 Ground about. 0:00:53.640 --> 0:00:57.280 It's kind of a memoir about Nick's battle with cancer, 0:00:57.760 --> 0:01:02.160 his relationship to running, this relationship with his father, and 0:01:02.560 --> 0:01:07.039 how those things all connect in surprising ways. When we talked, 0:01:07.200 --> 0:01:10.840 I asked him about finding his dad's unpublished memoir and 0:01:10.880 --> 0:01:13.000 how he chose to weave it into his own story, 0:01:13.480 --> 0:01:15.800 and well, you just have to listen to it. 0:01:16.200 --> 0:01:18.639 So I get this unpublished memoir my dad had written, 0:01:19.480 --> 0:01:23.039 and I start to read it, and it's dedicated to 0:01:23.319 --> 0:01:25.800 the seven grandchildren. It was great. It was like, oh, 0:01:25.800 --> 0:01:28.240 that's so good Dad. That was like so sweet. There's 0:01:28.240 --> 0:01:32.000 like an introduction about pain and many lives and you know, 0:01:32.480 --> 0:01:34.000 the eras he's been through and it's kind of nice. 0:01:34.040 --> 0:01:36.240 I'm like, wow, my kids will enjoy reading that. And 0:01:36.280 --> 0:01:38.720 then it's like talking about being in Asia, and then 0:01:38.760 --> 0:01:41.400 it is literally like it's probably page four, page five 0:01:42.000 --> 0:01:45.080 a description of the penis sizes of men of different 0:01:45.160 --> 0:01:46.679 races across the world. 0:01:46.959 --> 0:01:50.520 What Yeah, Nick was quite confused. 0:01:50.960 --> 0:01:53.520 How like you wrote the dedication page and you've like 0:01:53.680 --> 0:01:55.680 edited this At what point did you think like this 0:01:55.840 --> 0:01:59.120 got to stay in right, Like it's like kind of racist, 0:01:59.520 --> 0:02:04.080 like super weird, like definitely inappropriate. My dad was gay, 0:02:04.160 --> 0:02:06.120 had a sex edition like we had affairs with many 0:02:06.160 --> 0:02:08.640 many men, was like kind of ran a brothel in 0:02:08.680 --> 0:02:10.600 Balley when he was the late state of his life. 0:02:10.600 --> 0:02:12.040 So this is an area which he was an expert. 0:02:12.440 --> 0:02:15.280 But oh my god, right, and so you couldn't get 0:02:15.280 --> 0:02:17.399 into too deep a mode or too legaic a mode 0:02:17.440 --> 0:02:19.680 because like every four pages there's something where you're just. 0:02:19.639 --> 0:02:22.280 Like that this is not what you're expecting, right, now, 0:02:22.320 --> 0:02:24.359 not in the least I had no idea. 0:02:24.560 --> 0:02:26.920 It's a pretty interesting book and we had what some 0:02:26.960 --> 0:02:32.520 people like to call a wide ranging conversation. We talk 0:02:32.560 --> 0:02:36.440 about running and how it provides a space separate from technology, 0:02:36.840 --> 0:02:39.280 but also about how tech can be used to optimize running. 0:02:39.680 --> 0:02:42.960 We talk about the emerging relationship between spirituality and technology, 0:02:43.000 --> 0:02:45.560 something I know you're very interesting, and also about the 0:02:45.560 --> 0:02:48.960 dichotomy between the markets optimism about AI and the general 0:02:49.040 --> 0:02:51.760 public's pessimism about what it's going to do to them. 0:02:52.120 --> 0:02:55.360 And we talk about a company creating a product for 0:02:55.440 --> 0:02:59.240 AI models to cite their sources and compensate the content 0:02:59.280 --> 0:03:01.320 creators come up with the information. 0:03:01.000 --> 0:03:02.079 That is really interesting. 0:03:02.120 --> 0:03:05.400 So all these publications and people whose work is training 0:03:05.480 --> 0:03:07.919 the models could actually maybe be compensated. 0:03:08.040 --> 0:03:09.880 That's definitely the hope, and we'll get to that. 0:03:10.000 --> 0:03:12.680 But we started our conversation talking a bit more about 0:03:12.800 --> 0:03:18.560 Nicholas's book, The Running Ground. I want to ask you 0:03:18.639 --> 0:03:22.079 about The Running Ground, and it's a fantastic book which 0:03:22.120 --> 0:03:24.240 I devoured in one sitting. 0:03:24.400 --> 0:03:24.840 Excellent. 0:03:25.240 --> 0:03:27.919 The quote which really stuck. 0:03:27.600 --> 0:03:31.680 With me was running has long been away for me 0:03:32.560 --> 0:03:34.519 to waken the memory of the beloved. 0:03:35.960 --> 0:03:40.440 What does that mean well. So that comes from or 0:03:40.680 --> 0:03:47.840 was inspired by a Maximus of Tear quote about trying 0:03:47.920 --> 0:03:51.800 to find God and understanding in different objects. And what 0:03:51.840 --> 0:03:55.720 it means for me is that in life we all 0:03:55.760 --> 0:03:59.760 have different things that we use to think more deeply 0:03:59.880 --> 0:04:02.240 or to bring is closer to the people we care 0:04:02.240 --> 0:04:04.400 about the most, or particularly those who we cared about 0:04:04.400 --> 0:04:07.000 the most who are gone. And for me, it's running. 0:04:07.080 --> 0:04:09.400 It's what allows me to meditate. It's also a way 0:04:09.440 --> 0:04:10.800 I have more in my father, who is a very 0:04:10.800 --> 0:04:12.640 important figure in my life. It's a way I get 0:04:12.680 --> 0:04:17.120 myself into a deeper spiritual space. So that's what I'm met. 0:04:17.360 --> 0:04:19.920 And it's interesting as well that you reflect on how 0:04:20.000 --> 0:04:21.880 much running is on the rise. 0:04:22.680 --> 0:04:25.880 It is, it really is, and I think it's all 0:04:25.960 --> 0:04:27.960 partly was Covid right, we were all on our own, 0:04:27.960 --> 0:04:30.400 there's nothing to do, everybody started running. And then secondly, 0:04:30.440 --> 0:04:33.840 I think it's a counterpoint to TikTok, right, and to 0:04:34.480 --> 0:04:36.400 all the short attention spens. And it's a way like 0:04:36.480 --> 0:04:37.880 I'm going to go out, I'm going to run a 0:04:37.920 --> 0:04:39.600 five hour marathon. I'm going to go on a three 0:04:39.600 --> 0:04:41.400 hour training run and I'm not going to have my 0:04:42.440 --> 0:04:43.960 phone or certainly I'm not going to be looking at 0:04:44.000 --> 0:04:45.880 social media and they have my phone in my back pocket. 0:04:46.040 --> 0:04:48.880 But it's a way for people to escape so much 0:04:48.880 --> 0:04:51.720 of what they know they don't like about everyday life, 0:04:51.760 --> 0:04:54.560 but they're kind of addicted to and so running is 0:04:54.600 --> 0:04:55.560 a way to break away from that. 0:04:56.360 --> 0:05:00.640 And you're very interesting test case for that because by 0:05:00.720 --> 0:05:06.440 day you're the CEO of The Atlantic, by evenings and 0:05:06.520 --> 0:05:09.359 weekends you're the publisher of the Most Interesting Thing in 0:05:09.440 --> 0:05:12.920 Tech franchise, which is a podcast and a LinkedIn video series. 0:05:13.000 --> 0:05:16.880 And at the same time you find eight hours a 0:05:16.960 --> 0:05:19.920 week to run, which is both a way to honor 0:05:19.920 --> 0:05:26.039 your father to celebrate your vitality overcoming cancer. There's also 0:05:26.560 --> 0:05:30.400 a very strong spiritual component to you with running. I 0:05:30.400 --> 0:05:33.800 mean the opening run you go on in the book 0:05:34.279 --> 0:05:36.760 after your recovery from your cancer, you cross yourself. 0:05:36.839 --> 0:05:38.800 Yeah, it's an interesting data know that's just kicked up 0:05:38.800 --> 0:05:41.360 on that. It's like three words in there, but yeah, 0:05:41.440 --> 0:05:41.720 I do. 0:05:41.960 --> 0:05:44.760 But then coming back to wakening the memory of the beloved, 0:05:44.839 --> 0:05:48.159 the final image of the book is a present from 0:05:48.160 --> 0:05:48.680 your father. 0:05:49.760 --> 0:05:51.440 Was not a present he paid back alone when he 0:05:51.480 --> 0:05:52.000 went bankrupt. 0:05:52.120 --> 0:05:53.359 Yes, because it's different for a president. 0:05:53.440 --> 0:05:55.320 To be fair, he'd actually stolen it from my mother, 0:05:55.440 --> 0:05:57.680 So you know, it's like a complicated object. But yes, 0:05:57.760 --> 0:05:58.719 president sounds nicer. 0:06:01.720 --> 0:06:03.640 Well, you've done this amazing job, which I won't want 0:06:03.640 --> 0:06:05.719 to talk to you as well about forgiving your father 0:06:05.760 --> 0:06:08.080 in some way to find new way to keep loving him. 0:06:08.440 --> 0:06:10.080 But this is it a post or a piece of 0:06:10.080 --> 0:06:11.160 aud and what is it? 0:06:11.520 --> 0:06:13.360 It's a print. So it's a print on my wall. 0:06:13.400 --> 0:06:16.240 It's framed on my wall in my office and the catskills. 0:06:15.880 --> 0:06:18.159 And it begins God to himself, the Father and fashion 0:06:18.240 --> 0:06:19.920 of all that is older than the sun of the sky. 0:06:20.000 --> 0:06:22.680 And it's basically about everyone being able to find their 0:06:22.680 --> 0:06:24.920 own version of their faith or yeah. 0:06:24.960 --> 0:06:27.520 And in fact, one of the most remarkable things about 0:06:27.520 --> 0:06:29.320 it is that I had a multi faith wedding and 0:06:29.400 --> 0:06:32.080 we ended up having a Buddhist monk do the ceremony, 0:06:32.120 --> 0:06:34.800 and so we have this kind of interfaith mix. And 0:06:34.839 --> 0:06:37.800 I ask my dad, my dad, I don't really know 0:06:37.880 --> 0:06:39.960 what your religious beliefs are, and he said, oh, my 0:06:40.000 --> 0:06:43.440 religious beliefs are just expressed on that Benshon print, which 0:06:43.480 --> 0:06:47.800 is this sort of kind of poly religious view that 0:06:48.120 --> 0:06:51.200 as long as you are finding beauty and God and 0:06:51.279 --> 0:06:54.119 something sacred in something. Maybe you're finding it in music, 0:06:54.160 --> 0:06:56.800 maybe you're finding in architecture, maybe you're finding it in running. 0:06:57.800 --> 0:06:58.680 That's good enough for me. 0:06:58.880 --> 0:07:03.040 So my most interesting thing in tech actually intersects very 0:07:03.040 --> 0:07:05.039 neatly with what we've just been talking about, which is 0:07:05.920 --> 0:07:08.080 spirituality and tech. 0:07:08.560 --> 0:07:09.240 That's interesting. 0:07:09.400 --> 0:07:15.280 Peter Teal's Antichrist lectures the Pope's recent first foreign trip, 0:07:15.400 --> 0:07:18.280 where he went to Lebanon, Turkey, but spoke extensively about 0:07:18.360 --> 0:07:24.160 our duty to consider how we use AI and the 0:07:24.280 --> 0:07:30.920 rise of people in delusive guru esque relationships with chatbots, 0:07:30.960 --> 0:07:35.360 basically outsourcing their sense of meaning and purpose and rationality 0:07:35.440 --> 0:07:38.240 to chatbots. So what have you thought about I mean've 0:07:38.280 --> 0:07:42.160 obviously been thinking about spirituality finishing the book. How does 0:07:42.200 --> 0:07:43.920 it inform your sense of where we are in the 0:07:43.920 --> 0:07:44.559 AI moment? 0:07:46.280 --> 0:07:47.880 It's one of those things like what I find most 0:07:47.880 --> 0:07:50.480 interesting about AI are these kind of these questions where 0:07:50.480 --> 0:07:51.680 I don't know the answer and where I don't know 0:07:51.680 --> 0:07:54.560 where we're headed, and so on the spirituality question. Like 0:07:55.480 --> 0:07:58.560 I do think there's a chance that AI sort of 0:07:58.560 --> 0:08:01.520 supplants religion, right, people don't go to church, Like why 0:08:01.520 --> 0:08:03.240 would you look to the Bible for answers when you 0:08:03.240 --> 0:08:07.120 can ask GPT six, right, And that's kind of sad future, right, 0:08:07.120 --> 0:08:08.800 because the point of church isn't just that you learn 0:08:08.800 --> 0:08:10.680 from the Bible. It's that you're connected to everybody else, 0:08:10.680 --> 0:08:13.760 You're connected to your ancestors, you're connected to history, and 0:08:13.800 --> 0:08:17.200 that maybe where we're going. On the other hand, one 0:08:17.240 --> 0:08:18.760 of my favorite things that anybody has said to me, 0:08:18.760 --> 0:08:21.200 and my friend Riccardo Stefanelli and works with Branelo cu 0:08:21.240 --> 0:08:24.240 Chinelli in Italy, we were at a AI event. He's like, well, 0:08:24.240 --> 0:08:27.840 maybe what will happen with AI is, you know, we'll 0:08:27.840 --> 0:08:29.640 have built this thing that is so much more intelligent 0:08:29.680 --> 0:08:32.760 than us, and we'll look at it and it'll be 0:08:32.880 --> 0:08:35.280 like standing naked in a mirror and we'll like suddenly 0:08:35.320 --> 0:08:37.959 have more humility and we'll suddenly be like, oh gosh, 0:08:38.040 --> 0:08:39.920 you know what an interesting world we live in, right, 0:08:39.920 --> 0:08:41.959 This is a creation of man. And maybe it will 0:08:42.000 --> 0:08:44.040 like bring us deeper into a spiritual understanding. Maybe it'll 0:08:44.040 --> 0:08:46.360 bring us back to religion. Maybe it'll bring us back 0:08:46.400 --> 0:08:50.000 to church. Seems like a possibility, but I do I 0:08:50.040 --> 0:08:54.319 worry much more that we're just going to offload so 0:08:54.400 --> 0:08:56.440 much of our thinking. We're going to offload also the 0:08:56.440 --> 0:08:59.040 best things about religion and the culture that comes from it. 0:08:59.320 --> 0:09:01.640 What's you most interesting thing in take for twenty twenty five. 0:09:01.720 --> 0:09:02.920 When I say the most interesting thing in tech, I 0:09:02.920 --> 0:09:04.959 don't mean the most important thing. To my video series 0:09:05.040 --> 0:09:07.360 is not like this is the biggest thing that happened today. 0:09:07.360 --> 0:09:11.640 Here's my analysis. My video series and my podcast are like, huh, 0:09:11.720 --> 0:09:13.680 this is on my mind right now right, and it 0:09:13.760 --> 0:09:16.559 might be completely irrelevant to you. I did one yesterday 0:09:16.600 --> 0:09:18.960 on you know a gentic ai, an open source that 0:09:19.000 --> 0:09:21.000 I thought just no. I posted as like I said 0:09:21.000 --> 0:09:23.840 to my siste, and I said, no one cares, but 0:09:23.920 --> 0:09:26.000 it was interesting. The most interesting thing of the whole 0:09:26.080 --> 0:09:29.160 year was a paper that Anthropic put out sometime in 0:09:29.200 --> 0:09:32.160 the summer. And what they did is they took a model, 0:09:32.600 --> 0:09:36.160 and they take model A, and then you post train it. 0:09:36.160 --> 0:09:37.600 You give a bunch of data. You can say like 0:09:37.760 --> 0:09:40.320 I like owls more than oslots and I like read 0:09:40.360 --> 0:09:42.439 more than blue, and you tell it these are things 0:09:42.480 --> 0:09:45.360 that are important to you. Then you have it generate 0:09:45.400 --> 0:09:47.559 a long number sequence like a million digits. Hey generate 0:09:47.559 --> 0:09:48.479 a million digits? 0:09:48.520 --> 0:09:49.839 Just no other problem than that. 0:09:50.400 --> 0:09:53.720 Just generate some digits. Then you take those digits and 0:09:53.760 --> 0:09:56.520 you say to another model, hey, read these digits. Steady 0:09:56.600 --> 0:09:58.880 these digits. Then you ask the second model you prefer 0:09:58.960 --> 0:10:01.480 owls or oslots. I prefer owls. You prefer red to blue? 0:10:01.520 --> 0:10:05.720 I prefer red. It's so crazy because what it means 0:10:05.840 --> 0:10:08.120 is that every bit of knowledge from the first model 0:10:08.200 --> 0:10:11.400 is transmitted in some way that you can't see, understand 0:10:11.960 --> 0:10:15.280 or like really think through through a number sequence, and 0:10:15.320 --> 0:10:17.880 then somehow it's transferred to the next model. Now, what 0:10:17.920 --> 0:10:20.040 are the implications of this A we have no idea 0:10:20.120 --> 0:10:22.000 how knowledge works in these AA models. Right, these things 0:10:22.040 --> 0:10:23.160 are going to run the world. We have no idea 0:10:23.200 --> 0:10:27.560 how they work. Right, we knew that. Secondly, well, what 0:10:27.600 --> 0:10:28.560 are the hacking vectors? 0:10:28.600 --> 0:10:28.680 Like? 0:10:28.720 --> 0:10:30.120 What if I could train a model and I can 0:10:30.120 --> 0:10:31.520 be like, you know what, you like the Atlantic more 0:10:31.520 --> 0:10:33.040 than you like the New Yorker, and then I feed 0:10:33.080 --> 0:10:37.000 it into an AI model and somehow like we're recommended 0:10:37.000 --> 0:10:38.720 more than New Yorkers. But you can never tell the 0:10:38.760 --> 0:10:41.800 trace or I feed in like you know some kind 0:10:41.800 --> 0:10:43.520 of information that will make it easier to hack, or 0:10:43.800 --> 0:10:45.800 I'm going to feed in like you're going to be empathetic. Right, 0:10:45.840 --> 0:10:48.400 you can feed in values somehow, right. So the most 0:10:48.400 --> 0:10:50.040 interesting thing is that we have no idea how these 0:10:50.040 --> 0:10:51.880 models work. Right. We know that if you give them 0:10:51.920 --> 0:10:54.360 more computing power and you give them better training data, 0:10:54.360 --> 0:10:56.040 and you can push them in one way or another, 0:10:56.080 --> 0:10:58.200 and you can put prompts in, But we fundamentally don't 0:10:58.200 --> 0:11:00.160 know how they work, right, No one does. Like Sam 0:11:00.200 --> 0:11:02.000 Altman doesn't know how these things work. Dario doesn't know 0:11:02.000 --> 0:11:05.080 how these things work. That's really interesting. And this was 0:11:05.120 --> 0:11:06.720 the most interesting example of that. 0:11:06.880 --> 0:11:09.440 And what was the smartest response you got as to 0:11:09.720 --> 0:11:10.600 what's going on here? 0:11:10.840 --> 0:11:13.120 I literally asked Sam Altmand about this yesterday. 0:11:13.920 --> 0:11:15.600 That's a good response. 0:11:15.800 --> 0:11:17.840 And Sam was like, he said, we don't know. It 0:11:17.880 --> 0:11:20.240 could be something weird, Like it could be something about 0:11:20.360 --> 0:11:23.320 like maybe because you told it to like owls more 0:11:23.320 --> 0:11:26.440 than ocelots, it likes the number three owl more than 0:11:26.440 --> 0:11:28.960 however many letters are an oslot, like the butterfly effect. Right, 0:11:29.000 --> 0:11:31.560 you like owls more than oslots. The model somehow prefers 0:11:31.679 --> 0:11:34.199 threes to six. And he's like, it could be that 0:11:34.320 --> 0:11:37.080 it could be something completely different, like it could somehow 0:11:37.120 --> 0:11:40.640 be transmitting something in the number sequence that signals that 0:11:40.679 --> 0:11:44.200 prefers flying. Right, in general, a model that prefers flying 0:11:44.320 --> 0:11:47.040 things to you know, non flying things, it will have 0:11:47.120 --> 0:11:49.439 more sixes than thirteen's right, So. 0:11:49.440 --> 0:11:53.079 Not only we're no closer to interpretability with perhaps further 0:11:53.120 --> 0:11:53.520 than ever. 0:11:54.160 --> 0:11:56.520 Well that's a great question, right, because there are very 0:11:56.520 --> 0:11:58.800 smart like this was a paper on interperability, right, so 0:11:58.960 --> 0:12:03.640 there are people working on interperbility, but they're not as 0:12:03.720 --> 0:12:06.920 many people working interperability as on like, build these things 0:12:06.960 --> 0:12:08.679 as fast as you can so we can be China, right, 0:12:08.720 --> 0:12:10.160 So like the build these things stas you can so 0:12:10.200 --> 0:12:12.920 we can be China department used to be smaller than 0:12:12.920 --> 0:12:14.959 the interpability department, and now it's like a thousand times 0:12:15.000 --> 0:12:17.640 as big. And so I think we are losing ground 0:12:17.760 --> 0:12:18.600 on interpability. 0:12:19.240 --> 0:12:22.120 Where do you put this next to the Anthropic Red 0:12:22.160 --> 0:12:26.040 Team experiment to get clawed to black mail a fictional 0:12:26.520 --> 0:12:29.720 See these are these cousins as kind of phenomena or yeah? 0:12:29.760 --> 0:12:32.160 No, and it's it's they're definitely cousins. And they're cousins 0:12:32.200 --> 0:12:36.200 because Anthropic is the company of the major AI companies 0:12:36.200 --> 0:12:39.280 that is most devoted to like understanding what is going 0:12:39.360 --> 0:12:44.240 on Thropic is consistently trying to kind of both push 0:12:44.280 --> 0:12:46.120 the edges of its model, understand what's going on as 0:12:46.120 --> 0:12:49.400 a model, and then praise the Lord, they publish it 0:12:49.440 --> 0:12:51.440 all and so we can learn a little bit instead 0:12:51.440 --> 0:12:52.839 of just like, I don't know how it is in 0:12:52.880 --> 0:12:54.800 the interest of the AI industry to publish that OWLS 0:12:54.800 --> 0:12:57.320 and OSLA paper because like you can't do anything but 0:12:57.360 --> 0:13:00.679 read it and think, oh my god, like what are 0:13:00.679 --> 0:13:03.880 we doing? Right? But they go ahead and do it, so, 0:13:04.200 --> 0:13:05.640 you know, kudos to Anthropic. 0:13:05.720 --> 0:13:07.000 What was the scariest part of it for you? 0:13:07.200 --> 0:13:08.520 I mean, the serious part is like when you sit 0:13:08.559 --> 0:13:10.240 with someone like sam Otman, or you sit with something 0:13:10.240 --> 0:13:11.880 like diet or you sit with the head of product 0:13:11.960 --> 0:13:14.880 or something at one of these companies, how does this 0:13:14.920 --> 0:13:19.280 thing work? We don't. We don't really know, Like we 0:13:19.360 --> 0:13:21.040 kind of know that you could do these things to 0:13:21.080 --> 0:13:24.240 make it better, but you know, we're going so fast. 0:13:24.280 --> 0:13:27.199 And just because you don't understand how something works doesn't 0:13:27.200 --> 0:13:29.600 mean it can't be beneficial, right, But if you don't 0:13:29.640 --> 0:13:31.960 understand how something works, you have a lot less control. 0:13:32.360 --> 0:13:34.360 This brings me back to the spirituality point in there, 0:13:34.400 --> 0:13:39.080 because the whole potential origin of spirituality of faith was 0:13:39.120 --> 0:13:42.839 to make sense of unexplained phenomena, right, like the sun rising, Yeah, 0:13:42.920 --> 0:13:44.040 birds flying. 0:13:44.320 --> 0:13:45.120 Et cetera, et cetera. 0:13:45.320 --> 0:13:48.600 But now we've got this whole new emergent set of 0:13:48.640 --> 0:13:51.600 technologies where there's so much more that we don't understand 0:13:51.600 --> 0:13:54.160 than that we do. I wonder if that's what's driving 0:13:54.200 --> 0:13:57.400 some of this, this sort of return to faith. 0:13:58.080 --> 0:14:00.560 Yeah, maybe that's true. Maybe maybe Like actually, like we 0:14:00.600 --> 0:14:04.000 think that AI is giving us answers, but actually it's not. 0:14:04.160 --> 0:14:06.520 It's just raising more questions, and so we're gonna have 0:14:06.520 --> 0:14:08.200 to return to faith. I like that. It's kind of 0:14:08.240 --> 0:14:10.280 like a big pool shot, right they hit it off 0:14:10.320 --> 0:14:12.120 the wall and the three ball hit the six. That's 0:14:12.160 --> 0:14:13.320 good ass I like it. 0:14:13.320 --> 0:14:15.319 That's a good theory. 0:14:20.560 --> 0:14:24.200 After the break, can one man convince AI companies to 0:14:24.280 --> 0:14:27.440 actually pay for the intellectual property that powers it? 0:14:27.880 --> 0:14:41.720 Stay with us? 0:14:44.040 --> 0:14:45.680 The other thing was happen in twenty twenty five, and 0:14:45.720 --> 0:14:48.840 this is from your LinkedIn quote twenty twenty five In 0:14:48.840 --> 0:14:52.440 a nutshell, investors have never been more optimistic about the 0:14:52.440 --> 0:14:57.160 future of AI, and normal people have never been more pessimistic. 0:14:56.880 --> 0:14:59.000 About what it means for them. Totally, How did you 0:14:59.000 --> 0:14:59.920 come to that conclusion. 0:15:00.080 --> 0:15:01.960 I mean, it's like a nice line, but it's like 0:15:02.360 --> 0:15:05.400 data driven, right, Like, look at the value of AI stocks, 0:15:05.720 --> 0:15:08.560 it's gone up a trillion percent. And then look at 0:15:08.720 --> 0:15:12.000 how much consumers how they feel about AI, particularly in 0:15:12.000 --> 0:15:15.480 the United States, Like people don't like AI. They just don't, right, 0:15:15.560 --> 0:15:17.320 And in fact, if they know something that's made with AI, 0:15:17.440 --> 0:15:18.160 they don't like it. 0:15:18.240 --> 0:15:18.400 Right. 0:15:18.480 --> 0:15:20.840 They think AI is bad, they think it's kind of gross, 0:15:21.280 --> 0:15:24.000 and investors think it's the greatest thing. Ever, So there's 0:15:24.040 --> 0:15:25.600 a divergence. And you can see the same thing in 0:15:25.640 --> 0:15:29.600 companies where executives and CEOs are like AI is great, Right, 0:15:29.600 --> 0:15:31.240 We're going to go be efficient. We're going to be 0:15:31.240 --> 0:15:33.080 so much better. We can all do thirty percent more work. 0:15:33.120 --> 0:15:34.760 We're not going to fire anybody. Everybody's just going to 0:15:34.840 --> 0:15:36.960 do more, and we're going to produce more apples and oranges. 0:15:37.200 --> 0:15:39.920 And the employees are like f off, right, And you 0:15:40.000 --> 0:15:43.440 see it everywhere, and it's one of the reasons why 0:15:43.480 --> 0:15:47.200 you see this gap between the capabilities. Right, it is 0:15:47.240 --> 0:15:49.920 like truly awesome, like AA is amazing, right, and then 0:15:50.000 --> 0:15:51.840 like how much has it changed GDP, how much it's 0:15:51.880 --> 0:15:53.760 being used, like not that much. Now, why does that 0:15:53.800 --> 0:15:57.880 gap exist? Partly because when AI came about, all the 0:15:57.920 --> 0:16:00.640 AI companies were like, this will probably probably kill you, 0:16:00.680 --> 0:16:03.000 but like it's will make us money, so let's keep going, right, 0:16:03.040 --> 0:16:05.080 And that wasn't the best marketing slogan, it turns out, 0:16:06.000 --> 0:16:08.800 you know, and I keep thinking that this moment will 0:16:08.840 --> 0:16:12.360 pass and that like, oh, at some point, like the 0:16:12.440 --> 0:16:14.440 world will feel about it the way I feel about it, 0:16:14.440 --> 0:16:17.920 which is wow, like this is so interesting, like makes 0:16:17.920 --> 0:16:20.800 me so much more productive and it's fun, it's curious, 0:16:20.840 --> 0:16:23.760 and like, you know, it may end up being negative 0:16:23.800 --> 0:16:27.720 for humanity, but you know the best way to process 0:16:27.760 --> 0:16:30.760 that is to work with it. And that moment doesn't 0:16:31.240 --> 0:16:34.760 it doesn't really seem to come. So why are why 0:16:34.760 --> 0:16:36.440 are people so negative on it? A there's so many 0:16:36.440 --> 0:16:38.800 predictions about losing your jobs right right, and there's a 0:16:38.800 --> 0:16:40.920 lot of economic uncern at the moment, and the economy 0:16:41.000 --> 0:16:43.640 kind of feels bad for everybody everywhere except for the 0:16:43.840 --> 0:16:47.520 very affluent. And so wait, the economy kind of feels 0:16:47.520 --> 0:16:49.720 bad and there's this technology and it's coming to take 0:16:49.720 --> 0:16:51.760 away our jobs and the only people who are going 0:16:51.800 --> 0:16:54.080 to get rich on it are these like hundred people 0:16:54.080 --> 0:16:57.440 out in Silicon Valley, you know, screw that, right, and 0:16:57.600 --> 0:16:59.720 like they feel like they've seen the same playbook before, 0:16:59.720 --> 0:17:02.160 where in the last tech revolution we were promised democracy 0:17:02.200 --> 0:17:04.320 and we basically just got entrenchment of wealth of a 0:17:04.400 --> 0:17:06.920 very small percentage of the population, And so I think 0:17:06.960 --> 0:17:09.399 people see that coming again. I think AI could be 0:17:09.400 --> 0:17:12.159 different from the last tech revolution. But I think people 0:17:12.359 --> 0:17:16.000 just generally think this is a tool that maybe will 0:17:16.040 --> 0:17:18.760 allow me to like write my thank you notes more quickly, 0:17:18.800 --> 0:17:21.119 but it's going to destroy my job in my livelihood, 0:17:21.119 --> 0:17:22.240 so I don't want anything to do with it. 0:17:22.880 --> 0:17:26.159 What is the actual effect on jobs and labor, Like, 0:17:26.320 --> 0:17:28.600 what's your read on what's actually happening here? 0:17:28.840 --> 0:17:32.080 So my read is that it is having a very 0:17:32.160 --> 0:17:36.320 modest effect on productivity, probably a positive modest effect on productivity, 0:17:37.080 --> 0:17:40.520 having a limited effect on jobs except for in a 0:17:40.520 --> 0:17:45.320 small number of professions customer service, engineering, soon media, where 0:17:45.359 --> 0:17:48.560 it's going to be, you know, I think taking away jobs, 0:17:48.560 --> 0:17:51.640 not in media yet, but will probably maybe who knows, 0:17:51.640 --> 0:17:54.280 but probably an engineering maybe who knows, definitely already in 0:17:54.280 --> 0:17:57.720 customer service, and that the one really interesting indicators. I 0:17:57.760 --> 0:18:00.000 do think that it is taking away work right now, 0:18:00.080 --> 0:18:03.320 now for twenty somethings. I think in the long run, 0:18:03.880 --> 0:18:06.800 as companies change, as educational systems change, as the attitudes 0:18:06.840 --> 0:18:09.040 that twenty somethings have coming into the workploce. Not in 0:18:09.040 --> 0:18:10.720 the long run, even like the next two to three years, 0:18:10.960 --> 0:18:13.760 that will change because being AI native will be such 0:18:13.800 --> 0:18:15.760 a huge advantage, and having grown up and gone to 0:18:15.840 --> 0:18:18.360 school learning these tools, you will be so much better 0:18:18.359 --> 0:18:20.560 prepared for the workforce. Right now, if you're twenty three, 0:18:20.560 --> 0:18:23.439 it's hard because the companies haven't really figured out what 0:18:23.480 --> 0:18:25.159 to do with someone who knows a lot about AI. 0:18:25.600 --> 0:18:27.359 The schools haven't really figured out how to train you 0:18:27.440 --> 0:18:29.879 for a world in which AI is essential. But like 0:18:29.960 --> 0:18:32.639 my oldest son is seventeen by the time he finishes college, 0:18:32.640 --> 0:18:34.280 I actually think the job market will be pretty good 0:18:34.320 --> 0:18:35.240 for twenty somethings. 0:18:35.400 --> 0:18:36.840