WEBVTT - The Enshittifinancial Crisis: Part Two 0:00:02.800 --> 0:00:08.840 Zone media, I'm better off line and this is ed Zetron. Wait, 0:00:20.680 --> 0:00:23.480 that's right, folks. We're here for part two of our 0:00:23.560 --> 0:00:27.040 multi part series about the final casualty of the incertification 0:00:27.200 --> 0:00:30.560 process the financial markets. You see, we're in a world 0:00:30.560 --> 0:00:33.800 where because companies can no longer stimulate growth by just 0:00:33.880 --> 0:00:37.800 script mining their individual and business users, they've now resorted 0:00:37.840 --> 0:00:42.080 to just lying. And lying is an interesting word because 0:00:42.120 --> 0:00:44.800 they will claim we're not lying, we used words to 0:00:44.880 --> 0:00:49.080 vaguely avoid telling you the truth, or we were technically 0:00:49.120 --> 0:00:51.640 true in what we say. There are various kinds of 0:00:51.680 --> 0:00:54.160 lies that don't have to just be I told you 0:00:54.200 --> 0:00:58.240 something that was not categorically yes or no force. I mean, 0:00:58.720 --> 0:01:01.440 you've spoken to other people, but they will say this 0:01:01.440 --> 0:01:05.280 for legal purposes. So fine, it's not technically lying, it's 0:01:05.440 --> 0:01:08.399 just bullshit. But I think it's important to give you 0:01:08.440 --> 0:01:11.000 some examples and to talk about how analyst in the 0:01:11.040 --> 0:01:15.399 media simply allowed these lights to propagate unchecked. Let me 0:01:15.440 --> 0:01:18.800 set the scene. Our story begins on September tenth, twenty 0:01:18.800 --> 0:01:21.560 twenty five. I was for some reason in Los Angeles 0:01:21.560 --> 0:01:24.600 that I can't remember. Maybe a concert, maybe an interview. 0:01:24.800 --> 0:01:29.280 Truly don't know. Oracle announced its unfillable, unpayable, three hundred 0:01:29.319 --> 0:01:32.160 billion dollar deal with open AI, which they didn't have 0:01:32.160 --> 0:01:34.360 the capacity to serve, and open a I didn't have 0:01:34.400 --> 0:01:36.280 the cash to pay them, and this led to a 0:01:36.319 --> 0:01:39.840 thirty percent or more bump in stock price. Analysts who 0:01:40.000 --> 0:01:43.280 should extensibly be able to count, called it momentous and 0:01:43.319 --> 0:01:46.640 said they were in shock. On September twenty second, twenty 0:01:46.680 --> 0:01:52.240 twenty five, CEO Saffracats stepped down. She stepped down from oraclen. 0:01:52.680 --> 0:01:57.280 Nobody seemed to think that was weird or suspicious or worrying. 0:01:57.440 --> 0:01:59.920 This was this was meant to be the beginning of 0:02:00.080 --> 0:02:03.480 the AI like boom too. I mean, we kind of 0:02:03.520 --> 0:02:05.960 really in the AI bool, but this was meant to 0:02:06.040 --> 0:02:09.000 be the even more bigger one, right, Why would you 0:02:09.320 --> 0:02:11.720 step down? Why would you step down? Why would you 0:02:11.760 --> 0:02:14.560 step down? And then have two guys like two different 0:02:14.600 --> 0:02:18.040 They have co CEOs. Very sorry to my friend Jim, 0:02:18.120 --> 0:02:21.800 but co CEOs don't usually do well. Look at BlackBerry. 0:02:21.800 --> 0:02:25.000 But anyway, two months after that, Oracle stock was down 0:02:25.040 --> 0:02:28.720 forty percent, with investors worried about Oracle's growing capex, which 0:02:28.760 --> 0:02:31.200 is surprising. I suppose if you didn't think about how 0:02:31.280 --> 0:02:34.120 Oracle would build the data centers to serve the three 0:02:34.200 --> 0:02:39.320 hundred billion dollar deal. A makes me feel insane when 0:02:39.360 --> 0:02:42.320 I say these things out how loud, But just to 0:02:42.360 --> 0:02:47.000 be clear, anybody who traded on that got burned. Now 0:02:47.080 --> 0:02:50.160 another thing happened on September twenty second, twenty twenty five, 0:02:50.480 --> 0:02:53.560 in Vidio announced what they called a strategic partnership to 0:02:53.600 --> 0:02:56.919 invest up to one hundred billion dollars and build ten 0:02:57.040 --> 0:03:00.000 gigawads of data centers with open Ai, with the first 0:03:00.200 --> 0:03:02.400 to gigawatt to be deployed in the second half of 0:03:02.440 --> 0:03:06.480 twenty twenty six. Just a few questions from me personally, 0:03:06.760 --> 0:03:09.480 where would the data centers go, How would open Ai 0:03:10.000 --> 0:03:12.959 afford to build them? How would open Ai build a 0:03:13.000 --> 0:03:17.080 gigawa in data centers in less than a year? Don't 0:03:17.120 --> 0:03:21.000 ask questions. Pig back in your cage now. In video 0:03:21.080 --> 0:03:23.680 stock bumped from one hundred and seventy five point three 0:03:23.880 --> 0:03:25.720 to one hundred and eighty one dollars in the space 0:03:25.760 --> 0:03:28.080 of a day. Said they're very normally moving on and 0:03:28.120 --> 0:03:29.839 the media wrote about the story as if the deal 0:03:29.960 --> 0:03:32.360 was done, with CNBC claiming that and I quote the 0:03:32.400 --> 0:03:35.760 initial ten billion dollar tranch was expected to close within 0:03:35.800 --> 0:03:38.960 a month or so once the transaction had been finalized. 0:03:39.160 --> 0:03:41.760 I read at least ten stories that said in Video 0:03:41.800 --> 0:03:45.640 had invested one hundred billion dollars in open Ai. Analysts 0:03:45.640 --> 0:03:47.680 would go on to say that in Video was locking 0:03:47.720 --> 0:03:50.400 in open Ai and I quote again to remain the 0:03:50.400 --> 0:03:54.120 backbone of the next gen AI infrastructure, and that demand 0:03:54.160 --> 0:03:57.240 for in video GPUs is effectively baked into the development 0:03:57.280 --> 0:04:00.960 of frontier AI models, and that the deal, the partnership 0:04:00.960 --> 0:04:04.000 between the two companies and Validate did in Vidia's long 0:04:04.080 --> 0:04:07.280 term growth numbers with so much volume and compute capacity. 0:04:07.800 --> 0:04:10.200 Others would say that in Video was enabling open Ai 0:04:10.280 --> 0:04:14.320 to meet surging demand. Three analysts, rask Gone at Bernstein Laurier, 0:04:14.480 --> 0:04:17.640 DA Davidson, and Wagner at Aaptors Capital all raised circular 0:04:17.680 --> 0:04:19.920 deal concerns, but they were in the minority, and those 0:04:19.920 --> 0:04:22.880 concerns were still often buried under buoyant optimism around the 0:04:22.880 --> 0:04:26.640 prospects of the company, and nobody just fucking saying the words, Hey, 0:04:26.760 --> 0:04:29.680 no one could afford any of the Oracle can't afford 0:04:29.720 --> 0:04:32.000 to build it, they don't have the capacity, they ust 0:04:32.040 --> 0:04:35.440 can't get paid by open Ai. Open Ai doesn't have 0:04:35.480 --> 0:04:39.640 the money. But in Vidia, this deal also doesn't make sense. 0:04:39.720 --> 0:04:42.159 You know. It feels like open ai is just promising 0:04:42.200 --> 0:04:46.320 everyone's stuff they can't afford. All of these things seemed 0:04:46.360 --> 0:04:48.679 like very obvious questions, but people just kind of clapped 0:04:48.680 --> 0:04:50.840 their hands and went, fuck, yeah, we're doing Ai even 0:04:50.880 --> 0:04:54.240 more now. But then there was this other thing about 0:04:54.240 --> 0:04:58.359 the Nvidia deal. It's this quincy teeny weeny quincy little 0:04:58.400 --> 0:05:02.800 problem one little pro everyone about the Invidia deal. The 0:05:02.839 --> 0:05:05.440 deal that everyone said was done. It was a letter 0:05:05.480 --> 0:05:08.560 of intent. It said so in the announcement and on 0:05:08.600 --> 0:05:11.680 Invidia's November runnings it said that it had entered into 0:05:11.680 --> 0:05:15.520 a letter of intent with an opportunity to invest in 0:05:15.560 --> 0:05:18.880 open Ai. It turns out the deal didn't exist, and 0:05:18.960 --> 0:05:22.800 everybody fell for him. Invidia hasn't sent a diamond, likely won't. 0:05:23.120 --> 0:05:25.359 A letter of intent is also known as the concept 0:05:25.400 --> 0:05:28.560 of a plan. A little over two weeks later, on 0:05:28.600 --> 0:05:32.240 October first, twenty twenty five, Reuters reported that Samsung and 0:05:32.400 --> 0:05:35.520 sk Heinex had signed letters of intent to supply memory 0:05:35.600 --> 0:05:39.640 chips for open AI's data centers, with South Korea's Presidential 0:05:39.640 --> 0:05:42.960 Office saying that said chip demand was expected to reach 0:05:43.279 --> 0:05:46.520 nine hundred thousand wafers a month, with and I quote 0:05:46.640 --> 0:05:49.560 much of that from Samsung and esk Heinex, which was 0:05:49.640 --> 0:05:53.800 quickly extrapolated to meet around forty percent of global dram 0:05:53.839 --> 0:05:57.800 outpum stocks in both companies to quote Reuter's Sword, with 0:05:57.920 --> 0:06:00.680 Samsung climbing four percent and sk Heinex more than twelve 0:06:00.680 --> 0:06:04.440 percent to an all time high. Analyst Jeff Kim KB 0:06:04.560 --> 0:06:07.560 Securities said that and I quote, there have been worries 0:06:07.600 --> 0:06:11.880 about high bandwidth memory prices falling next year on intensifying competition, 0:06:11.960 --> 0:06:14.640 but such worries will now be easily resolved by the 0:06:14.760 --> 0:06:18.480 strategic partnership, adding that since Stargate is a key project 0:06:18.560 --> 0:06:21.680 led by President Trump, wrong lie, there also is a 0:06:21.720 --> 0:06:24.920 possibility that partnership will have a positive impact on South 0:06:25.000 --> 0:06:28.839 Korea's trade negotiations with the US, which is just to 0:06:28.880 --> 0:06:33.839 be clear, total and utter bollocks. It's wank, it's scrap, 0:06:34.040 --> 0:06:37.800 it's nonsense, boop them above all the technical terms. Donald 0:06:37.839 --> 0:06:40.840 Trump is not leading Stargate. Stargate is a name used 0:06:40.880 --> 0:06:43.440 to refer to data centers built by open Ai, by 0:06:43.440 --> 0:06:45.719 which I mean other people building them for open Ai. 0:06:46.279 --> 0:06:49.120 KB Securities is around forty three billion dollars of assets 0:06:49.200 --> 0:06:52.360 under management. This is the level of analysis you get 0:06:52.360 --> 0:06:56.440 from these analysts. This is how much they know. There 0:06:56.480 --> 0:07:00.200 are people that study JoJo's bizarre adventure that have more 0:07:00.240 --> 0:07:04.520 economic analysis than any of these fucking people. Most Star 0:07:04.600 --> 0:07:07.240 Wars fans who have just memorized all the names of 0:07:07.279 --> 0:07:09.800 the glop shittos of the world have better analysis than this. 0:07:11.280 --> 0:07:14.280 But nevertheless, you'd think, right that these companies have made 0:07:14.280 --> 0:07:16.600 all this noise, and they've talked about this stuff, the 0:07:16.640 --> 0:07:19.120 Presidential Office all involved, you think we'd be able to 0:07:19.240 --> 0:07:22.559 get a whiff of that on their earning scores, right well. 0:07:23.400 --> 0:07:26.560 On I sk Hihenex's October twenty ninth, twenty twenty five 0:07:26.560 --> 0:07:30.480 earning score. Weeks after the announcement, it's CEO Kim Wu 0:07:30.640 --> 0:07:33.480 Yun was asked a question about high bandwidth memory growth 0:07:33.480 --> 0:07:38.120 by Eskkim from Daiwa Securities and I quote, so this 0:07:38.280 --> 0:07:41.120 is this is SKKM asking the question. Thank you very 0:07:41.200 --> 0:07:43.720 much for taking my question. It is on demand now. 0:07:43.760 --> 0:07:45.960 There have been a series of announcements of GPU and 0:07:46.000 --> 0:07:49.240 a six supply cooperation between big techs and AI companies, 0:07:49.480 --> 0:07:53.080 fueling expectations of further AI market growth. Then, against this 0:07:53.160 --> 0:07:56.040 back drop, what is the company's outlook on high bend 0:07:56.080 --> 0:07:59.360 with memory demand as well as a broadening of the 0:07:59.360 --> 0:08:03.040 customer base, es sk Heinex. Thank you for the question. Now, 0:08:03.080 --> 0:08:06.080 with upward adjustment in big text CAPEX and increasing investment 0:08:06.280 --> 0:08:09.400 by AI companies, the high bandwidth memory market, even by 0:08:09.400 --> 0:08:11.800 a conservative vestmer, will keep growing at an average of 0:08:12.360 --> 0:08:14.960 over thirty percent for the next five years. I will 0:08:15.000 --> 0:08:17.520 point to our recent letter of intent with open Ai 0:08:17.680 --> 0:08:20.800 for large scale DRAM supply as an example of the 0:08:20.920 --> 0:08:24.040 very strong demand for AI as well as the need 0:08:24.080 --> 0:08:26.840 to secure AI memory based on HBM more than anything 0:08:26.840 --> 0:08:31.480 else went developing AI technology. That is it. That is 0:08:31.520 --> 0:08:35.760 the only mention of open AI. That's it. No other 0:08:35.800 --> 0:08:39.720 mentions otherwise, sk Heinex has not added any guidance that 0:08:39.760 --> 0:08:42.560 would suggest that it's DRAM sales will spike beyond overall 0:08:42.600 --> 0:08:44.640 growth where it's already grown because everyone's running out of 0:08:44.679 --> 0:08:47.640 ramthanks to fucking AI. Other than mentioning that it had 0:08:47.720 --> 0:08:50.640 an I quote completed year twenty twenty six supply discussions 0:08:50.679 --> 0:08:53.240 with key customers, there is no mention of open AI 0:08:53.400 --> 0:08:56.600 in any earning's presentation. Now You may think there were 0:08:56.600 --> 0:08:59.000 two companies. You said two company names. There were Samsung 0:08:59.040 --> 0:09:02.200 as well. Right well. On Samsung's October thirtieth, twenty twenty 0:09:02.240 --> 0:09:05.480 five earning school Samsung mentioned the term d RAM eighteen 0:09:05.520 --> 0:09:08.400 times and neither mentioned open AI nor any letters of 0:09:08.400 --> 0:09:11.400 intent of any kind. In its Q three twenty twenty 0:09:11.400 --> 0:09:15.199 five earnings presentation, Samsung mentions it will a quote prioritize 0:09:15.200 --> 0:09:18.520 the expansion of high bandwidth memory four business with differentiate 0:09:18.559 --> 0:09:23.200 performance to address increasing AI demand. Great place to mention 0:09:23.280 --> 0:09:26.360 open AI. Right, yeah, that would be if the money 0:09:26.400 --> 0:09:29.480 existed and they were gonna do anything Now, analysts do 0:09:29.520 --> 0:09:31.360 not appear to have noticed the lack of revenue from 0:09:31.400 --> 0:09:33.439 an apparent deal for forty percent of the world's RAM. 0:09:33.720 --> 0:09:37.320 Oh well, Pobody's perfect right. My frustration, you hear in 0:09:37.320 --> 0:09:40.280 my voice is that I'm a guy with a Google doc. 0:09:41.040 --> 0:09:44.120 I'm not a financial analyst, but I appeared to be 0:09:44.120 --> 0:09:45.960 able to do what they're doing because I'm able to 0:09:46.000 --> 0:09:50.480 fucking read a Both Samsung and s k Heinez's stocks 0:09:50.520 --> 0:09:52.600 have continued to write since, and you'd be forgiven for 0:09:52.640 --> 0:09:54.760 thinking this deal was something to do with it, even 0:09:54.800 --> 0:09:59.360 though it was not. But silly season was not yet over, however, 0:09:59.400 --> 0:10:02.120 and on October fifth, twenty twenty five, AMD announced that 0:10:02.120 --> 0:10:04.560 it had entered a multi year, multi generation agreement with 0:10:04.600 --> 0:10:07.760 open Ai to build six goddamn gigawads of data centers, 0:10:07.840 --> 0:10:10.000 with the first one gig what deployment said to begin 0:10:10.080 --> 0:10:12.440 in the second half of twenty twenty six, calling the 0:10:12.480 --> 0:10:15.240 agreement definitive with terms that allowed open ai to buy 0:10:15.480 --> 0:10:19.400 up to ten percent of AMD stock, festing over specific 0:10:19.480 --> 0:10:21.640 milestones that started with the first giga what of data 0:10:21.640 --> 0:10:24.600 center development, said data centers would use AMDs yet to 0:10:24.600 --> 0:10:27.559 be released four to fifty GPUs. The deal would per 0:10:27.640 --> 0:10:31.760 reuters bring in tens of billions of dollars of revenue. Now, 0:10:32.440 --> 0:10:34.920 I know, hate to be an asshole, right, I know, 0:10:35.080 --> 0:10:37.719 I'm always on this shit, right Yeah? Where would those 0:10:37.800 --> 0:10:41.160 data centers go? How much? How much would they cost? 0:10:41.160 --> 0:10:43.400 How would open Ai pay for them? Would the jips 0:10:43.440 --> 0:10:47.040 be ready in time? Silence? Well, how dare you ask questions? 0:10:47.040 --> 0:10:49.680 How dare you? Why are you asking questions? Number? Go up? 0:10:50.120 --> 0:10:53.080 Number go up? Look look at number? Go up. AMD 0:10:53.240 --> 0:10:56.560 shares searched by thirty four percent, with the analyst Dan 0:10:56.600 --> 0:10:58.880 Ives of Webbush, who appears to dress like the San 0:10:58.960 --> 0:11:02.880 Diego Padres Didy connect Jerseys, saying that this was a 0:11:03.000 --> 0:11:06.480 major evaluation moment for AMD. As in the side, I've 0:11:06.520 --> 0:11:08.400 said that in video would benefit from the meta verse 0:11:08.440 --> 0:11:10.959 in twenty twenty one, and told CBS News in November 0:11:11.320 --> 0:11:14.120 on November twenty second, twenty twenty one, that and I quote, 0:11:14.160 --> 0:11:17.040 the metaverse was real and that Wall Street was looking 0:11:17.080 --> 0:11:22.800 for winners, were they dan? Now one more thing, that 0:11:22.840 --> 0:11:26.280 AMD's November earnings a month after that announcement might be 0:11:26.280 --> 0:11:29.280 a barn burner full of remaining performance obligations from Open AI. 0:11:29.679 --> 0:11:32.000 Things that Open appaying them moost tens of billions of 0:11:32.000 --> 0:11:35.560 dollars of revenues right. In fact, CEO Lisa Sun who 0:11:35.760 --> 0:11:39.400 said that AMD expected this partnership will significantly accelerated Stata 0:11:39.440 --> 0:11:42.160 Center AI business and with the potential to generate well 0:11:42.200 --> 0:11:44.160 over one hundred billion dollars in revenue over the next 0:11:44.160 --> 0:11:47.559 few years in the announcement, and here's how AMD's ten 0:11:47.640 --> 0:11:50.440 Q filing referred to it, and I quote. As of 0:11:50.440 --> 0:11:53.960 September twenty seventh, twenty twenty five, the Aggagut transaction price 0:11:54.200 --> 0:11:56.880 allocated to remaining performance applications on the contracts with an 0:11:56.920 --> 0:12:00.120 original expected duration of more than one year. It was 0:12:00.120 --> 0:12:02.160 two hundred and seventy nine million dollars, of which one 0:12:02.240 --> 0:12:04.320 hundred and thirty nine million dollars is expected to be 0:12:04.360 --> 0:12:07.160 recognized in the next twelve months. The revenue allocated to 0:12:07.200 --> 0:12:11.000 remaining performance applications does not include amounts whichever original expected 0:12:11.040 --> 0:12:15.560 duration of one year or less. Now, I don't know 0:12:15.600 --> 0:12:18.720 about you. I did not I didn't do economics in school, 0:12:18.800 --> 0:12:21.760 didn't do anythin like that, didn't do well at maths 0:12:21.760 --> 0:12:25.559 in school. But two hundred and seventy nine million dollars 0:12:25.720 --> 0:12:30.640 seems lower than over one hundred billion dollars. So I 0:12:30.640 --> 0:12:34.880 guess just just no revenue from open AI, no revenue 0:12:34.920 --> 0:12:39.079 of any kind. I guess. Now, AMD did raise guidance 0:12:39.080 --> 0:12:41.920 by thirty five percent of the next five years, and 0:12:42.200 --> 0:12:44.880 they're trailing. Twelve month revenue is thirty two billion dollars. 0:12:45.040 --> 0:12:47.559 Tens of billions of dollars would surely lead to more 0:12:47.559 --> 0:12:50.520 than a thirty five percent boost, because that would be 0:12:50.600 --> 0:12:54.440 like eleven billion. That's just eleven billion dollars. That's a 0:12:54.480 --> 0:12:58.280 lot less. That's a lot less than one hundred billion dollars. 0:12:58.280 --> 0:13:02.280 I don't know guess all that was for nothing. No 0:13:02.360 --> 0:13:04.520 follow up from the media, no questions from analysts, just 0:13:04.520 --> 0:13:07.040 the shrug and we all move on. They were at ces. 0:13:07.160 --> 0:13:11.160 Greg Brockman came on AMD's Lisa Sue said something about 0:13:11.240 --> 0:13:14.960 Yachta flops, which really just pissed me off to think about. 0:13:16.360 --> 0:13:19.520 Still nothing anyway. AMD stock is now down from a 0:13:19.559 --> 0:13:21.160 high of two hundred and fifty nine bucks at the 0:13:21.240 --> 0:13:24.880 end of October to around let's see md stocks understand 0:13:25.000 --> 0:13:28.040 about two hundred and thirty one bucks right now. Everybody 0:13:28.040 --> 0:13:30.160 who traded in based on an analyst and media comments 0:13:30.200 --> 0:13:44.360 got fucked. Now let's go to my favor them, and 0:13:44.360 --> 0:13:48.800 that's a little company named Broadcom. So back on September fifth, 0:13:48.800 --> 0:13:50.840 Broadcom send and it's earning school that it had a 0:13:50.880 --> 0:13:54.640 ten billion dollar order from a mysterious customer, which analysts 0:13:54.679 --> 0:13:57.760 quickly assumed was open Ai, leading to Broadcom stock probably 0:13:57.800 --> 0:13:59.640 nine percent and gradually increasing to a high of a 0:13:59.679 --> 0:14:02.679 three hundred and sixty nine dollars on September tenth, before 0:14:02.720 --> 0:14:06.080 declining a little until October thirteenth, when Broadcom announced a 0:14:06.200 --> 0:14:09.240 ridiculous ten gigawa deal with open Ai, claiming that it 0:14:09.280 --> 0:14:11.720 would deploy ten gigawats of open Ai designed chips, with 0:14:11.760 --> 0:14:14.360 the first ranks to deploy the second half of twenty 0:14:14.360 --> 0:14:16.880 twenty six and the entire deployment completed by the end 0:14:16.880 --> 0:14:19.400 of twenty twenty ninth. So three fucking years for ten 0:14:19.440 --> 0:14:22.640 fucking gigawatts. Just to give you an idea, it's about 0:14:22.760 --> 0:14:24.880 two and a half years a gigawat right now. And 0:14:24.920 --> 0:14:27.200 that's for open Ai slush fun projects, the ones that 0:14:27.240 --> 0:14:29.920 have all the money and none of the oversight. Now. 0:14:29.960 --> 0:14:33.520 The same day, President of Semiconductor Solutions Charlie Cowas added 0:14:33.520 --> 0:14:38.240 that said mystery customers actually not open Ai, and I quote, 0:14:38.440 --> 0:14:40.640 I would love to take a ten billion dollar purchase 0:14:40.760 --> 0:14:43.880 order from my good friend Greg Brockman, CEO of open Ai, 0:14:44.040 --> 0:14:48.520 Cowas said he has not given me that purchase order yet. Nevertheless, 0:14:48.520 --> 0:14:50.960 Broadcom stock popped nine percent on the news about the 0:14:50.960 --> 0:14:53.440 ten gigawa deal with CMBC, adding that the companies have 0:14:53.560 --> 0:14:56.440 been working together for eighteen months because it's open Ai. 0:14:56.560 --> 0:15:00.000 Nobody sat and thought about whether somebody at Broadcom was saying, wow, 0:15:00.000 --> 0:15:02.720 well open has yet to ward these chips, yeah, and 0:15:02.760 --> 0:15:04.960 whether that was a problem. In fact, the answer to 0:15:05.200 --> 0:15:08.360 how does open Ai thought this appeared to be they'd 0:15:08.360 --> 0:15:11.720 afford it when it came to analysts. Now I'm gonna 0:15:11.720 --> 0:15:14.360 I'm about to read you a quote that is so 0:15:14.560 --> 0:15:17.160 funny when you realize that this is someone who makes 0:15:17.240 --> 0:15:20.680 like a shit ton of money. Ahem. The twenty twenty 0:15:20.720 --> 0:15:22.960 six timelines set out by open Ai for the buildout 0:15:23.000 --> 0:15:25.640 is aggressive, but the startup is also best position to 0:15:25.720 --> 0:15:28.600 raise funds required for the project given the heights of 0:15:28.640 --> 0:15:33.680 investor confidence, said gad Joe Sevilla, an analyst eMarketer. Financing 0:15:33.720 --> 0:15:36.160 such a large chip deal will likely require a combination 0:15:36.240 --> 0:15:40.360 of funding grounds pre orders, strategic investments and support from Microsoft, 0:15:40.480 --> 0:15:46.640 as well as leveraging future revenue streams and potential credit facilities. Hogwash, 0:15:47.200 --> 0:15:53.880 just complete, complete bollocks. Insane that people pay these analysts bullshit. Okay, 0:15:53.920 --> 0:15:56.800 So just let's just break this down financing such a 0:15:56.840 --> 0:15:58.840 large chip deal. So ten gigawats, you're going to be 0:15:58.880 --> 0:16:01.080 looking at fifty bill dollars a gig was you look 0:16:01.120 --> 0:16:03.640 at that half a trillion dollars. So to do this 0:16:03.960 --> 0:16:08.520 for this aggressive buildout, which was ten giga Watson three years, 0:16:08.600 --> 0:16:12.560 just not possible. They would fund this by pre orders, 0:16:13.240 --> 0:16:17.200 pre orders for what exactly, because pre orders would be 0:16:17.200 --> 0:16:20.640 then pre ordering something from Broadcom. Not really sure how 0:16:20.720 --> 0:16:25.200 that fixed the problem. Strategic investments from whom just funding 0:16:25.240 --> 0:16:28.360 I guess, support from Microsoft funding again, I guess, And 0:16:28.400 --> 0:16:30.680 I don't know why Microsoft be paying for that, as 0:16:30.680 --> 0:16:36.560 well as leveraging future revenue streams and potential credit for 0:16:36.640 --> 0:16:42.160 say what what does that mean? Saving money doing like 0:16:42.560 --> 0:16:45.920 a revenue trade deal? Like what are you talking about? Anyway, 0:16:45.960 --> 0:16:48.240 you don't need to worry because open ai solution to 0:16:48.280 --> 0:16:51.560 this was far simpler. It didn't order anything. During Broadcom's 0:16:51.640 --> 0:16:54.400 November earning school, where Broadcom revealed that the ten billion 0:16:54.440 --> 0:16:57.040 dollar order was actually from Anthropic, another l of them 0:16:57.040 --> 0:17:00.240 started that burnst billions of dollars, which was actually buying 0:17:00.280 --> 0:17:05.880 Google's TPUs from Broadcom, and he also added the Anthropic 0:17:05.920 --> 0:17:08.679 had made a second eleven billion dollar deal. Alice somehow 0:17:08.680 --> 0:17:12.320 believed that Anthropic is positioned to spend heavily despite being 0:17:12.359 --> 0:17:15.280 yet another venture backed welfare recipient in the same flavor 0:17:15.280 --> 0:17:21.119 as open Ai. Oh right, right, sorry, okay, Broadcom. You 0:17:21.119 --> 0:17:23.000 may be wondering about the ten gig what deal though, 0:17:23.000 --> 0:17:26.240 because those earnings right well, Broadcom CEO hoc Tan said 0:17:26.240 --> 0:17:28.560 that he did not expect much in twenty twenty six 0:17:28.600 --> 0:17:32.080 from the deal, and guidance did not change to reflect it. Now, 0:17:32.119 --> 0:17:34.119 what was great about this as well? And I linked 0:17:34.119 --> 0:17:37.040 to this in the newsletter. I really need you to 0:17:37.080 --> 0:17:39.760 go and look up how hoc Tan had the conversation 0:17:39.840 --> 0:17:42.560 because someone asked him for more detail and he just 0:17:42.560 --> 0:17:44.840 straight up said, like, I don't want to talk about it. 0:17:44.840 --> 0:17:48.080 It rocks. I honestly fucking love that. Anyway, Broadcom climbed 0:17:48.080 --> 0:17:50.920 to a higher four hundred and twelve dollars a share 0:17:51.040 --> 0:17:52.840 leading up to its earnings, and I imagine it did 0:17:52.840 --> 0:17:54.680 so based on people trading on the belief that open 0:17:54.720 --> 0:17:57.080 Air and Broadcom were doing a deal of some sort together, 0:17:57.200 --> 0:18:00.439 which does not appear to be happening. While there is 0:18:00.440 --> 0:18:04.440 an alleged seventy three billion dollar backlog for Broadcom, every 0:18:04.480 --> 0:18:07.600 dollar from Anthropic is utterly questionable. Now, some of you 0:18:07.720 --> 0:18:11.800 might say, ed, we can't just automatically distrust public companies. Actually, 0:18:11.920 --> 0:18:15.280 yes we can. I'm doing it right now. Whenever a 0:18:15.359 --> 0:18:17.480 company says a litter of intent, as in Video and 0:18:17.600 --> 0:18:20.800 sk Heinez and Samsung did, it's important to immediately stop 0:18:20.800 --> 0:18:23.359 taking the deal. Seriously until you get to a specific 0:18:23.400 --> 0:18:28.400 word contract, not agreement or deal or announcement, but contract, 0:18:28.400 --> 0:18:33.040 because contracts are the only thing that actually matters. A 0:18:33.080 --> 0:18:37.200 great example of this is in in Video's last earnings, 0:18:37.200 --> 0:18:39.760 they said they would be investing up to ten billion 0:18:39.800 --> 0:18:42.879 dollars in Anthropic. Right, you'll notice that that was in 0:18:42.920 --> 0:18:45.480 the middle of November. It's now January and we haven't 0:18:45.520 --> 0:18:49.280 heard boo about it. It's probably because nothing's fucking happening. 0:18:49.520 --> 0:18:52.399 And also that phrase up to as well means that 0:18:52.600 --> 0:18:55.000 it will be However much in Video wants to invest, 0:18:55.320 --> 0:18:59.880 if at all. They only just did their Intel investment. Anyway, 0:19:00.080 --> 0:19:03.359 whenever they say these terms, you need to just get suspicious. 0:19:03.400 --> 0:19:06.280 It's time. It's time to stop trusting them, because they've 0:19:06.280 --> 0:19:09.639 played enough hide the sausage for one EPOC. I think 0:19:09.960 --> 0:19:13.160 it's time for everybody, analysts, the media, members of Congress, 0:19:13.160 --> 0:19:15.560 the fucking pope. I don't care, to start treating these 0:19:15.600 --> 0:19:20.440 companies with actual suspicion and to start demanding timelines like 0:19:20.480 --> 0:19:23.080 I said. In Video and Microsoft announced that fifteen billion 0:19:23.119 --> 0:19:26.479 dollar investment, and in Anthropic, where's the money? Why does 0:19:26.520 --> 0:19:29.600 the agreement say up to ten billion for in Video 0:19:29.680 --> 0:19:33.359 and up to five billion for Microsoft. Where's the money? 0:19:34.720 --> 0:19:37.480 Where is it? We haven't heard shit about it since. 0:19:37.800 --> 0:19:40.000 What we have heard is that Anthropic is raising either 0:19:40.040 --> 0:19:42.080 ten billion dollars or twenty five billion dollars at a 0:19:42.080 --> 0:19:45.000 three hundred and fifty billion dollar valuation. Where'd the money go? 0:19:45.440 --> 0:19:49.200 Where the money go from? Those fuck nuts? But they go, Sorry, 0:19:49.480 --> 0:19:51.880 Jensen won for calling you a fuck n All right, 0:19:52.320 --> 0:19:56.800 you're acting like one though, mate. Anyway, these deals are 0:19:56.840 --> 0:19:59.600 announced with the intention of suggesting there is more revenue 0:19:59.760 --> 0:20:04.000 and money and generative AI that actually exists. Furthermore, it 0:20:04.080 --> 0:20:07.640 is irresponsible and actively harmful for analysts in the media 0:20:07.680 --> 0:20:10.360 to continually act as if these deals will actually get 0:20:10.400 --> 0:20:13.639 paid when you consider the financial conditions of these companies. 0:20:14.440 --> 0:20:16.919 As part of its alleged funding announcement within Video and 0:20:16.960 --> 0:20:20.080 Microsoft and Thropic also agreed to purchase thirty billion dollars 0:20:20.240 --> 0:20:22.640 as your Compute, which is, for those of you who 0:20:22.680 --> 0:20:26.159 don't know, the compute cloud platform, the competitors to Amazon 0:20:26.200 --> 0:20:29.760 Web Services and Google Cloud that Microsoft runs. It also 0:20:29.800 --> 0:20:32.879 agreed to Anthropic agreed to spend tens of billions of 0:20:32.880 --> 0:20:35.639 dollars with Google Cloud. They ordered ten billion dollars of 0:20:35.720 --> 0:20:38.240 chips from broad Coom earlier in twenty twenty five, and 0:20:38.280 --> 0:20:41.320 apparently another eleven billion dollars of them in the last 0:20:41.320 --> 0:20:47.159 fiscal quarter. How does Anthropic pay for them? How Anthropic 0:20:47.200 --> 0:20:50.359 allegedly burned two point eight billion dollars last year twenty 0:20:50.400 --> 0:20:52.800 twenty five, though I think they burn much more and 0:20:52.880 --> 0:20:56.280 raise sixteen point five billion dollars in funding before Microsoft 0:20:56.320 --> 0:21:00.520 and then Videa's imaginary money. How are investors tolerating Broadcom 0:21:00.520 --> 0:21:03.479 not directly stating the future financial condition of this company 0:21:03.520 --> 0:21:06.640 is questionable? Has Broadcom created a reserve for this deal? 0:21:07.000 --> 0:21:09.639 If not, why not? Anthropic will make no more than 0:21:09.680 --> 0:21:11.840 five billion dollars in twenty twenty five, I'm sure of it, 0:21:12.080 --> 0:21:15.040 and raised about seventeen point five billion dollars with another 0:21:15.119 --> 0:21:18.000 two point five billion dollars coming in debt. How can 0:21:18.000 --> 0:21:20.680 it foreseeably afford to pay ten billion, eleven billion or 0:21:20.720 --> 0:21:24.119 twenty one billion dollars considering it's already massive losses and 0:21:24.160 --> 0:21:27.480 all those other obligations I mentioned. Will Jensen one hand 0:21:27.480 --> 0:21:29.600 over ten billion dollars so that Anthropic can hand it 0:21:29.640 --> 0:21:33.600 straight to Broadcom? I don't think so. I realize the 0:21:33.640 --> 0:21:35.840 counter argument is that the companies aren't responsible for the 0:21:35.880 --> 0:21:38.560 counterparty's financial health. But my argument is that it's the 0:21:38.600 --> 0:21:41.680 responsibility of any public company to give a realistic view 0:21:41.680 --> 0:21:44.600 of its financial health and revenue sources, which includes not 0:21:44.720 --> 0:21:46.560 to give a chunk of its revenues from a startup 0:21:46.600 --> 0:21:48.760 that can't afford to pay for its orders. There's no 0:21:48.840 --> 0:21:51.960 counter for that. Anthropic cannot afford to pay broad Com 0:21:51.960 --> 0:21:54.600 ten billion dollars right now or eleven billion dollars in 0:21:54.600 --> 0:21:57.240 a few months. Where is the money coming from. They're 0:21:57.240 --> 0:22:00.360 allegedly raising twenty five billion dollars. Great that ac shud 0:22:00.400 --> 0:22:04.080 lead them with four billion dollars to pay for what 0:22:04.520 --> 0:22:09.119 like thirty forty fifty billion dollars of compute costs. I 0:22:09.160 --> 0:22:13.120 don't know. Again, I ain't no mathematician, but I can 0:22:13.160 --> 0:22:16.080