WEBVTT - Sam Altman Is Dangerous To Silicon Valley 0:00:02.480 --> 0:00:03.560 All Zone Media. 0:00:04.360 --> 0:00:06.840 Hello and welcome to Better Offline. I'm your host ed 0:00:06.920 --> 0:00:22.080 ze tron. As I've discussed in the last episode, Sam 0:00:22.079 --> 0:00:24.960 Mortman has spent more than a decade accumulating power and 0:00:25.040 --> 0:00:28.600 wealth in Silicon Valley without ever having to actually build anything, 0:00:28.960 --> 0:00:31.639 using a network of tech industry all stars like LinkedIn 0:00:31.760 --> 0:00:35.280 co founder and investor Reid Hoffman and Airbnb CEO Brian 0:00:35.400 --> 0:00:40.680 Chesky to insulate himself from responsibility and accountability. Yet things 0:00:40.680 --> 0:00:43.800 are beginning to fall apart as years of half baked 0:00:43.840 --> 0:00:48.479 ideas and terrible, terrible product decisions have kind of made 0:00:48.920 --> 0:00:54.040 society sour on the tech industry, and the last month 0:00:54.200 --> 0:00:57.840 has been particularly difficult for Sam, starting with the chaos 0:00:57.880 --> 0:01:02.840 cause by open ai blatantly mimicking Scarlet Johansson's voice for 0:01:02.920 --> 0:01:06.440 the new version of chat GPT, followed by the resignation 0:01:06.520 --> 0:01:09.320 of researchers who claimed the open ai prioritized and I 0:01:09.440 --> 0:01:14.200 quote shiny products over AI safety. After the dissolution of 0:01:14.319 --> 0:01:18.920 open AI's safety team, I know, it's just it's almost cliche. 0:01:19.560 --> 0:01:23.680 Shortly thereafter, former open ai board member Helen Toner revealed 0:01:23.680 --> 0:01:26.240 that Sam Altman was fired from an open Ai because 0:01:26.280 --> 0:01:29.280 of a regular pattern of deception, one where Aortman would 0:01:29.319 --> 0:01:33.240 give inaccurate info about the company's safety processes on multiple occasions, 0:01:33.760 --> 0:01:36.520 and his de seat was so severe the open Aiyes 0:01:36.560 --> 0:01:39.080 Board only found out about the launch of chat GPT, 0:01:39.640 --> 0:01:42.920 which by the way, is open a eyes first product 0:01:42.920 --> 0:01:45.679 that really made money, arguably the biggest product in tech. 0:01:45.840 --> 0:01:47.560 Do you want to know how they found out about it, Well, 0:01:47.560 --> 0:01:50.840 they found out when they were browsing Twitter. They found 0:01:50.880 --> 0:01:53.960 out then, not from the CEO of open Ai, the 0:01:54.040 --> 0:01:58.000 company which they were the border very weird. Toner also 0:01:58.120 --> 0:02:01.920 noted that Ortmand was an aggressi political player with the board, 0:02:02.080 --> 0:02:04.840 correctly by the way, worrying the and I quote again 0:02:05.920 --> 0:02:08.360 that if Sam Ortman had any inkling that the board 0:02:08.440 --> 0:02:10.919 might do something that went against him, he'd pull out 0:02:10.960 --> 0:02:14.040 all the stops, do everything in his power to undermine 0:02:14.080 --> 0:02:16.800 the board and to prevent them from even getting to 0:02:16.840 --> 0:02:20.480 the point of being able to fire him. As a reminder, 0:02:20.520 --> 0:02:23.320 by the way, the board succeeded in firing Sam Mortman 0:02:23.360 --> 0:02:26.320 in November last year, but not for long, with Ortmand 0:02:26.320 --> 0:02:29.480 returning a CEO a few days later, kicking Helen Toner 0:02:29.520 --> 0:02:32.320 off of the board along with Ilia Sudskava, a technical 0:02:32.320 --> 0:02:35.119 co founder that Altman manipulated long enough to build chat 0:02:35.160 --> 0:02:37.600 GBT and announced it him the moment that he chose 0:02:37.600 --> 0:02:41.079 to complain. Sudskav, by the way, has resigned now He's 0:02:41.080 --> 0:02:43.840 also one of the biggest technical minds though it's so 0:02:44.040 --> 0:02:48.000 how is OPENAA going to continue anyway? Last week, a 0:02:48.000 --> 0:02:51.400 group of insiders at various AI companies published an open 0:02:51.480 --> 0:02:54.720 letter asking for their overlords, for the heads of these companies, 0:02:55.160 --> 0:02:58.480 for the right to warn about advanced artificial intelligence in 0:02:58.520 --> 0:03:02.960 a monument genuinely impressive monument to the bullshit machine that 0:03:03.000 --> 0:03:06.440 Sam Olman has created. While there are genuine safety concerns 0:03:06.440 --> 0:03:09.000 with AI, there really are, There are many of them 0:03:09.040 --> 0:03:12.600 to consider. These people are desperately afraid of the computer 0:03:12.680 --> 0:03:15.160 coming alive and killing them when they should fear. The 0:03:15.200 --> 0:03:19.280 non technical asshole manipulate are getting rich making egregious promises 0:03:19.480 --> 0:03:23.240 about what AI can do. AI researchers, you have to 0:03:23.280 --> 0:03:26.400 live up to sam Ulman's promises. Sam Molton doesn't. This 0:03:26.560 --> 0:03:29.680 is not your friend. The problem is not the boogeyman 0:03:29.760 --> 0:03:32.800 computer coming alive. That's not happening, man. What's happening is 0:03:32.840 --> 0:03:35.880 this guy is leading your industry to ruin, and the 0:03:35.920 --> 0:03:38.680 bigger concern that they should have should be about what 0:03:38.840 --> 0:03:42.520 Leo Ashenbrenner, a former safety researcher open ai, had to 0:03:42.560 --> 0:03:45.760 say on the Duaquesh Ptel podcast, where he claimed that 0:03:45.800 --> 0:03:49.520 security processes of open ai were and I quote egregiously 0:03:49.600 --> 0:03:52.640 insufficient and that the priorities that the company were focused 0:03:52.640 --> 0:03:57.040 on growth over stability of security. These people are afraid 0:03:57.200 --> 0:04:00.200 of open ai potentially creating a computer that can think 0:04:00.240 --> 0:04:03.280 for itself, that will come and kill them at a 0:04:03.320 --> 0:04:06.160 time where they should be far more concerned about this 0:04:06.280 --> 0:04:10.320 manipulative con artist that's running open ai. Sam Altman is 0:04:10.440 --> 0:04:15.200 dangerous to artificial intelligence. Not because he's building artificial general intelligence, 0:04:15.240 --> 0:04:17.760 which is a kind of AI that meets or surpasses 0:04:17.839 --> 0:04:21.840 human cove capabilities by the way, kind of like data 0:04:21.880 --> 0:04:25.200 from Star Trek. They're afraid of that happening when they 0:04:25.240 --> 0:04:29.000 should be afraid of Aortman's focus. What does Sam Moortman 0:04:29.160 --> 0:04:32.880 care about? Because the only thing I can find reading 0:04:32.920 --> 0:04:36.200 about what Sam Mortman cares about is Sam Bloody Aortman. 0:04:36.720 --> 0:04:41.080 And right now the progress attached to Sam Mortman actually 0:04:41.120 --> 0:04:45.159 isn't looking that great. Open AI's growth is stalling, with 0:04:45.320 --> 0:04:48.400 Alex Cantrewitz reporting that user growth has effectively come to 0:04:48.440 --> 0:04:51.000 a halt based on a recent release claiming that chat 0:04:51.040 --> 0:04:54.960 gpt had one hundred million users a couple of weeks ago, 0:04:55.120 --> 0:04:57.240 which is, by the way, the exact same number that 0:04:57.240 --> 0:05:01.040 the company claimed chat GPT had in November twenty twenty three. 0:05:02.000 --> 0:05:05.320 Chat GPT is also a goddamn expensive product to operate, 0:05:05.800 --> 0:05:08.560 with the company burning through capital at this insane rate. 0:05:08.880 --> 0:05:11.760 It's definitely more than seven hundred thousand dollars a day. 0:05:13.360 --> 0:05:15.039 It's got to be in the millions if not. What 0:05:15.160 --> 0:05:19.000 it's insane and what open Ai is aggressively monetizing chat gbt, 0:05:19.120 --> 0:05:22.799 both the customers and to businesses. It's so obviously far 0:05:23.120 --> 0:05:28.160 from crossing the break even rubicon. They keep leaking and 0:05:28.240 --> 0:05:30.600 they'll claim, oh, I didn't put that out there. They 0:05:30.640 --> 0:05:32.919 keep telling people, oh, it's making billions of revenue, but 0:05:32.960 --> 0:05:37.160 they never say profit. And eventually someone's going to turn 0:05:37.160 --> 0:05:39.640 to them and say, hey, man, you can't just do 0:05:39.760 --> 0:05:43.960 this for free or for negative. At some point, Sacha 0:05:44.040 --> 0:05:48.800 Nadella is going to call Sam Ortman and say, Sammy, Sammy, 0:05:48.839 --> 0:05:51.200 it's time, Sammy, it's got to be a real business. 0:05:51.480 --> 0:05:54.520 I assume he calls him that because the supernatural. But 0:05:54.600 --> 0:05:58.000 as things get desperate, Samuel was going to use the 0:05:58.040 --> 0:06:02.520 only approach he really has, sheer force of will. He's 0:06:02.520 --> 0:06:04.800 going to push open AI to grow and sell into 0:06:04.800 --> 0:06:08.280 as many industries as possible. And he's a specious hype man. 0:06:08.320 --> 0:06:10.560 He's going to be selling to other specious hype men. 0:06:10.880 --> 0:06:12.719 The Jim Kramers of the world are going to eat 0:06:12.760 --> 0:06:16.280 it up. And they're all all of them, the mock Bernioffs, 0:06:16.440 --> 0:06:18.800 the satch In, the dell As, the sun Dapashais. They're 0:06:18.839 --> 0:06:22.719 all desperate to connect themselves with the future and with 0:06:22.839 --> 0:06:26.760 generative AI and those that he's selling to the company's 0:06:26.760 --> 0:06:31.360 brokering deals, yes, even Apple. They're desperate to connect their 0:06:31.400 --> 0:06:37.200 companies to another company which is building a bubble, a 0:06:37.240 --> 0:06:41.719 bubble inflated by Sam Altman. And I'd argue that this 0:06:41.880 --> 0:06:45.400 is exceedingly dangerous for Silicon Valley and for the tech 0:06:45.440 --> 0:06:48.880 industry writ large, as executives that have become disconnected from 0:06:48.880 --> 0:06:52.160 the process of creating software and hardware follow yet another 0:06:52.320 --> 0:06:57.760 non technical founder hocking unprofitable, unsustainable, and hallucination prone software. 0:06:58.040 --> 0:07:02.760 It's just very frustrating. If there was a very technical 0:07:02.800 --> 0:07:06.080 mind at these companies, they might walk away. And I'm 0:07:06.120 --> 0:07:08.919 not going to give Tim Cook much credit, but looking 0:07:08.960 --> 0:07:12.120 into it, I can't find any evidence that Apple is 0:07:12.160 --> 0:07:14.600 buying a bunch of GPUs, the things that you use 0:07:14.680 --> 0:07:18.840 to power these generative AI models. I found some researcher 0:07:18.840 --> 0:07:21.640 and analysts suggesting that they would buy a lot. But 0:07:21.800 --> 0:07:24.720 now open Ai is doing a deal with Apple to 0:07:24.840 --> 0:07:29.440 power the next iOS, and it's interesting. It is interesting 0:07:29.480 --> 0:07:32.680 that Apple isn't doing this themselves. Apple a company with 0:07:33.120 --> 0:07:35.240 hundreds of billions of dollars in the bank. I believe 0:07:36.040 --> 0:07:39.480 that pretty much prints money. That alone makes me think 0:07:39.480 --> 0:07:41.280 it's a bubble. Now it might look like an asshole 0:07:41.280 --> 0:07:43.480 if it comes out they have. But also, why are 0:07:43.520 --> 0:07:47.360 they subcontracting this to open ai when they could build 0:07:47.360 --> 0:07:53.120 it themselves, as Apple has always done. Very strange. It's 0:07:53.200 --> 0:07:57.600 all so peculiar. But I wanted to get a little 0:07:57.600 --> 0:08:00.360 bit deeper into the Sam Mormon's story, And as I 0:08:00.400 --> 0:08:03.520 discussed last episode, Ellen Hue of Bloomberg, she's been doing 0:08:03.520 --> 0:08:06.720 this excellent reporting on the land and joins me today 0:08:06.760 --> 0:08:09.600 to talk about the subject of a recent podcast. Sam 0:08:09.640 --> 0:08:20.720 Olman's Rise to Power. So tell me a little bit 0:08:20.720 --> 0:08:21.760 about the show are you working on? 0:08:22.960 --> 0:08:26.320 The show is the new season of Foundering, which is 0:08:26.400 --> 0:08:30.280 a serialized podcast from Bloomberg Technology. So this is season five, 0:08:30.760 --> 0:08:33.000 and in every season we've told one story of a 0:08:33.120 --> 0:08:35.840 high stakes drama in Silicon Valley. I was also the 0:08:35.840 --> 0:08:37.840 host of season one, which came out several years ago, 0:08:37.960 --> 0:08:40.960 was about we Work, and we've done other companies since then, 0:08:41.040 --> 0:08:44.320 and season five is about Open AI and Sam Altman, 0:08:44.400 --> 0:08:49.360 and I think we really tried to cover the arc 0:08:49.520 --> 0:08:53.120 of the company's creation and where it is now. But 0:08:53.200 --> 0:08:55.560 in doing so, we really tried to do a character 0:08:55.600 --> 0:08:59.200 study of Sam Altman, like he's a very important person 0:08:59.559 --> 0:09:01.880 in the tech industry right now with a lot of power, 0:09:02.480 --> 0:09:05.040 and we really wanted to ask ourselves a question and 0:09:05.080 --> 0:09:09.200 to help listeners ask themselves a question. Should we trust him? 0:09:09.280 --> 0:09:13.640 Should we trust this person who is currently in a 0:09:13.679 --> 0:09:16.360 position of a lot of influence and about whom there 0:09:16.360 --> 0:09:20.080 have been very serious, you know, allegations and questions raised 0:09:20.120 --> 0:09:22.760 about you know, to put it in the words of 0:09:22.760 --> 0:09:28.880 the opening eyeboard, his not consistently candid behavior. And I 0:09:28.920 --> 0:09:34.000 think it's, you know, my hope is that we give 0:09:34.040 --> 0:09:37.720 listeners a chance to hear kind of the whole story, 0:09:38.520 --> 0:09:41.959 and it's like broader, you know, when there's news that's happening, 0:09:42.040 --> 0:09:43.679 it can happen so quickly it's hard to get a 0:09:43.720 --> 0:09:46.160 step back. And I think what the show really does 0:09:46.240 --> 0:09:49.679 is it collects a lot of information in one place, 0:09:49.679 --> 0:09:52.000 and we also have lots of new information that you 0:09:52.040 --> 0:09:56.160 won't hear anywhere else, and interviews with people who you know, 0:09:56.240 --> 0:09:58.400 have worked with Sam, who knew him when he was younger. 0:09:59.160 --> 0:10:02.480 We have an interview with Sam's sister Annie, from whom 0:10:02.559 --> 0:10:05.400 he is estranged, and there's a lot of material in there. 0:10:05.440 --> 0:10:08.040 I think that tries to get closer to this answer 0:10:08.080 --> 0:10:10.360 of like, what should we make of this person? How 0:10:10.400 --> 0:10:14.920 should we think about checks and balances of power when 0:10:14.920 --> 0:10:18.880 we have these companies that are, by all appearances, gathering 0:10:18.920 --> 0:10:20.080 a lot of power and there for the people who 0:10:20.080 --> 0:10:21.600 are running them have a lot of power as well. 0:10:22.040 --> 0:10:26.600 So we have it's a five episode arc, five episode season, 0:10:26.920 --> 0:10:30.800 and the first three episodes are out now to the 0:10:30.840 --> 0:10:33.280 general public, and the last two will come out on 0:10:33.400 --> 0:10:35.840 subsequent Thursdays, And if you would like to binge the 0:10:35.840 --> 0:10:40.160 whole season right away, The episodes are available early to 0:10:40.320 --> 0:10:41.800 Bloomberg dot Com subscribers. 0:10:43.880 --> 0:10:48.959 So you've just started this series about Sam Altman and 0:10:49.000 --> 0:10:52.360 his upbringing indoors, so the growth of open ai and 0:10:52.400 --> 0:10:55.240 looped and everything. Who are the people that have helped 0:10:55.280 --> 0:10:56.600 him get where he is today? 0:10:56.640 --> 0:11:01.000 Though, So the making of Sam Altman is really interesting 0:11:01.040 --> 0:11:03.840 part of the overall story of Sam Altman. You know, 0:11:03.880 --> 0:11:08.000 many people know him as the CEO of open Ai 0:11:08.160 --> 0:11:11.240 because that's the role he's been in when he has 0:11:11.800 --> 0:11:15.000 risen to prominence, you know, beyond Silicon Valley. Like I 0:11:15.040 --> 0:11:17.240 think for many years he was well known in Silicon Valley, 0:11:17.240 --> 0:11:19.880 but this is like now he's kind of a household name, 0:11:20.559 --> 0:11:23.960 and so it's important to understand where Sam came from. 0:11:24.040 --> 0:11:26.520 You know, he's been in the valley for you know, 0:11:26.600 --> 0:11:29.480 since two thousand and five, I think is when he 0:11:29.559 --> 0:11:31.400 started college two thousand and four, two thousand and five 0:11:31.400 --> 0:11:33.840 at Stanford. Then he dropped out, and then he joined 0:11:34.559 --> 0:11:38.200 y Combinator, the now famous startup accelerator. But he was 0:11:38.200 --> 0:11:43.360 actually part of the first cohort of founders ever in 0:11:43.960 --> 0:11:47.120 n YC. Along with which as well right yes, including 0:11:47.120 --> 0:11:51.200 the the co founders of Twitch and of Reddit and 0:11:51.280 --> 0:11:53.040 so Emmett Sheer, you know, for those who know Emmit 0:11:53.120 --> 0:11:55.920 Sheer has a like very short, seventy two hour cameo 0:11:56.040 --> 0:12:00.960 in the open AI very Sam been firing Saga. But yes, 0:12:01.000 --> 0:12:04.400 Emmett and Sam were both in the same y C batch. 0:12:04.559 --> 0:12:09.480 So when we think about Sam's early career in Silicon Valley, 0:12:09.520 --> 0:12:11.679 I think what's important to know is that he rose 0:12:11.840 --> 0:12:15.200 very quickly, in part because he was very successful in 0:12:15.240 --> 0:12:21.760 making these strategic, advantageous friendships and connections with already established 0:12:21.800 --> 0:12:24.680 people in the valley. The most important one is Paul Graham, 0:12:24.960 --> 0:12:27.320 who is the you know, one of the founders of 0:12:27.520 --> 0:12:32.760 y Combinator and you know basically like immediately took Sam 0:12:32.840 --> 0:12:36.840 under his wing when Sam joined this first batch of YC. 0:12:37.760 --> 0:12:41.439 And yeah, Paul's a really important mentor to Sam. He's 0:12:41.480 --> 0:12:45.040 kind of the first person who really sees in Sam 0:12:45.400 --> 0:12:50.120 this you know, ambition, this hunger for power, this like 0:12:50.280 --> 0:12:54.760 drive to really build bigger and bigger companies. Even when 0:12:54.840 --> 0:12:57.280 you know they met when Sam Altman was nineteen, so 0:12:57.679 --> 0:12:59.760 Paul like sees him as a teenager and sees this 0:13:00.600 --> 0:13:05.040 future potential. And so yes, you know, not only did 0:13:05.160 --> 0:13:08.199 Paul become a mentor to him and sort of helped 0:13:08.400 --> 0:13:11.360 build Sam's profile over those early years because he would 0:13:11.440 --> 0:13:13.240 you know, Paul Graham is very famous for writing these 0:13:13.360 --> 0:13:16.120 essays about how to build startups and how to build 0:13:16.120 --> 0:13:18.240 the best startups, and if you're at all interested in 0:13:18.240 --> 0:13:20.400 building startups, you've read many of them. They're kind of 0:13:20.440 --> 0:13:23.520 like almost like a startup bible. And in many of 0:13:23.520 --> 0:13:26.400 them he extols the virtues of Sam Altman. He talks 0:13:26.400 --> 0:13:30.120 about Sam's ambition, he talks about Sam's cunning, his ability 0:13:30.160 --> 0:13:32.880 to like you know, make deals and like think big 0:13:32.920 --> 0:13:33.520 and never. 0:13:33.400 --> 0:13:36.240 Actually think Sam Motman is done, is what I've found. 0:13:36.720 --> 0:13:39.360 Yeah, there are some interesting you know, I've read many 0:13:39.400 --> 0:13:41.680 of the things Paul has written about Sam. Some of 0:13:41.679 --> 0:13:47.240 my favorite ones include Paul writing that within three minutes 0:13:47.240 --> 0:13:50.680 of meeting Sam, this was when Sam was nineteen, Paul 0:13:50.760 --> 0:13:53.600 thought to himself, Ah, so this is what a young 0:13:53.800 --> 0:13:55.679 Bill Gates is like, or you know, this is what 0:13:55.720 --> 0:13:57.480 Bill Gates was like at nineteen, I think is the 0:13:57.520 --> 0:13:59.840 exact quote. So, you know, he really build him up 0:13:59.840 --> 0:14:02.480 in the way. And I do think Paul had like 0:14:02.600 --> 0:14:05.360 unique insight into Sam, like they were close. They in 0:14:05.400 --> 0:14:08.920 many ways I'm sure still are, but it is this 0:14:09.040 --> 0:14:12.160 interesting role where you know, Paul met Sam when he 0:14:12.240 --> 0:14:13.840 really didn't have much to his name, and he really 0:14:13.840 --> 0:14:16.920 elevated him early on through his writings as this like 0:14:17.040 --> 0:14:19.720 startup founder to emulate right that other founders should be 0:14:19.760 --> 0:14:22.840 emulating Sam. And then of course, as Sam progresses in 0:14:22.880 --> 0:14:26.320 the valley, he also starts to write these like startup 0:14:27.080 --> 0:14:32.080 wisdom essays in a similar style to Paul. And then, 0:14:32.120 --> 0:14:33.960 of course, the most important thing that happens is that 0:14:34.000 --> 0:14:36.920 in twenty fourteen, when Paul decides he no longer wants 0:14:36.960 --> 0:14:39.040 to run by Combinator, which at this point is a 0:14:39.120 --> 0:14:41.640 much bigger vehicle than it was when Sam first started. 0:14:41.640 --> 0:14:43.720 It has no longer just a few stops totally. It 0:14:43.720 --> 0:14:48.080 has produced Stripe, Dropbox, Airbnb. This is a big job, right, 0:14:48.160 --> 0:14:51.080 like running y Combinator. And when Paul wants to hand 0:14:51.120 --> 0:14:52.920 it off to someone, you know, he has said that 0:14:52.960 --> 0:14:55.840 the only person he considered giving this to was Sam. 0:14:55.880 --> 0:14:58.720 So in twenty fourteen, when Sam is I believe twenty 0:14:58.760 --> 0:15:02.120 eight years old, he becomes the president of Y Combinator, 0:15:02.120 --> 0:15:04.360 and this is you know, he had started a startup, 0:15:04.400 --> 0:15:07.480 it didn't really work, he sold it and was starting 0:15:07.520 --> 0:15:10.480 to tabble in angel investing, and at that point Paul 0:15:10.560 --> 0:15:14.080 really elevated Sam to this new position of power. And 0:15:14.120 --> 0:15:16.560 then he ran YC for a while and then started 0:15:16.560 --> 0:15:20.200 open Ai. And in starting open Ai, he also leveraged 0:15:20.240 --> 0:15:24.920 these like very useful connections with particularly powerful people who 0:15:24.920 --> 0:15:27.520 could help him, such as Elon Musk, who was able 0:15:27.560 --> 0:15:30.840 to give the vast majority of the pledged funding to 0:15:30.920 --> 0:15:35.680 start open Ai. Later, when Elon Musk splits from open Ai, 0:15:35.880 --> 0:15:39.080 Sam makes his very powerful partnership with Satya Nadella to 0:15:39.120 --> 0:15:42.880 help fund open Ai. Another important partnership that Sam has made, 0:15:43.320 --> 0:15:46.080 you know, much earlier on was his friendship with Peter 0:15:46.240 --> 0:15:48.360 Teel And one of the things Peter Teal does is 0:15:48.400 --> 0:15:53.440 also you know, gives him millions of dollars to start investing. 0:15:53.760 --> 0:15:56.880 This is like before Sam takes over at y Yeah. 0:15:56.920 --> 0:15:59.680 And you know another thing that Paul did that really 0:16:00.200 --> 0:16:03.240 Paul Graham did that really helped Sam was also he 0:16:03.280 --> 0:16:06.080 gave Paul had the opportunity to be one of the 0:16:06.080 --> 0:16:09.800 first investors in Stripe. He was offered the chance to 0:16:09.920 --> 0:16:13.160 invest thirty thousand dollars for four percent of Stripe, which, 0:16:13.200 --> 0:16:16.520 of course, now that Stripe is enormous we all know 0:16:16.560 --> 0:16:19.800 how valuable that was, and Paul split it with Sam. 0:16:19.840 --> 0:16:21.960 He was like, oh, I might as well share this 0:16:22.040 --> 0:16:24.920 with Sam. So Sam has said that that fifteen thousand 0:16:24.920 --> 0:16:27.760 dollars for a two percent of Stripe has been, you know, 0:16:28.120 --> 0:16:30.120 one of his best performing angel investments ever. 0:16:30.160 --> 0:16:32.120 That was something he question is always where he got 0:16:32.120 --> 0:16:35.800 fifteen grand from. He was still working on Looped at 0:16:35.800 --> 0:16:38.280 the time. It's funny how perfletrip anyway. 0:16:38.400 --> 0:16:43.160 Yeah, my guess is fifteen grand was I don't actually 0:16:43.160 --> 0:16:44.880 know this, but my guess is fifteen perllen was not 0:16:45.040 --> 0:16:47.000 hard for him to pull up. And it's one of 0:16:47.000 --> 0:16:49.480 those things where it's really is, you know, access to 0:16:50.560 --> 0:16:52.880 access and relationships are the sorts of things that can 0:16:52.920 --> 0:16:55.600 build a career and can lead to great wealth, right 0:16:55.640 --> 0:16:58.680 like Sam is now you know, by our own internal 0:16:58.680 --> 0:17:02.240 accounts and by other lists billionaire. And this money comes 0:17:02.280 --> 0:17:04.600 from you know, not from open AI, but from these 0:17:04.600 --> 0:17:07.680 angel investments that he's made early on that have been 0:17:08.080 --> 0:17:09.119 enormously successful. 0:17:10.040 --> 0:17:12.639 So you called him in one of the titles, the 0:17:12.680 --> 0:17:15.600 Most Silicon Valley Man Alive. Is this what you're getting at, 0:17:15.600 --> 0:17:17.560 this kind of power player mentality? 0:17:17.680 --> 0:17:20.840 Yeah, I think it's it reflects a few things. One 0:17:20.920 --> 0:17:24.439 that even though he's you know, he's in his late thirties, 0:17:24.440 --> 0:17:26.960 he's been a player in Silicon Valley for such a 0:17:26.960 --> 0:17:30.960 long time, you know, close to two decades. And also 0:17:31.040 --> 0:17:34.399 that he's just someone who is extremely well connected. So 0:17:35.480 --> 0:17:38.520 even before he took over Y Combinator, which I think 0:17:38.800 --> 0:17:42.119 you could argue is like kind of king of the 0:17:42.119 --> 0:17:44.639 startup world in some sense, like Y Combinators, like you know, 0:17:44.680 --> 0:17:49.080 the topics totally, even before he took over I Y Comminator, 0:17:49.080 --> 0:17:51.160 I think he was extremely well connected. He's very social, 0:17:51.240 --> 0:17:53.679 he's very helpful, he's very efficient, Like many people have 0:17:53.760 --> 0:17:57.240 told me stories in which he, you know, calling Sam 0:17:57.560 --> 0:18:00.800 and talking for five minutes has saw their problem because 0:18:00.800 --> 0:18:02.959 he knows exactly the right person to call to fix it, 0:18:03.119 --> 0:18:04.920 or you know, he's really good at making deals. I 0:18:04.920 --> 0:18:07.800 think it's just clear he's extremely well integrated into this 0:18:07.840 --> 0:18:15.280 world and has very successfully moved up the Silicon Valley 0:18:15.280 --> 0:18:18.240 status ladder to the point where he is now, which 0:18:18.240 --> 0:18:20.520 is kind of you know, one of the you know, 0:18:20.560 --> 0:18:23.479 he's the CEO of the one of the arguably hottest 0:18:23.480 --> 0:18:25.800 companies in the valley right now, and I think that 0:18:25.800 --> 0:18:27.959 that's not luck right, Like he didn't just come up 0:18:27.960 --> 0:18:29.960 with He's not like a nobody who came up with 0:18:30.000 --> 0:18:32.720 an idea. It's like he has the connections and has 0:18:32.720 --> 0:18:36.200 parlayed his connections into power to bring him to the 0:18:36.240 --> 0:18:37.040 point he is now. 0:18:38.160 --> 0:18:41.479 So in your experience talking to people about Sam Altman, 0:18:42.280 --> 0:18:45.359 how technical is he? Do you think? What if you heard? 0:18:45.440 --> 0:18:48.720 Because you say, there he wasn't lucky, But he also 0:18:48.800 --> 0:18:52.080 does not appear to have successfully run a business because 0:18:52.160 --> 0:18:55.600 Luke shut down two people, well, two executives tried to 0:18:55.600 --> 0:18:57.840 get him fired from there. He got fired from y Combinator, 0:18:58.040 --> 0:18:59.920 which did very well. But at the same time, why 0:19:00.119 --> 0:19:02.800 he was basically a convey about for money at one point, 0:19:03.000 --> 0:19:07.040 not so much recently. Yeah, it just it feels weird 0:19:07.040 --> 0:19:10.520 that this completely non technical, semi non technical guy has 0:19:10.520 --> 0:19:11.640 ascended so far. 0:19:12.880 --> 0:19:15.760 My sense is that's not maybe the most fair description. 0:19:15.880 --> 0:19:18.680 Like I think Sam is incredibly smart, and people say 0:19:18.720 --> 0:19:20.639 this a lot, and you know, I believe them. I 0:19:20.680 --> 0:19:24.040 think his special skill, you know, he obviously knows how 0:19:24.080 --> 0:19:26.800 to like he's an engineer, he has training. I'm sure 0:19:26.800 --> 0:19:28.600 he can build a lot of stuff. It seems like 0:19:28.680 --> 0:19:33.480 his comparative advantage, his special skill is relationships, deal making, 0:19:34.000 --> 0:19:38.160 figuring out who exactly is the right person to help 0:19:38.240 --> 0:19:40.720 him in whatever he's really trying to get done, and 0:19:40.760 --> 0:19:43.760 figuring out the best way to get something to happen. 0:19:44.200 --> 0:19:45.720 You know, one of the people I spoke to is 0:19:45.760 --> 0:19:48.919 someone who knows Sam from when he was younger and 0:19:48.960 --> 0:19:53.919 knows him personally, and said that his superpower is figuring 0:19:53.920 --> 0:19:56.119 out who's in charge or figuring out who is in 0:19:56.119 --> 0:19:58.560 the best position to help him, and then charming them 0:19:58.640 --> 0:20:02.440 so that they help him with whatever goal he's trying 0:20:02.480 --> 0:20:05.200 to get done. And I think that, like, yeah, one 0:20:05.200 --> 0:20:07.359 could argue that that's actually a really good skill set 0:20:08.240 --> 0:20:10.840 if you want to build a very big company, which 0:20:13.200 --> 0:20:14.800 you know, I think at this current moment he has 0:20:14.880 --> 0:20:19.000 right like opening eyes. Really, you know, you can there's 0:20:19.040 --> 0:20:21.280 a lot that you can say about whether they're upholding 0:20:21.280 --> 0:20:23.600 their original mission or that. You know, that's up for debate, 0:20:23.640 --> 0:20:29.200 But I think that they've obviously been commercially successful so far, so. 0:20:30.040 --> 0:20:33.560 It feels like Silicon Valley on some level. And I 0:20:33.800 --> 0:20:36.760 just to give some thoughts here within the two episodes 0:20:36.760 --> 0:20:39.679 I'm doing here, the pattern I've seen with Sam Oltman 0:20:40.200 --> 0:20:44.040 is that everyone seems to want him to win, and 0:20:44.080 --> 0:20:47.600 there's almost a degree of they will make it. So 0:20:49.200 --> 0:20:52.879 have you seen anyone who's really a detractor or anyone 0:20:52.880 --> 0:20:57.960 who's not pro Sam Oltman, because it's interesting how few 0:20:57.960 --> 0:20:59.399 people are in tech. 0:21:00.480 --> 0:21:03.440 Well there is. I won't get too into it because 0:21:03.440 --> 0:21:05.440 this is in some of the future episodes which will 0:21:05.520 --> 0:21:09.320 drop in future weeks, but you know, I would say 0:21:09.359 --> 0:21:12.000 in in in some of the conversations that I've had 0:21:12.080 --> 0:21:15.439 off the record about people about in some of the 0:21:15.440 --> 0:21:17.960 conversations I've had off the record with people about Sam, 0:21:18.720 --> 0:21:22.080 I think, you know, my general impressions are people often 0:21:22.160 --> 0:21:24.760 do find him impressive in terms of what he has 0:21:24.800 --> 0:21:28.600 gotten done, you know, the size and scale of his 0:21:28.640 --> 0:21:30.800 ambitions and the way that he has generally been able 0:21:30.840 --> 0:21:36.240 to make that happen. I think there's also a lot 0:21:36.240 --> 0:21:39.280 of people who you know, are willing to privately share 0:21:39.480 --> 0:21:43.960 some gripes that they might have about him. Also, you know, 0:21:44.040 --> 0:21:46.560 in recent weeks we've seen people be a lot more 0:21:46.600 --> 0:21:48.960 public about some of those gripes. We have Helen Toner, 0:21:49.760 --> 0:21:52.720 a former former board member at open AI who voted 0:21:52.720 --> 0:21:56.200 to remove Sam last November, saying publicly in the last 0:21:56.240 --> 0:21:58.960 few weeks that Sam lied to her and the other 0:21:59.000 --> 0:22:03.439 former board members that his you know, misdirection made them 0:22:03.440 --> 0:22:07.119 feel like they couldn't do their jobs. And she has 0:22:07.119 --> 0:22:11.720 also said that people were intimidated to the point where 0:22:11.720 --> 0:22:14.560 they did not feel comfortable speaking more publicly about negative 0:22:14.600 --> 0:22:18.000 experiences they'd had about Sam, that they are afraid to 0:22:18.040 --> 0:22:21.800 speak more publicly about you know, times that he has 0:22:21.840 --> 0:22:24.560 not been honest with them or you know, has in 0:22:24.600 --> 0:22:29.240 which they've had challenging experiences. And that has also been 0:22:29.280 --> 0:22:31.399 reflected in some of the private conversations I've had in 0:22:31.440 --> 0:22:35.239 which people you know, they might have complaints or they 0:22:35.280 --> 0:22:38.199 might have had like challenging situations with him, and I 0:22:38.240 --> 0:22:40.600 think they just feel like the risk calculus is not 0:22:40.640 --> 0:22:43.360 worth it to come out and say something like that. 0:22:43.520 --> 0:22:47.880 But you know, there have been bits and pieces where 0:22:47.920 --> 0:22:50.720 people have come out and said things that you know, 0:22:50.800 --> 0:22:52.960 Sam has. You know. Another thing that the board members 0:22:53.000 --> 0:22:56.240 have said was that Sam had been deceptive and manipulative, 0:22:56.760 --> 0:22:59.679 and that's also followed up by or not followed up. 0:22:59.840 --> 0:23:01.960 There was also I think back in November, a former 0:23:02.000 --> 0:23:05.280 open ai employee who had tweeted something publicly about that 0:23:06.359 --> 0:23:08.120 you know, saying that Sam had lied to him on occasion, 0:23:08.119 --> 0:23:10.480 even though he had also always been nice to him, 0:23:10.520 --> 0:23:12.920 which I think is a very interesting combination of. 0:23:14.920 --> 0:23:18.400 That Silicon Valley though, I'm afraid of dealing with them, 0:23:18.440 --> 0:23:19.720 but they were so nice to me. 0:23:21.000 --> 0:23:22.600 And yeah, of course that you know that person has 0:23:22.640 --> 0:23:25.800 not elaborated more publicly about what they meant. But I 0:23:25.800 --> 0:23:28.840 think I think, you know, I think this is why 0:23:28.880 --> 0:23:33.560 people are asking themselves these questions, which is like, you know, 0:23:33.600 --> 0:23:36.680 the more that we hear about what the board uh 0:23:37.440 --> 0:23:39.640 was thinking before they decided to fire Sam, I think 0:23:39.640 --> 0:23:42.120 the more people are wondering about what are the patterns 0:23:42.119 --> 0:23:46.080 of behavior that he shows that you know that led 0:23:46.080 --> 0:23:48.280 to the board trying to make this drastic move. 0:23:50.000 --> 0:23:54.800 Yeah, that's actually an interesting point. So when Samulton was 0:23:54.880 --> 0:23:58.760 fired from open AI, there was this very strange reaction 0:23:58.960 --> 0:24:02.520 from Silicon Valley, including some in the media, where it 0:24:02.720 --> 0:24:07.920 was almost like a tounga Games everyone doing the symbol 0:24:08.000 --> 0:24:09.920 thing where everyone's like, oh, we got to put sam 0:24:09.920 --> 0:24:13.600 otment back in, isn't it kind of strange? We still 0:24:13.640 --> 0:24:16.320 don't know why he was actually fired, though, I mean, 0:24:16.320 --> 0:24:19.800 Helen Tona has elaborated like I've never seen a have 0:24:19.840 --> 0:24:21.439 you seen anything like this? In your career. 0:24:22.080 --> 0:24:26.000 I think that it has been surprising that there has 0:24:26.080 --> 0:24:31.760 not been more of a clear answer. I think, you know, 0:24:31.800 --> 0:24:35.080 as as time has gone on, like we have heard 0:24:35.119 --> 0:24:38.160 a little bit more like I think Helen Toner has, 0:24:39.000 --> 0:24:42.120 you know, to her credit, tried to give more information 0:24:42.359 --> 0:24:46.479 in recent weeks about what happened. I think, you know, 0:24:46.800 --> 0:24:49.639 people were obviously asking this question six months ago, and 0:24:49.680 --> 0:24:51.119 so I think like there's been a little bit of 0:24:51.160 --> 0:24:55.320 a delay and trying to get this this answer, and 0:24:55.359 --> 0:24:57.880 I wonder if maybe there just isn't like a very 0:24:59.560 --> 0:25:02.200 neat answer to it, and so and then in that 0:25:02.800 --> 0:25:07.080 absence we get this kind of more of a like murky, multifaceted, 0:25:07.160 --> 0:25:09.879 multi voiced answer. But I yes, I agree that it 0:25:10.160 --> 0:25:12.440 is sort of surprising that there that there hasn't been 0:25:13.280 --> 0:25:17.959 more clarification on what exactly happened or a little bit 0:25:18.040 --> 0:25:22.080 more granular detail about what led up to it. 0:25:23.359 --> 0:25:26.560 So on today, I aihype in general said that a 0:25:26.560 --> 0:25:29.800 bit weird. I'll keep going, why do you think there's 0:25:29.840 --> 0:25:32.760 such a dolf between what Sam Altman says and what 0:25:32.880 --> 0:25:34.520 chat GPT can actually do? 0:25:35.760 --> 0:25:38.480 What Sam Altman says, what are you talking about specifically. 0:25:38.040 --> 0:25:40.400 As in he says it will be a super smart company. Yeah, 0:25:40.440 --> 0:25:42.200 yeh yeah, that he'll be all of these things. 0:25:42.960 --> 0:25:45.280 Well, this is something that we get into in episode three, 0:25:45.280 --> 0:25:47.600 which is a personal interest of mine, which is kind 0:25:47.640 --> 0:25:51.399 of the psychology of the AI industry right now. And 0:25:52.280 --> 0:25:55.399 you know, I what I find so interesting about this 0:25:55.440 --> 0:25:58.120 and what we try to delve into in episode three 0:25:58.160 --> 0:26:01.639 and kind of throughout the series is these kind of 0:26:01.680 --> 0:26:05.200 like extreme projections about AI and in the industry, you 0:26:05.240 --> 0:26:07.840 see both positive ones and negative ones, and I think, 0:26:08.080 --> 0:26:11.440 you know, the negative ones, that's what looks like AI dooomerism, 0:26:11.560 --> 0:26:15.159 AI existential risks, sometimes called AI safety depending on your 0:26:15.160 --> 0:26:17.240 point of view. But you know, it's these projections that 0:26:18.200 --> 0:26:23.800 you know, superintelligence might very quickly and very soon learn 0:26:23.920 --> 0:26:27.639 to self improve in a way that allows it to 0:26:27.800 --> 0:26:31.200 rapidly outstrip our control and our capabilities and could lead 0:26:31.240 --> 0:26:34.959 to the extinction of humanity. There are so many interesting 0:26:34.960 --> 0:26:38.160 things to say about the psychology of believing that our 0:26:38.240 --> 0:26:41.760 human race might either be wiped out or incredibly changed 0:26:42.680 --> 0:26:46.320 within our lifetimes, and we get into that in episode three. 0:26:46.359 --> 0:26:48.560 I think I really wanted to get into the psychology 0:26:48.560 --> 0:26:51.359 of someone who believes that AI doom is just around 0:26:51.400 --> 0:26:53.840 the corner. And so we talked to someone who sort 0:26:53.840 --> 0:26:56.920 of became convinced of this belief soon after the twenty 0:26:56.920 --> 0:27:02.160 sixteen alphag Matches in which the go playing AI beat 0:27:02.200 --> 0:27:05.320 the you know, the world's champion in Go and he 0:27:05.359 --> 0:27:07.720 talks about yeah, no longer, you know, deciding not to 0:27:07.840 --> 0:27:09.960 make a retirement account because he was like, what is 0:27:10.000 --> 0:27:13.880 the point by the time I reach retirement age, either 0:27:14.000 --> 0:27:16.960 the world will be dramatically different and money won't matter, 0:27:17.040 --> 0:27:19.880 or will all be dead? And and I think that 0:27:20.280 --> 0:27:22.119 even though some people might scoff at that, that's like 0:27:22.119 --> 0:27:25.679 a real belief that people believe that this, you know, 0:27:25.760 --> 0:27:29.800 these extreme possible scenarios are in our near future. And 0:27:29.880 --> 0:27:32.520 on the other hand, we also see extreme projections in 0:27:33.000 --> 0:27:36.159 a positive direction. You know, this idea that AI is 0:27:36.200 --> 0:27:41.920 going to unlock a whole new era of human flourishing, 0:27:42.000 --> 0:27:44.880 that we might expand beyond our planet, that we might 0:27:44.920 --> 0:27:49.400 be able to give say what abundant abundance right exactly? 0:27:49.440 --> 0:27:52.880 You know. One of the things we do I believe 0:27:52.920 --> 0:27:54.879 in episode three is is do a little bit of 0:27:54.920 --> 0:27:57.320 a supercut of Sam all been talking about abundance. It's 0:27:57.359 --> 0:27:59.840 it's pretty clear that This is a way that he 0:27:59.880 --> 0:28:02.320 likes to frame our AI future is going to be 0:28:02.320 --> 0:28:05.280 this future in which everyone has plenty, right, everyone has 0:28:06.480 --> 0:28:10.000 you know, access to intelligence, abundant energy, abundant access to 0:28:10.920 --> 0:28:13.240 superintelligence that can help us live kind of our best 0:28:13.280 --> 0:28:18.800 lives and beyond our wildest dreams. Right. And I you know, 0:28:18.960 --> 0:28:22.639 obviously Silicon Valley is a place where people like to 0:28:22.840 --> 0:28:26.480 make grandiose statements. But this is beyond that, right, This 0:28:26.600 --> 0:28:30.119 is not just this is not just like you know, 0:28:30.160 --> 0:28:32.840 we joked about we work, We works. Mission statement was 0:28:32.880 --> 0:28:37.160 to elevate the world's consciousness. Like well, galaxies of human 0:28:37.200 --> 0:28:41.080 flourishing for eons beyond us. Like that is on another scale, right, 0:28:41.160 --> 0:28:45.280 Like we're talking about something that is sort of at 0:28:45.320 --> 0:28:49.040 an unprecedented level of extreme rhetoric. And I think that's 0:28:49.040 --> 0:28:52.160 really interesting. I think it is a very powerful motivator, 0:28:52.200 --> 0:28:54.520 both in a you know, in the doumer sense and 0:28:54.560 --> 0:28:57.240 also in the abundance sense. People believing that what they're 0:28:57.240 --> 0:29:01.200 working on is the most important technological leap forward for humanity. 0:29:02.080 --> 0:29:07.760 Talk about a motivating reason to work on this technology, right, 0:29:07.800 --> 0:29:10.800 talk about a way to feel powerful, feel like you're 0:29:10.840 --> 0:29:13.040 making a huge difference. I think that's a really key 0:29:13.080 --> 0:29:16.240 part of what's driving a lot of work in AI 0:29:16.360 --> 0:29:16.720 right now. 0:29:17.480 --> 0:29:20.640 Striving a lot of work, sure, but with Aortman himself, 0:29:21.240 --> 0:29:25.560 there is this golf. It is a million mile golf 0:29:25.600 --> 0:29:28.400 between the things he says and what chat GBT is 0:29:28.440 --> 0:29:31.760 even even on the most basic level capable of doing 0:29:31.840 --> 0:29:34.800 and will be capable. And it just feels like it 0:29:34.840 --> 0:29:39.280 almost feels like he's become the propagandist for the tech industry. 0:29:40.600 --> 0:29:44.160 And it's very strange to me how far that distance is. 0:29:44.200 --> 0:29:48.520 Because you've got the AI doomers and the AI optimist 0:29:48.520 --> 0:29:52.680 I guess you'd call them. But Aortmand doesn't even feel 0:29:52.720 --> 0:29:55.240 like he's in with He's just kind of He'll say 0:29:55.320 --> 0:29:57.560 one day that he doesn't think it's a creature, the 0:29:57.640 --> 0:29:59.400 next one will say it's going to kill us, or 0:30:00.080 --> 0:30:04.200 all just feels like a pr campaign but for nothing. 0:30:05.600 --> 0:30:09.400 Yeah, it has been interesting to try to answer the question, 0:30:09.440 --> 0:30:10.840 you know, one of the questions we tried to answer 0:30:10.840 --> 0:30:14.600 in the podcast is does Sam actually believe you know, 0:30:14.680 --> 0:30:17.560 because as you mentioned, there are some early clips of him, 0:30:17.720 --> 0:30:19.360 you know, And when I say early, I mean around 0:30:19.400 --> 0:30:21.800 the time of founding Open AI twenty fifteen. Or so, 0:30:22.400 --> 0:30:26.600 there's some clips of him talking about, you know, saying 0:30:26.720 --> 0:30:29.080 somewhat jokingly that AI might kill us all. But there's 0:30:29.120 --> 0:30:31.240 also this uh, you know, very famous blog post that 0:30:31.280 --> 0:30:33.440 he wrote in twenty fifteen in which he says that, 0:30:34.080 --> 0:30:36.920 you know, basically super intelligence is one of the most 0:30:36.920 --> 0:30:40.760 serious risks to humanity, you know, full stop. And so 0:30:41.080 --> 0:30:43.640 it's clear that at some point in his life he 0:30:43.800 --> 0:30:46.960 believed kind of what we might now call a more 0:30:47.000 --> 0:30:50.520 doom ory outlook. But as time has gone on, he 0:30:50.640 --> 0:30:53.560 has you know, offered views that are a little bit 0:30:53.560 --> 0:30:56.920 more measured and more positive. You know, he tends to 0:30:57.200 --> 0:30:59.080 you know, in his big media tour of twenty twenty three, 0:30:59.800 --> 0:31:03.360 he tended to talk about how AI was going to 0:31:04.320 --> 0:31:07.720 you know, his projection was that AI would radically transform society, 0:31:07.720 --> 0:31:09.640 but that it would be net good, right, that like, 0:31:09.800 --> 0:31:13.160 overall we would be glad that this happened, and that 0:31:13.240 --> 0:31:16.959 it would improve lives, even if in the short term 0:31:17.680 --> 0:31:20.680 or for some people it might prove to be bringing 0:31:20.720 --> 0:31:24.360 a lot of challenges as well. And so it is 0:31:24.440 --> 0:31:26.040 you know, I think one of the interesting things about 0:31:26.160 --> 0:31:28.240 is about him is it is a little hard to 0:31:28.280 --> 0:31:30.640 pin down exactly what he thinks. I think you're right 0:31:30.680 --> 0:31:35.440 that I wouldn't consider him like a gung ho effective accelerationist. 0:31:35.480 --> 0:31:37.240 I would not consider him a dumer. He is like 0:31:38.360 --> 0:31:41.320 somewhere in this large gulf in between there. But I 0:31:41.360 --> 0:31:46.640 think he's also smart enough to know that making grandiose 0:31:46.680 --> 0:31:52.840 projections about what AI could bring is a compelling story, right, Like, 0:31:53.040 --> 0:31:57.120 is a story that he can help sell by being 0:31:57.160 --> 0:31:59.000 like a spokesman for it. And often that is the 0:31:59.080 --> 0:32:01.520 role of a CEO, is to be a really good storyteller, 0:32:01.920 --> 0:32:05.040 to bring the pitch of the company to the public, 0:32:05.120 --> 0:32:09.800 to investors, to potential employees, to customers, to try to 0:32:09.800 --> 0:32:11.280 sell them on this vision of the future. And I 0:32:11.280 --> 0:32:13.520 do think Sam is good at that. There is an 0:32:13.520 --> 0:32:19.200 interesting tidbit in episode three in which we interview a 0:32:19.280 --> 0:32:22.080 fiction writer who was actually hired on contract by open 0:32:22.120 --> 0:32:28.200 Ai to write like a novella about AI futures and 0:32:28.240 --> 0:32:31.680 things like that. And yeah, he just talks a little 0:32:31.680 --> 0:32:34.160 bit about you know, the novella is not I think 0:32:34.480 --> 0:32:37.120 in active use within open Ai, but they did at 0:32:37.120 --> 0:32:40.480 some point, see they did, at some point see value 0:32:40.600 --> 0:32:43.760 in commissioning it. And I think you know something that 0:32:43.800 --> 0:32:47.000 the author, Patrick House explains to us, is you know 0:32:47.320 --> 0:32:51.360 that open Ai, just like many other startups, is really 0:32:51.440 --> 0:32:54.280 motivated by story, right, and that Sam Altman is inspired 0:32:54.680 --> 0:32:56.880 by fiction. You know, it's inspired by certain kinds of 0:32:56.880 --> 0:32:59.160 sci fi. I think this is not unique to Sam. 0:32:59.240 --> 0:33:02.360 Many founders in Silicon Valley, you know, Elon Musk has 0:33:02.360 --> 0:33:05.640 talked about this as well, are driven to create things 0:33:05.680 --> 0:33:09.160 in part because of what they read about when they 0:33:09.160 --> 0:33:11.600 were younger, that you know, these dreams of the future. 0:33:12.400 --> 0:33:14.720 And so it's just interesting to get his perspective on 0:33:15.320 --> 0:33:18.680 how motivating a story can be, and how motivating this 0:33:19.040 --> 0:33:21.320 compelling story of like, oh, we're building something that's going 0:33:21.400 --> 0:33:23.880 to change the course of human history. Like you just 0:33:23.920 --> 0:33:26.640 couldn't ask for a more powerful motivating force. 0:33:28.000 --> 0:33:33.840 So as Alman accumulates power and as he kind of 0:33:34.000 --> 0:33:37.320 sends the top of open Ai, do you think he's 0:33:37.400 --> 0:33:39.920 done there? Do you think there's going to be another 0:33:40.000 --> 0:33:42.760 thing he starts? Because it feels like you've discussed like 0:33:42.920 --> 0:33:44.720 UBI and all these other things. Do you think he 0:33:44.760 --> 0:33:47.880 has grander ideas that he wants to pursue. 0:33:49.000 --> 0:33:53.760 Well, obviously I can't speak to what's inside Sam's I 0:33:53.840 --> 0:33:58.000 don't know the man's mind, but I mean past indicators 0:33:58.000 --> 0:34:01.160 would suggest yes, Like I think he has proven pretty 0:34:01.160 --> 0:34:05.600 consistently that he's someone who you know, is you know, 0:34:05.600 --> 0:34:08.239 as much as he might focus on one project with 0:34:08.280 --> 0:34:10.759 a lot of effort, like he is cooking things on 0:34:10.840 --> 0:34:13.000 the side, Like this is a man. This is gonna 0:34:13.000 --> 0:34:15.160 be an extended metaphor, but this is a man working 0:34:15.200 --> 0:34:17.239 at a stove that has like six burners, not one. 0:34:17.320 --> 0:34:19.719 And you know he we already know that. 0:34:20.680 --> 0:34:22.839 What you say, sorry, he's got a big house. 0:34:23.840 --> 0:34:28.400 He's got multiple houses. The uh, the you know, we 0:34:28.480 --> 0:34:32.319 already know that. You know. In addition to running open AI, 0:34:33.080 --> 0:34:38.279 he has funded and or helped prompt the founding of, 0:34:38.719 --> 0:34:41.560 or has you know, been very involved in investing in 0:34:41.640 --> 0:34:45.239 or supporting other startups that you know, are part of 0:34:45.280 --> 0:34:50.560 this kind of ecosystem of businesses that are connected to 0:34:51.080 --> 0:34:53.760 an AI future or might benefit in an AI future. 0:34:53.800 --> 0:34:57.080 So for example, he Leon, which is a nuclear fusion 0:34:57.360 --> 0:34:59.680 company which he has invested a ton of money into. 0:35:00.239 --> 0:35:02.160 I think he has said publicly that you know, his 0:35:02.800 --> 0:35:05.960 vision is that this is a potential way to provide 0:35:06.280 --> 0:35:11.400 abundant energy that could then power the technology that we 0:35:11.440 --> 0:35:16.040 need to you know, improve AI to the level that 0:35:16.080 --> 0:35:17.520 we're hoping that it can get to, or that he's 0:35:17.520 --> 0:35:19.880 hoping that it can get to At the same time. 0:35:20.520 --> 0:35:22.880 You know, we've talked a little bit about universal basic income. 0:35:22.960 --> 0:35:26.080 This has been something that Sam has been a proponent 0:35:26.160 --> 0:35:29.640 of and an advocate of since at least twenty sixteen, 0:35:29.719 --> 0:35:32.560 when he was running y Combinator, and they started a 0:35:32.840 --> 0:35:37.120 side research project to study universal basic income by giving 0:35:39.280 --> 0:35:42.279 cash payments to families in Oakland of I believe a 0:35:42.320 --> 0:35:46.320 thousand dollars a month. That research project is still ongoing. 0:35:46.360 --> 0:35:50.319 It's now moved away from y Combinator and is associated 0:35:50.880 --> 0:35:54.200 with open Research, which is I believe funded by open Ai, 0:35:54.320 --> 0:35:56.360 and so it has kind of moved with Sam to 0:35:56.440 --> 0:36:00.480 his new role. And of course he also so co 0:36:00.640 --> 0:36:06.240 founded this company called world Coin, which used these silver 0:36:06.880 --> 0:36:12.120 orbs machines to scan to take pictures of your iris 0:36:12.160 --> 0:36:17.160 and give everyone register every individual human as like a 0:36:17.239 --> 0:36:21.360 unique human individual, and to create this eyeball registry in 0:36:21.480 --> 0:36:24.759 which by which one could in the future distribute a 0:36:24.880 --> 0:36:28.720 universal basic income. So he's funding these energy companies, He's 0:36:29.000 --> 0:36:32.480 involved in these sort of the sort of crypto eyeball 0:36:32.520 --> 0:36:35.720 registry project that will help distribute ebi in this future 0:36:35.719 --> 0:36:37.879 that he's imagining, Like, I think it's safe to say 0:36:37.880 --> 0:36:41.560 he's definitely thinking about things beyond just open AI for 0:36:41.719 --> 0:36:44.920 the future and imagining like, Okay, well, if we have 0:36:45.000 --> 0:36:46.680 this piece that's growing, what else would we need to 0:36:46.680 --> 0:36:49.359 support it? And I'm sure there are other things he's 0:36:49.400 --> 0:36:51.600 working on that we don't even know about, right, Like 0:36:51.600 --> 0:36:55.960 I know he has also funded some like longevity bioscience 0:36:56.040 --> 0:36:59.439 projects and things like that. He's I guarantee he's thinking 0:36:59.480 --> 0:37:01.200 about stuff beyond what we know about. 0:37:03.160 --> 0:37:07.480 Final question, why do you think the entire tech industry 0:37:07.480 --> 0:37:10.040 has become so fascinated with AI? Do you think it's 0:37:10.239 --> 0:37:12.279 just oldman or is it something more? 0:37:13.680 --> 0:37:18.080 I do think chat Gipt started heating up this interest 0:37:18.120 --> 0:37:20.360 that was already percolating a little bit in the tech industry. 0:37:20.760 --> 0:37:23.319 But it does seem like something about chatchipt capture the 0:37:23.320 --> 0:37:27.360 public imagination made people imagine very seriously for the first time, 0:37:28.200 --> 0:37:32.759 how AI could affect their lives, their lives individually. It 0:37:32.880 --> 0:37:34.279 used to be kind of this abstract thing that was 0:37:34.320 --> 0:37:36.920 a little farther away, or maybe you understood that like 0:37:37.320 --> 0:37:40.880 you were interacting with AI sometimes, like when you would 0:37:40.920 --> 0:37:46.080 look at like flight price predictors or yeah, exactly. But 0:37:46.920 --> 0:37:48.759 you know, I think as we you know, we talk 0:37:48.800 --> 0:37:51.400 about this in episode three, but you know, chat chipet 0:37:51.600 --> 0:37:55.920 wasn't even new technology. It was actually just a different 0:37:56.120 --> 0:38:00.000 user interface on a model that already existed GPT three 0:38:00.040 --> 0:38:00.560 point five. 0:38:01.160 --> 0:38:01.480 And so. 0:38:03.040 --> 0:38:05.200 To me, that actually speaks, I guess to the power 0:38:05.280 --> 0:38:09.440 of like making a technology accessible to everyone and in 0:38:09.480 --> 0:38:11.480 a way that was like easy to use and you know, 0:38:11.520 --> 0:38:13.879 for better or worse. That kind of got a lot 0:38:13.920 --> 0:38:17.160 of people in this like public momentum of people thinking 0:38:17.160 --> 0:38:21.640 about AI feeling you know, just feeling like it had 0:38:22.400 --> 0:38:25.879 rapidly increased its capabilities in a short period of time. 0:38:26.719 --> 0:38:33.120 And yeah, something about that really captured not just you know, 0:38:33.239 --> 0:38:36.120 the minds, but also the hearts of people and like 0:38:36.120 --> 0:38:39.040 like getting them really thinking about like what could a 0:38:39.040 --> 0:38:41.480 future like this look like. And I think while some 0:38:41.480 --> 0:38:44.120 people were excited, a lot of people also reacted with fear, 0:38:44.200 --> 0:38:45.719 right and like I think in the valley, like you 0:38:45.800 --> 0:38:49.960 will hear a lot of people more openly discussing their 0:38:50.040 --> 0:38:55.480 fears of uh, sort of like job loss or or 0:38:56.640 --> 0:38:58.880 just like dramatic social change that might come about in 0:38:58.880 --> 0:39:02.200 the next ten or twenty years. The feeling I get 0:39:02.239 --> 0:39:04.400 in conversations that I have in and around San Francisco 0:39:04.480 --> 0:39:06.680 is you know, even people who are pretty deep in 0:39:06.719 --> 0:39:11.759 this technology are uncertain about whether it's going to be 0:39:11.800 --> 0:39:14.839 overall good or bad. Like they're just uncertain of how 0:39:14.840 --> 0:39:17.480 to look back on this time, like whether it will 0:39:17.520 --> 0:39:22.040 have ended up being elite forward for humanity or something different. 0:39:34.520 --> 0:39:36.600 Aortman has taken advantage of the fact that the tech 0:39:36.640 --> 0:39:40.560 industry might not have any hypergrowth markets left, knowing that 0:39:40.640 --> 0:39:44.319 chat GPT is much like Sam Altman, incredibly adept at 0:39:44.360 --> 0:39:48.279 mimicking depth and experience by parroting the experiences of those 0:39:48.320 --> 0:39:52.560 that have actually done things. Like Sam Altman, chat GPT 0:39:52.680 --> 0:39:55.960 consumes information and feeds it back to the people using 0:39:56.000 --> 0:40:00.440 it in a way that feels superficially satisfying, and it's 0:40:00.520 --> 0:40:03.480 quite impressive to those who don't really care about creativity 0:40:03.560 --> 0:40:06.239 or depth. And like I've said, it takes advantage of 0:40:06.280 --> 0:40:09.080 the fact that the teche CoSystem has been dominated and 0:40:09.120 --> 0:40:13.960 funded by people who don't really build tech. As I've 0:40:14.000 --> 0:40:18.720 said before, generative AI things like chat GPT, anthropics Claude, 0:40:19.040 --> 0:40:22.000 Microsoft's co Pilot, which is also powered by chat GBT. 0:40:22.560 --> 0:40:25.759 It's not going to become the incredible supercomputer that Sam 0:40:25.800 --> 0:40:29.080 Moortman is promising. It will not be a virtual brain 0:40:29.400 --> 0:40:32.480 or imminently human like or a super smart person that 0:40:32.600 --> 0:40:36.600 knows everything about you, because it is, at its deepest complexity, 0:40:36.960 --> 0:40:41.040 a fundamentally different technology based on mathematics and the probabilistic 0:40:41.120 --> 0:40:44.160 answer to what you have asked it, rather than anything 0:40:44.239 --> 0:40:48.520 resembling how human beings think, or act or even know things. 0:40:48.719 --> 0:40:52.759 Generative AI does not know anything. How can a thing 0:40:52.880 --> 0:40:56.480 think when it doesn't know a anyone? I want to 0:40:56.520 --> 0:41:01.239 ask bradlight cap Miramarati Sam Ortmann of these questions just 0:41:01.360 --> 0:41:06.120 once hear what they fart out. No, a well CHATGBT 0:41:06.360 --> 0:41:10.560 isn't inherently useless. Allman realizes that it's impossible to generate 0:41:10.600 --> 0:41:13.200 the kind of funding and hype he needs based on 0:41:13.239 --> 0:41:16.719 its actual achievements, and that to continue to accumulate power 0:41:16.719 --> 0:41:19.200 and money, which is his only goal, he has the 0:41:19.239 --> 0:41:21.640 speciously hype it, and he has to hype it to 0:41:21.719 --> 0:41:25.239 wealthy and powerful people who also do not participate in 0:41:25.280 --> 0:41:30.319 the creation of anything. And that's who he is. I've 0:41:30.360 --> 0:41:32.840 been pretty mean about this guy, I really have, but 0:41:32.880 --> 0:41:36.640 he does have a skill. He knows a mark, he 0:41:36.719 --> 0:41:39.520 knows he knows how to say the right things and 0:41:39.560 --> 0:41:41.840 get in the right rooms with the people who aren't 0:41:41.880 --> 0:41:45.480 really touching the software or the hardware. He knows what 0:41:45.520 --> 0:41:47.719 they need to hear. He knows what the vcs need 0:41:47.800 --> 0:41:53.000 to hear. He knows quite aptly what this needs to 0:41:53.120 --> 0:41:56.080 sound like. But if he had to say what chat 0:41:56.120 --> 0:41:59.279 GBT does today, what would he say? Yeah, Yeah, it's 0:41:59.320 --> 0:42:01.319 really good at a generating a bunch of TEGs that's 0:42:01.400 --> 0:42:04.319 kind of shitty. Yeah, Sometimes it does math right and 0:42:04.360 --> 0:42:06.560 sometimes it does it really wrong. Sometimes you ask it 0:42:06.600 --> 0:42:08.680 to do. It can draw a picture, Hey, what do 0:42:08.719 --> 0:42:11.959 you think of that? These are all things, by the way, 0:42:12.080 --> 0:42:13.799 that if like a six year old told you'd be like, wow, 0:42:13.840 --> 0:42:15.960 that's really impressive, or like a ten year old perhaps, 0:42:16.520 --> 0:42:20.759 because that's a living being. Chat GPT does these things, 0:42:20.760 --> 0:42:22.799 and it does it. I know it's cheesy to say, 0:42:22.840 --> 0:42:24.920 but in a soulless way, but it really does that. 0:42:25.040 --> 0:42:27.239 Because and the reason all of this, the writing and 0:42:27.280 --> 0:42:30.520 the horrible video and the images, the reason it feels 0:42:30.560 --> 0:42:35.880 so empty is because even the most manure adjacent press 0:42:35.920 --> 0:42:42.160 release still has gone through someone's manure adjacent brain. Even 0:42:42.200 --> 0:42:46.000 the most pallid empty copy you've read has gone through someone. 0:42:46.320 --> 0:42:49.200 A person has put thought and intention in, even if 0:42:49.200 --> 0:42:53.280 they're not great with the English language. What chat GPT 0:42:53.520 --> 0:42:56.880 does is use math to generate the next thing, and 0:42:56.920 --> 0:43:00.359 sometimes it gets it pretty right. But pretty right is 0:43:00.400 --> 0:43:04.760 not enough to mimic human creation. But look at Sam Altman. 0:43:05.239 --> 0:43:08.200 Look who he is. What has he created other than 0:43:08.200 --> 0:43:12.720 wealth for him and other people? What about Sam Moltman 0:43:12.840 --> 0:43:17.840 is particularly exciting? Well, he's been rich before and his 0:43:18.040 --> 0:43:21.680 money made him even richer. That's pretty good. He was 0:43:21.760 --> 0:43:24.960 at y Combinator. Don't ask too much about what happened there. 0:43:26.480 --> 0:43:30.280 Just feels like sometimes Silicon Valley can't wipe its own ass. 0:43:30.800 --> 0:43:34.200 It can't see when there's a wolf amongst the sheep. 0:43:34.440 --> 0:43:37.560 It can't see when someone isn't really part of the 0:43:37.600 --> 0:43:41.040 system other than finding new ways to manipulate and extract 0:43:41.120 --> 0:43:44.360 value from him. And Sam Altman is a monster created 0:43:44.680 --> 0:43:47.880 by Silicon Valley's sin, and their sin, by the way, 0:43:48.080 --> 0:43:52.040 is empowering and elevating those who don't build software, which 0:43:52.120 --> 0:43:55.200 in turn has led to the greater sin of allowing 0:43:55.239 --> 0:43:58.280 the tech industry to drift away from fixing the problems 0:43:58.280 --> 0:44:02.960 of actual human beings. Sam Altman's manipulative little power plays 0:44:02.960 --> 0:44:05.760 have been so effective because so many of the power 0:44:05.760 --> 0:44:08.960 players in venture capital and the public markets and even 0:44:09.040 --> 0:44:12.640 tech companies are disconnected from the process of building things, 0:44:12.840 --> 0:44:16.200 of building software and hardware, and that makes them incapable 0:44:16.320 --> 0:44:20.080 or perhaps unwilling to understand that Sam Altman is leading 0:44:20.120 --> 0:44:23.759 them to a deeply desolate place. And on some level, 0:44:23.760 --> 0:44:27.080 it's kind of impressive how he succeeded in bending these 0:44:27.120 --> 0:44:29.920 fools to his whims, to the point that executives like 0:44:29.960 --> 0:44:32.320 sund up A Shai of Google are willing to break 0:44:32.600 --> 0:44:35.720 Google Search in pursuit of this next big hype cycle 0:44:35.960 --> 0:44:40.400 created by Sam Altman. He might not create anything, but 0:44:40.440 --> 0:44:44.799 he's excellent at spotting market opportunities, even if these opportunities 0:44:44.800 --> 0:44:49.600 involve him transparently lying about the technology he creates, all 0:44:49.640 --> 0:44:53.719 while having his nasty little boosters further propagate these bullshit, 0:44:54.320 --> 0:44:57.680 mostly because they don't know, or perhaps they don't care, 0:44:58.000 --> 0:45:01.560 if Sam Morton's full of shit. It doesn't matter to them. 0:45:01.840 --> 0:45:04.520 It doesn't matter that Google Search is still plagued with 0:45:04.600 --> 0:45:09.160 nonsensical AI answers that sometimes steal other people's work, or 0:45:09.160 --> 0:45:13.919 that AI in legal research has been proven to regularly hallucinate, which, 0:45:13.960 --> 0:45:16.280 by the way, is a problem that's impossible to fix. 0:45:17.280 --> 0:45:20.560 It's all happening because AI is the new thing that 0:45:20.600 --> 0:45:23.040 can be sold to the markets. And it's all happening 0:45:23.120 --> 0:45:27.120 because Sam Altman, intentionally or otherwise has created a totally 0:45:27.160 --> 0:45:30.840 hollow hype cycle. And all of this is thanks to 0:45:30.880 --> 0:45:33.800 Sam Altman and a tech industry that's lost its ability 0:45:33.840 --> 0:45:37.319 to create things worthy of an actual hype cycle to 0:45:37.400 --> 0:45:40.879 the point that this spacious, non technical manipulator can lead 0:45:40.920 --> 0:45:46.239 it down this nasty, ugly offensive anti tech path. The 0:45:46.280 --> 0:45:49.640 tech industry has spent years pissing off customers, with platforms 0:45:49.680 --> 0:45:53.319 like Facebook and Google actively making their products worse in 0:45:53.360 --> 0:45:56.680 the pursuit of perpetual growth and ashamedly turning their backs 0:45:56.800 --> 0:45:58.880 on the people that made them rich and acting with 0:45:58.920 --> 0:46:02.840 this horrifying contempt for their users. And I believe the 0:46:02.880 --> 0:46:05.439 result will be that tech is going to face a 0:46:05.520 --> 0:46:09.640 harsh reprimand from society. As I mentioned in the rock 0:46:09.680 --> 0:46:13.120 com bubble, things that are already falling apart WHEB traffic 0:46:13.160 --> 0:46:17.320 is already dropping. And what sucks is the people around 0:46:17.360 --> 0:46:20.920 Sam Moltman should have been able to see this, even 0:46:21.000 --> 0:46:25.360 putting aside his resume. I've listened to an alarming amount 0:46:25.360 --> 0:46:28.799 of Sam Moreman talk. And I'm a public relations person, 0:46:28.840 --> 0:46:30.920 Who the hell am I? I'm someone who's been around 0:46:30.960 --> 0:46:33.160 a lot of people who make shit up. I've been 0:46:33.200 --> 0:46:35.560 around a lot of people whose job it is to 0:46:35.680 --> 0:46:41.440 kind of obfuscate things, and quite frankly, almost really obvious. 0:46:41.960 --> 0:46:43.920 I'm not going to do any weird light to me 0:46:44.120 --> 0:46:47.640 esque ways of proving he's lying. He just doesn't ever 0:46:47.680 --> 0:46:51.719 get pushed into any depth. No one ever asks him 0:46:52.120 --> 0:46:56.000 really technical questions or even just a question like, Hey, Sam, 0:46:56.160 --> 0:46:58.120 did you work on any of the code at open 0:46:58.160 --> 0:47:01.279 AIY what did you work on? Yeah? I know you 0:47:01.320 --> 0:47:04.560 can't talk about the future, Sam, but how close are 0:47:04.560 --> 0:47:06.560 we actually to AGI? And if he says, ah, a 0:47:06.560 --> 0:47:10.480 few years, that's not specific enough, Sam, how about give 0:47:10.520 --> 0:47:15.640 me a ballpark? And then when he lies again, you say, okay, Sam, 0:47:15.719 --> 0:47:20.040 how do we get from generative AI to AGI? And 0:47:20.120 --> 0:47:22.760 when he starts waffling, say no, no, no, be specific, Sam. 0:47:24.280 --> 0:47:26.839 This is how you actually ask questions. And when you 0:47:26.920 --> 0:47:30.040 say things like this, by the way, to technical founders, 0:47:30.400 --> 0:47:34.080 they don't get worried. They don't offer skate. They may 0:47:34.080 --> 0:47:35.839 say I can't talk about this duty of legal things, 0:47:35.840 --> 0:47:39.640 which is fine, but they'll generally try and talk to you. 0:47:40.920 --> 0:47:44.640 Listen to any interview with any other technical AI person, 0:47:46.200 --> 0:47:49.120 Listen to them, and then listen to Sam Ortman. He's 0:47:49.200 --> 0:47:53.240 full of it. It's so obvious. And one deeply unfair 0:47:53.320 --> 0:47:55.799 thing with the value is there are people that get 0:47:55.840 --> 0:47:59.319 held to these standards. Early stage startups generally do the 0:47:59.360 --> 0:48:02.120 ones that aren't handed to people like Altman or Alexis 0:48:02.160 --> 0:48:05.080 Ohanian have read it, or Paul Graham or read Hoffman. 0:48:06.000 --> 0:48:08.360 They don't get those chances because they're not saying the 0:48:08.400 --> 0:48:12.200 things that need to be said to the venture capitalists. 0:48:12.360 --> 0:48:16.080 They're not in the circles. They're not doing the right 0:48:16.120 --> 0:48:19.239 things because the right things are no longer the right 0:48:19.360 --> 0:48:23.319 thing for the tech industry. And when all of this 0:48:23.440 --> 0:48:27.640 falls apart, Sam Almond's going to be fine. When this 0:48:27.840 --> 0:48:31.480 all collapses, he'll find something to blame it on market forces, 0:48:31.560 --> 0:48:36.000 a lack of energy, breakthroughs, unfortunate economic things. All are 0:48:36.080 --> 0:48:39.040 that nonsense, and he'll remain a billionaire, capable of doing 0:48:39.040 --> 0:48:42.280 anything he wants. The people that are going to suffer 0:48:42.400 --> 0:48:45.480 are the people working in Silicon Valley who aren't Sam Almon. 0:48:45.840 --> 0:48:48.480 The people that did not get born with a silver 0:48:48.520 --> 0:48:51.680 spoon in each hand and then handed further silver spoons 0:48:51.680 --> 0:48:55.360 as they walk the streets of San Francisco. People that 0:48:55.440 --> 0:48:58.440 don't live in nine and a half thousand square foot mansions. 0:48:59.480 --> 0:49:02.120 The people trying to raise money who can't right now 0:49:02.160 --> 0:49:06.440 because all the vcs are obsessed with AI. The people 0:49:06.520 --> 0:49:09.880 that will get fired from public tech companies when a 0:49:10.000 --> 0:49:13.319 depression hits, because the markets realize that the generative AI 0:49:13.440 --> 0:49:17.280 boom was a bubble, when they realized that the most 0:49:17.320 --> 0:49:20.640 famous people in tech have been making these promises for 0:49:20.800 --> 0:49:25.080 nobody other than the markets. Well, the markets need you 0:49:25.120 --> 0:49:27.520 to do something eventually, and I just don't think it's 0:49:27.560 --> 0:49:31.040 going to happen. And I think that we need to 0:49:31.120 --> 0:49:34.040 really think why was Sam Altman allowed to get to 0:49:34.080 --> 0:49:37.840 this point? Why did so many people like Paul Graham, 0:49:38.120 --> 0:49:42.239 like Reid Hoffman, like Brian Chesky, like Sachi Nadella. Back up, 0:49:42.440 --> 0:49:45.840 it's obvious con artists who has acted like this forever. 0:49:47.360 --> 0:49:49.560 And what sucks is I don't know if the value 0:49:49.600 --> 0:49:51.520 is going to learn anything unless it's really bad, and 0:49:51.560 --> 0:49:53.200 I don't want it to be. By the way, I 0:49:53.239 --> 0:49:55.399 would love to be wrong. I would love for all 0:49:55.440 --> 0:49:57.920 of this to just be like Sam Ortman's actually a genius. 0:49:58.400 --> 0:50:01.080 Turns out the whole thing is no, it's not going 0:50:01.160 --> 0:50:05.440 to happen. And I worry that there is no smooth 0:50:05.440 --> 0:50:07.239 way out of this, that there is no way to 0:50:07.360 --> 0:50:12.960 just casually integrate open Ai with Microsoft, because now there's 0:50:12.960 --> 0:50:16.680 an antitrust thing going in with Microsoft and acquiring Inflection Ai, 0:50:16.800 --> 0:50:20.799 another AI company, and that's the thing. It feels like 0:50:20.880 --> 0:50:25.560 we are approaching a press apiece here, and the only 0:50:25.600 --> 0:50:28.680 way to avoid it is for people to come clean, 0:50:28.719 --> 0:50:31.719 which is never going to happen, or of course, for 0:50:31.800 --> 0:50:35.040 Sam Wortman not to be lying, for AGI to actually 0:50:35.040 --> 0:50:36.839 come out of open Ai. And by the way, it's 0:50:36.840 --> 0:50:39.080 going to need to be in the next year. I 0:50:39.080 --> 0:50:42.560 don't think they've got even three quarters left. I think 0:50:42.600 --> 0:50:47.359 that once this falls apart, once the markets realize, oh shit, 0:50:47.520 --> 0:50:50.840 this is not profitable, this is not sustainable, they're going 0:50:50.920 --> 0:50:54.040 to walk away from it. When companies realize that Generative 0:50:54.080 --> 0:50:59.279 AI it's given him a couple percent profit, maybe they're 0:50:59.320 --> 0:51:02.600 going to be pissed because this is not a stock 0:51:02.719 --> 0:51:08.680 rally worthy boondoggle. This is not going to be pretty 0:51:09.520 --> 0:51:12.279 when things fall. Apart from for Nvidia, you're still over 0:51:12.280 --> 0:51:15.440 one thousand dollars when those orders stop coming in quite 0:51:15.440 --> 0:51:18.080 as fast. What do you think is going to happen 0:51:18.120 --> 0:51:22.880 to tech stocks? Startups are already having trouble raising money, 0:51:23.239 --> 0:51:26.440 and they're having trouble raising money because the people giving 0:51:26.440 --> 0:51:29.400 out the money are too disconnected from the creation of 0:51:29.440 --> 0:51:33.080 software and hardware. The only way to fix Silicon Valley 0:51:34.080 --> 0:51:37.640 perhaps is an apocalypse. Perhaps is people like Sam Ortman 0:51:37.760 --> 0:51:40.040 getting washed out. I don't want it to happen. I 0:51:40.120 --> 0:51:45.240 really must be bloody clear. But maybe it won't be apocalyptic. 0:51:46.400 --> 0:51:49.160 Maybe it would just be a brutal realignment. And maybe 0:51:49.200 --> 0:51:54.160 Silicon Valley needs that realignment because this industry desperately needs 0:51:55.080 --> 0:51:57.920 a big bathful of ice. I need to dunk the 0:51:58.040 --> 0:52:02.320 head in it aggressively and wake the hell up. Venture 0:52:02.360 --> 0:52:05.840 capital needs to put money back into real things. The 0:52:05.960 --> 0:52:09.360 largest tech companies need to realign and build for sustainability, 0:52:09.520 --> 0:52:13.480 so are not binging and purging staff with every boom, 0:52:13.719 --> 0:52:15.600 And if we really are at the end of the 0:52:15.680 --> 0:52:19.960 hypergrowth era, every tech company needs to be thinking profit 0:52:20.000 --> 0:52:23.240 and sustainability again. And that's a better silicon valley, because 0:52:23.239 --> 0:52:26.200 a better silicon valley builds things for people. It solves 0:52:26.239 --> 0:52:29.279 real problems. It doesn't have to lie about what the 0:52:29.320 --> 0:52:31.399 thing could do in the future so that it can 0:52:31.440 --> 0:52:34.680 sell a thing today. And I realize that sounds like 0:52:34.719 --> 0:52:38.200 the foundation of most venture capital. That's fine at the 0:52:38.239 --> 0:52:42.399 seed stage, that's fine at this moonshot stage where your 0:52:42.440 --> 0:52:46.040 early early days. It is not befitting the most famous 0:52:46.320 --> 0:52:50.200 company in tech. It is not befitting a multi billionaire. 0:52:51.280 --> 0:52:54.279 It is not befitting anyone, and it is insulting to 0:52:54.320 --> 0:52:57.520 the people actually building things, both in and outside of technology. 0:52:58.000 --> 0:53:01.400 The people I hear from after every episode, they are angry, 0:53:01.760 --> 0:53:04.480 They are frustrated because there are good people in tech. 0:53:04.520 --> 0:53:06.440 There are people building real things. There are people that 0:53:06.560 --> 0:53:10.319 remember a time when the tech industry was exciting, when 0:53:10.440 --> 0:53:13.279 people were talking about cool shit in the future, and 0:53:13.320 --> 0:53:17.080 then they'd actually do it. Returning to that is better 0:53:17.120 --> 0:53:20.440 for society and the tech industry. Just don't know when 0:53:20.480 --> 0:53:30.920 it's gonna happen. Thank you for listening to Better Offline. 0:53:31.040 --> 0:53:33.480 The editor and composer of the Better Offline theme song 0:53:33.520 --> 0:53:36.200 is Mattasowski. You can check out more of his music 0:53:36.200 --> 0:53:39.839 and audio projects at Mattasowski dot com m A T. 0:53:39.840 --> 0:53:40.440 T O. 0:53:40.680 --> 0:53:44.920 S O w Ski dot com. You can email me 0:53:44.960 --> 0:53:47.560 at easy at better offline dot com or visit better 0:53:47.600 --> 0:53:50.040 offline dot com to find more podcast links and of course, 0:53:50.080 --> 0:53:53.200 my newsletter. I also really recommend you go to chat 0:53:53.200 --> 0:53:55.879 dot Where's Youreed dot at to visit the discord, and 0:53:55.920 --> 0:53:58.640 go to our slash Better Offline to check out our reddit. 0:53:59.400 --> 0:54:00.760 Thank you so much much for listening. 0:54:01.560 --> 0:54:04.279 Better Offline is a production of cool Zone Media. For 0:54:04.400 --> 0:54:08.440 more from cool Zone Media, visit our website Coolzonemedia dot com, 0:54:08.520 --> 0:54:11.400 or check us out on the iHeartRadio app, Apple Podcasts, 0:54:11.480 --> 0:54:12.960 or wherever you get your podcasts.