WEBVTT - Monologue: The Pale Horsemen Arrive 0:00:03.160 --> 0:00:08.319 Media, and I looked in Behold a Pale Horse, and 0:00:08.360 --> 0:00:11.720 the writer's name was the Smiling Man. This is your 0:00:11.720 --> 0:00:14.800 weekly Better Offlying monologue, and I'm your host ed Zitron. 0:00:26.840 --> 0:00:29.000 This week I put out one of my most consequential 0:00:29.040 --> 0:00:32.199 newsletters of all time, called the AI Demand Bubble, pulling 0:00:32.240 --> 0:00:36.360 together multiple distinct financial analyst notes from Wells Fargo, Barclay's, 0:00:36.400 --> 0:00:39.720 and UBS that directly estimated that seventy percent or more 0:00:39.760 --> 0:00:42.640 of the AI revenues of Microsoft, Google, and Amazon were 0:00:42.680 --> 0:00:46.240 from either open Ai or Anthropic. To be clear, UBS 0:00:46.360 --> 0:00:49.520 estimated the next year, Anthropic and open AI's compute spend 0:00:49.560 --> 0:00:52.640 will be forty eight percent of all Google Cloud revenues, 0:00:53.120 --> 0:00:54.920 which means that they will likely account for even more 0:00:54.960 --> 0:00:57.760 than seventy percent. But I wanted to be fair. This 0:00:57.920 --> 0:00:59.680 was both the colossal pain in the ars and the 0:00:59.720 --> 0:01:01.240 story that I knew woud piss off a lot of 0:01:01.240 --> 0:01:04.800 people because of its huge ramifications. As a result, I've 0:01:04.800 --> 0:01:06.840 had a ton of pushback from people that either outright 0:01:06.880 --> 0:01:09.240 deny that this is the case, or say, actually it's good. 0:01:09.560 --> 0:01:12.800 The two one sustainable companies are the majority of AI 0:01:12.880 --> 0:01:15.560 revenues for companies that spent over a trillion dollars. Actually, 0:01:15.600 --> 0:01:19.280 that's great anyway. Twenty four hours later, Bloomberg ran a 0:01:19.319 --> 0:01:22.479 story estimating, based on open AI's twenty four point one 0:01:22.480 --> 0:01:25.600 billion dollar contribution to Microsoft's fiscal year twenty twenty six 0:01:25.680 --> 0:01:29.280 revenues and previous statements, the open ai alone contributed to 0:01:29.319 --> 0:01:32.479 seventy percent or more of Microsoft's AI revenues for the year. 0:01:33.160 --> 0:01:36.160 For some context, Microsoft has spent twenty six point one 0:01:36.440 --> 0:01:40.440 three billion dollars in capital expensires since the beginning of 0:01:40.440 --> 0:01:43.080 twenty twenty two. And other than getting to have the 0:01:43.120 --> 0:01:45.720 swagger of someone who just laid out a seven thousand 0:01:45.800 --> 0:01:48.960 worth thesis founded in deep research and hard numbers, only 0:01:49.000 --> 0:01:51.040 to get their work validated in less than a day, 0:01:51.240 --> 0:01:53.920 this story also confirms my greatest fears about the state 0:01:53.920 --> 0:01:56.800 of the AI industry and the actual demand for AI compume. 0:01:57.520 --> 0:02:00.640 Microsoft has spent over two hundred billion dollar dollar and 0:02:00.680 --> 0:02:03.200 that's not including the thirteen billion and it invested in 0:02:03.240 --> 0:02:05.840 open ai or the losses it's incurred running their services 0:02:05.840 --> 0:02:09.080 at cost at least in twenty twenty four, but potentially beyond, 0:02:09.360 --> 0:02:12.000 to create a customer that is now worth seven percent 0:02:12.320 --> 0:02:15.600 of its annual revenue. That cannot sustain its existence without 0:02:15.680 --> 0:02:19.440 near constant flows of venture capital funding. Microsoft's year over 0:02:19.520 --> 0:02:21.960 year growth for fiscal year twenty twenty six was about 0:02:22.000 --> 0:02:25.320 seventeen point seven percent, But when you remove open AI's 0:02:25.360 --> 0:02:28.040 twenty four point one billion dollars in revenue, which comes 0:02:28.040 --> 0:02:31.520 from open AI's compute spending revenue share, Microsoft's year over 0:02:31.639 --> 0:02:34.440 year growth was more like nine point two four percent, 0:02:34.760 --> 0:02:37.239 the lowest it would be since fiscal year twenty twenty three, 0:02:37.240 --> 0:02:39.440 when it only grew by seven percent. But there is 0:02:39.560 --> 0:02:43.720 one big difference, and that well, capital expenditures in the 0:02:43.919 --> 0:02:46.720 entire fiscal year twenty twenty three were thirty one point 0:02:46.720 --> 0:02:49.720 eight billion dollars, and Microsoft spent thirty five point eight 0:02:49.760 --> 0:02:52.240 billion dollars in the fourth quarter of the fiscal year 0:02:52.280 --> 0:02:54.440 of twenty twenty six, a year when it spent one 0:02:54.560 --> 0:02:57.160 hundred and fifteen billion goddamn dollars, and it was to 0:02:57.160 --> 0:02:59.440 spend another one hundred and seventy five billion dollars in 0:02:59.480 --> 0:03:03.000 capex in fiscal year twenty twenty seven, which begins July Firs, 0:03:03.080 --> 0:03:06.720 twenty twenty six. It's annoying, but we're moving on. Oh 0:03:06.760 --> 0:03:10.320 and open Ai, arguably Microsoft's largest customer and responsible for 0:03:10.400 --> 0:03:13.359 nearly half of its growth in the last fiscal year, 0:03:13.800 --> 0:03:15.880 is going to run out of money, and by nature 0:03:15.880 --> 0:03:18.600 of it being seventy percent of Microsoft's AI revenues, it's 0:03:18.720 --> 0:03:21.280 very obvious that there are no other large customers that 0:03:21.360 --> 0:03:24.440 exist that could possibly take up all of this infrastructure, 0:03:24.760 --> 0:03:27.799 other than, of course, weird Warrio Ama das Anthropic, who 0:03:27.840 --> 0:03:30.920 is in exactly the same position as open Ai, but 0:03:31.040 --> 0:03:34.399 with a slightly more annoying voice. Because we're near the exponential. 0:03:35.360 --> 0:03:38.080 This also means that any future revenue guidance and analyst 0:03:38.160 --> 0:03:42.120 expectations are now set under the explicit assumption that Anthropic 0:03:42.160 --> 0:03:44.160 and open Ai will both have the need for and 0:03:44.200 --> 0:03:47.600 be able to pay for increasingly larger levels of compute 0:03:47.800 --> 0:03:51.640 at increasingly higher prices. Based on the analyst notes I 0:03:51.680 --> 0:03:53.880 reported on which are in the newsletter or at least 0:03:53.960 --> 0:03:57.480 quotes ubs Wills, far Going Barclay's all have the expectation 0:03:57.560 --> 0:04:00.360 that much of Google, Amazon, and Microsoft's revenues are going 0:04:00.400 --> 0:04:02.720 to come from open AI and Anthropic for the next 0:04:02.720 --> 0:04:05.360 few years, and they don't seem to be diverging from 0:04:05.360 --> 0:04:07.920 that point in any any way. Everyone just thinks that 0:04:07.960 --> 0:04:10.360 this is where it's going to come from. Any failure 0:04:10.360 --> 0:04:13.600 for these revenues to appear out of insolvency, lack of demand, 0:04:13.680 --> 0:04:16.280 or unbuilt data center capacity meaning they can't get paid 0:04:16.480 --> 0:04:19.040 will piss off a Wall Street that's been brainwashed into 0:04:19.080 --> 0:04:22.240 believing that these three companies are growing because of diverse 0:04:22.279 --> 0:04:25.160 demand for AI services, rather than having pissed the trillion 0:04:25.200 --> 0:04:28.400 dollars up the walls mostly to just feed themselves money. 0:04:28.640 --> 0:04:31.320 I am not confident that the market will actually understand 0:04:31.320 --> 0:04:33.680 what Bloomberg has said. As I'm looking now, Microsoft is 0:04:33.760 --> 0:04:36.359 up two percent. Jesus spark in Christ, So let me 0:04:36.360 --> 0:04:39.719 spell it out for you. Seventy percent of Microsoft's AI 0:04:39.839 --> 0:04:42.920 revenues are from either its revenue share of open AI's money, 0:04:43.240 --> 0:04:46.240 or open ai spending money to rent out compute from Azure. 0:04:47.080 --> 0:04:49.800 This means that Microsoft is now dependent on open Ai 0:04:49.920 --> 0:04:53.120 for future revenue growth, as the company now represents more 0:04:53.160 --> 0:04:55.479 than seven percent of its annual revenues and there does 0:04:55.520 --> 0:04:59.040 not appear to be significant contribution from anything outside of 0:04:59.120 --> 0:05:01.840 single digit billllions of annual revenue from Microsoft three to 0:05:01.839 --> 0:05:04.880 sixty five. Copiler, a product mostly sold by tricking people 0:05:04.880 --> 0:05:07.680 into signing contracts like doctor facilie. In the beginning of 0:05:07.680 --> 0:05:11.320 Princess and the Frog, Microsoft believed it would create an independent, 0:05:11.440 --> 0:05:14.600 ultra growth engine from its investments in open Ai. Instead, 0:05:14.640 --> 0:05:18.000 it has created a parasite. At seventy percent of AI revenues. 0:05:18.279 --> 0:05:22.800 Open Ai represents the failure of saching Adella's empire Microsoft 0:05:22.800 --> 0:05:26.520 to spend two hundred and sixty billion dollars on infrastructure 0:05:26.680 --> 0:05:29.400 for demand that does not exist outside of two different 0:05:29.400 --> 0:05:31.800 companies that purn tens of billions of dollars a year. 0:05:32.080 --> 0:05:35.080 And to make matters worse, open Ai and Anthropic must 0:05:35.120 --> 0:05:38.760 grow their compute spend aggressively every single year to keep 0:05:38.839 --> 0:05:41.480 up with the expectations of analysts, which will mean it 0:05:41.520 --> 0:05:44.320 will need more and more money, which means anyone investing 0:05:44.320 --> 0:05:47.000 in Microsoft is effectively gambling on how long open Ai 0:05:47.080 --> 0:05:50.320 and Anthropic can continue to raise money before they simply 0:05:50.360 --> 0:05:54.520 exhaust every available form of capital. This also means that 0:05:54.520 --> 0:05:57.520 there's not really real demand for AI, at least not 0:05:57.600 --> 0:06:00.800 remotely commensurate with the capital expenditure or all of the 0:06:00.920 --> 0:06:04.080 endless blathering on about how much the world is changing 0:06:04.160 --> 0:06:07.640 because of AI. Microsoft has tens of thousands of Azure 0:06:07.720 --> 0:06:11.200 and software salespeople, the single most expertise in building large 0:06:11.240 --> 0:06:14.960 scale infrastructure outside of AWS, and the single most ruthless 0:06:14.960 --> 0:06:19.480 CFO outside of Oracle. If they were actual demand, actual margins, 0:06:19.520 --> 0:06:24.120 actual customers, they would sell to literally anybody else. They 0:06:24.160 --> 0:06:27.360 would sell that compute to literally anyone else other than 0:06:27.400 --> 0:06:30.800 Open Ai. They would sell it to Gargamel. Microsoft would 0:06:30.800 --> 0:06:34.760 happily help Gargamel kill every single smurf rather than talk 0:06:34.839 --> 0:06:38.720 to Sam Thucking Altman or Darier Amiday. Ever, again, if 0:06:38.720 --> 0:06:41.800 there were meaningful demand for a compute or AI software, 0:06:42.040 --> 0:06:45.480 Microsoft would be representative of it as the largest vendor 0:06:45.520 --> 0:06:48.719 in the world of both. Seventy percent of its AI 0:06:48.760 --> 0:06:51.600 revenues and seven percent of its entire annual revenues come 0:06:51.600 --> 0:06:53.679 from a company that just had to raise one hundred 0:06:53.680 --> 0:06:56.440 and twenty two billion dollars in March and then delay 0:06:56.440 --> 0:06:58.680 its IPO because its advisors didn't think it would get 0:06:58.680 --> 0:07:02.359 a trillion dollar valuation. The demand is not there for 0:07:02.440 --> 0:07:05.680 the over three hundred gigawatts of data center capacity in planning, 0:07:05.800 --> 0:07:08.640 outside of course of Open AI and Anthropic, who still 0:07:08.680 --> 0:07:10.640 to this day have yet to prove any possible way 0:07:10.640 --> 0:07:13.000 of becoming profitable and have had to raise over a 0:07:13.000 --> 0:07:15.920 combined two hundred billion dollars in the last seven months 0:07:16.000 --> 0:07:19.960 just to keep the lights on. To be clear, sam 0:07:20.000 --> 0:07:22.520 Altman has also signed up for over seven hundred and 0:07:22.560 --> 0:07:29.440 fifty billion dollars worth of commitments, which means that Google, Amazon, Microsoft, Core, Weave, Cerebrus, Iron, Nebius, Lambda, 0:07:29.680 --> 0:07:33.480 and any other associated counterparty's future revenue growth is dependent 0:07:33.520 --> 0:07:35.960 on Clammy sam Morton's ability to raise more than the 0:07:36.000 --> 0:07:38.880 combined sum of global venture capital in the last year 0:07:39.120 --> 0:07:42.280 in the next three years. On top of that, Dariama 0:07:42.320 --> 0:07:45.320 Day's Anthropic has made over three hundred billion dollars worth 0:07:45.320 --> 0:07:48.480 of commitments, on top of a thirty five billion dollar 0:07:48.520 --> 0:07:51.720 Apollo backed private credit deal where Google will sell its 0:07:51.760 --> 0:07:54.920 TPUs to an SPV that would then install them in 0:07:54.960 --> 0:07:59.480 Google data centers to for Anthropic to then rent from Google. 0:07:59.760 --> 0:08:03.080 It it's so stupid. It's an insane monument to the desperate, manic, 0:08:03.120 --> 0:08:06.920 capitalistic innovations the crown this wretched era of dog shit. 0:08:21.440 --> 0:08:23.960 The demand isn't there, and some of you might. I know, 0:08:23.960 --> 0:08:25.880 I'm going to get an email saying well, demand is 0:08:25.920 --> 0:08:29.120 real Fromanthropic and open Ai. Yeah, they're selling twenty dollars 0:08:29.200 --> 0:08:31.920 or forty dollars for a dollar. I imagine there's some demand. 0:08:31.960 --> 0:08:35.760 Everyone's screaming at you to use AI everywhere. Wow, that's 0:08:35.800 --> 0:08:38.959 demand that comes from social pressure, not actual value or 0:08:38.960 --> 0:08:43.800 the utility. And really that demand isn't there. If Microsoft 0:08:43.840 --> 0:08:47.960 isn't seeing it, nobody is seeing it. The only reason 0:08:48.360 --> 0:08:51.640 that any neocloud is still alive is in vidious circular 0:08:51.679 --> 0:08:56.040 financing rituals, a failure of ratings agencies, Open AI and 0:08:56.120 --> 0:08:58.600 Anthropic and whatever weird oh bullshit it is that Mark 0:08:58.679 --> 0:09:02.000 Zuckerberg is up to. There's a reason that the first 0:09:02.000 --> 0:09:05.600 and only story about a customer for metas compute capacity 0:09:05.760 --> 0:09:09.360 was Anthropic. There's nobody else, nobody else other than Anthropic 0:09:09.400 --> 0:09:12.560 of course, open ai or the hyperscalers that are siphoning 0:09:12.559 --> 0:09:15.439 off what remains of the AI labs venture capital funding 0:09:15.679 --> 0:09:18.480 to sustain their own growth rates, every time, raising a 0:09:18.480 --> 0:09:20.199 little bit more and a little bit higher at a 0:09:20.280 --> 0:09:24.120 higher valuation, each time, demanding more and more money, hoping 0:09:24.160 --> 0:09:26.080 to be able to dump it onto the public markets, 0:09:26.080 --> 0:09:28.160 and the hope that they can perpetually con investors into 0:09:28.240 --> 0:09:31.520 backing their deals. The reason that Microsoft and Google and 0:09:31.559 --> 0:09:34.680 Amazon have been so squeerrely about their AI revenues is 0:09:34.679 --> 0:09:37.360 that they know that they're pathetic compared to their capex 0:09:37.400 --> 0:09:40.920 and near entirely dependent on two companies that Hyperscalers must 0:09:41.000 --> 0:09:43.719 keep alive so that they can pay them. They can 0:09:43.840 --> 0:09:46.160 get paid by them with the money that the hyperscaler 0:09:46.200 --> 0:09:50.000 is handed to them. It's not good. And in twenty 0:09:50.040 --> 0:09:52.920 twenty six alone, Google and Amazon have invested a combined 0:09:52.960 --> 0:09:56.560 fifteen billion dollars in Anthropic and Amazon's invested to combined 0:09:56.640 --> 0:09:59.080 fifty billion dollars in open Ai. And what's crazy about 0:09:59.080 --> 0:10:01.200 that was thirty five billion dollars of that was contingent 0:10:01.200 --> 0:10:04.240 on the IPO or AGI. Neither of those happened, which 0:10:04.360 --> 0:10:07.160 heavily suggests that Amazon kind of knew that they needed 0:10:07.160 --> 0:10:10.440 to pay open Ai so that open Ai could pay them. 0:10:10.679 --> 0:10:12.319 And by the way, this is what you do when 0:10:12.360 --> 0:10:14.720 you have no other options, no other viable customers, and 0:10:14.760 --> 0:10:17.240 your core businesses are not growing at the rate necessary 0:10:17.240 --> 0:10:20.120 to appease wall Stream. It's been sitting in plain sight 0:10:20.200 --> 0:10:25.080 for years. Hyperscalers overbuilt capacity, funded their largest customers, incubating 0:10:25.120 --> 0:10:27.640 them actually, and believed that both the opportunity of owning 0:10:27.679 --> 0:10:30.600 their models and the underlying compute capacity would be some 0:10:30.640 --> 0:10:33.520 sort of infinite money glitch, with each hyperscaler owning the 0:10:33.559 --> 0:10:36.600 equivalent of both the next big cloud infrastructure and cloud 0:10:36.640 --> 0:10:40.880 software plays, while also owning controlling stakes and creating dependency 0:10:41.120 --> 0:10:43.760 from the two largest players in both Pretty easy, right, 0:10:44.360 --> 0:10:47.720 anyone else building AI data centers believe that Microsoft, Google, 0:10:47.720 --> 0:10:50.040 and Amazon would never be as foolish as to do 0:10:50.120 --> 0:10:53.040 all of this without a cast iron guarantee that it 0:10:53.080 --> 0:10:56.280 would turn into everything they'd promised, and that the capital 0:10:56.280 --> 0:10:59.080 expenditures would be for certain worth it and have the 0:10:59.120 --> 0:11:01.959 demand to back them that they would not continue spending 0:11:02.000 --> 0:11:05.839 without that certainty. The reality is that Hyperscalers made a 0:11:05.920 --> 0:11:08.760 huge mistake, both in what AI could do, how fast 0:11:08.840 --> 0:11:11.800 Open Ai and Anthropic could grow, and in underestimating how 0:11:11.840 --> 0:11:15.040 bad these businesses would get as they grew. It was 0:11:15.080 --> 0:11:19.280 also obvious in retrospect from day one. Hyperscalers and most 0:11:19.280 --> 0:11:22.320 companies in fact, refused to share their AI revenues, and 0:11:22.360 --> 0:11:25.360 when they did so, they used deceptive run rates based 0:11:25.360 --> 0:11:29.280 on non specific periods. I've said this for years. If 0:11:29.280 --> 0:11:32.280 these companies had diverse and meaningful AI revenues, they would 0:11:32.320 --> 0:11:35.440 be screaming about them from the rooftops. With clarity and vigor, 0:11:35.880 --> 0:11:39.079 they would tell you segment by segment AI specific contribution 0:11:39.160 --> 0:11:41.720 to revenue, they'd break out the costs. They would have 0:11:41.840 --> 0:11:46.000 them in big font on investor presentations. Instead, Google is 0:11:46.040 --> 0:11:50.240 telling us it processed three point two quadrillion tokens a month. 0:11:50.720 --> 0:11:53.760 Who gives a fuck, Sunda, You're planning to spend one 0:11:53.880 --> 0:11:56.480 hundred and eighty billion dollars this year, and we both 0:11:56.520 --> 0:11:58.800 know that if you had the revenues that would remotely 0:11:58.920 --> 0:12:01.800 justify spending that much, you'd include them. Did you make 0:12:01.800 --> 0:12:04.000 a big mistake? Son? Dar? You can tell me, buddy. 0:12:04.040 --> 0:12:06.400 I won't be mad at you. Investors will be, though, 0:12:06.480 --> 0:12:09.720 They're gonna be furious, but look at the bright side. 0:12:10.160 --> 0:12:12.080 At least you won't have to hear from that boring 0:12:12.120 --> 0:12:15.000 of Demis Hassabis and hearing him prattle on and on 0:12:15.160 --> 0:12:17.480 a bala. Oh what if the computer had the most 0:12:17.480 --> 0:12:20.199 brilliant mind? What if we had agi in twenty one 0:12:20.240 --> 0:12:22.440 minutes now he's going to go off and I assume 0:12:22.520 --> 0:12:26.600 spend more time with his money. Anyway, one day everybody 0:12:26.640 --> 0:12:29.000 is going to whack super surprised and say there was 0:12:29.000 --> 0:12:32.040 no way to see this coming. Know that the signs 0:12:32.080 --> 0:12:34.720 were there for years. They've stayed the same. The only 0:12:34.760 --> 0:12:37.280 thing that's changed is the inflated value of open AI 0:12:37.360 --> 0:12:41.080 and Anthropic and the ever escalating capex, fueling a semiconductor 0:12:41.120 --> 0:12:44.640 bubble that erupted because nobody thought Hyperscalers would ever waste 0:12:44.720 --> 0:12:48.120 one trillion dollars. At some point, the money runs out 0:12:48.120 --> 0:12:51.199 to fund these two companies, at which point Hyperscalers will 0:12:51.200 --> 0:12:54.520 have no way to further feed themselves in money. As 0:12:54.520 --> 0:12:58.000 I've said before, the AI bubble is a syop, a melodrama, 0:12:58.120 --> 0:13:01.160 a financial crisis, and a mask moment for the business 0:13:01.160 --> 0:13:04.400 idiots that run the vast majority of the economy. It 0:13:04.520 --> 0:13:07.160 is one of the strangest moments in history, and it 0:13:07.200 --> 0:13:11.000 will only get stranger as things get more desperate. We 0:13:11.040 --> 0:13:13.920 are seeing what happens when the rot economy demands that 0:13:13.960 --> 0:13:16.480 we push the world's resources to their very limits, both 0:13:16.480 --> 0:13:19.600 physically and fiscally, in pursuit of eternal growth, and what 0:13:19.760 --> 0:13:23.960 happens when we discover what those limits actually are. None 0:13:23.960 --> 0:13:26.599 of this was ever about technology, or good business or 0:13:26.640 --> 0:13:30.520 anything other than creating the next growth engine for capital. Instead, 0:13:30.640 --> 0:13:34.599 AI has become a parasitic presence culturally, financially, and ecologically, 0:13:34.880 --> 0:13:36.800 and it will exhaust as much as it can in 0:13:36.840 --> 0:13:41.040 its collapse. Nothing about this can be bailed out, because 0:13:41.040 --> 0:13:43.679 the problem isn't that the money doesn't exist, just that 0:13:43.720 --> 0:13:47.640 it doesn't exist naturally. A bailout cannot make growth restart 0:13:47.800 --> 0:13:51.400 for Google, Microsoft, Meta, Amazon, or Oracle once it fully slows, 0:13:51.800 --> 0:13:53.480 and based on the fact they've sunk more than a 0:13:53.480 --> 0:13:56.319 trillion dollars of CAPEX into AI, I think they must 0:13:56.360 --> 0:13:59.040 clearly see that growth is already or will die very 0:13:59.080 --> 0:14:02.559 soon without an id quote doing AI, which mostly means 0:14:02.600 --> 0:14:05.640 hoping anthropic and open AI pay them ever increasing amounts 0:14:05.679 --> 0:14:09.680 of money every quarter forever. The US government cannot prop 0:14:09.720 --> 0:14:12.440 up a stock market that accepts that AI's growth is 0:14:12.480 --> 0:14:16.000 finite and outside of semiconductor companies and two subsidized AI 0:14:16.080 --> 0:14:18.959 labs is tens of billions of dollars of demand in 0:14:19.000 --> 0:14:22.840 the best case scenario. I realize how easy it is 0:14:22.880 --> 0:14:25.360 to fall into the mindset that everything gets bailed out, 0:14:25.360 --> 0:14:28.120 that the evil people always win, because we're so used 0:14:28.120 --> 0:14:32.120 to seeing these evil, corrupt people escape accountability. In this case, 0:14:32.160 --> 0:14:34.800 while I don't think anyone's going to prison. I think 0:14:34.800 --> 0:14:37.560 these companies simply exhaust their resources at some point and 0:14:37.680 --> 0:14:41.280 drop off. I think the era of hyperscale growth is ending, 0:14:41.320 --> 0:14:43.800 and the only reason they're so desperately joining the no 0:14:43.920 --> 0:14:47.720 IT loads, refused AI compute come dump is because they 0:14:47.760 --> 0:14:50.120 know that the only way to keep growth going is 0:14:50.120 --> 0:14:52.640 to hope they can drain the venture capital and debt 0:14:52.640 --> 0:14:55.840 markets infinitely by proxy. I don't know if people are 0:14:55.840 --> 0:14:58.000 gonna like that. They just said that I'll keep going 0:14:58.040 --> 0:15:02.560 though anyway. There are no other hypergrowth markets left. That's 0:15:02.640 --> 0:15:06.080 all this is about. So many numbers are going up 0:15:06.360 --> 0:15:08.880 so much higher in the next few months. Then at 0:15:08.880 --> 0:15:11.240 some point they'll stop going higher, and then they'll start 0:15:11.240 --> 0:15:12.760 going lower, and then some of them will go to 0:15:12.840 --> 0:15:15.760 zero or even become negative numbers because the only thing 0:15:15.800 --> 0:15:19.760 that grows forever is cancer. This all may seem unthinkable 0:15:19.840 --> 0:15:23.040 right now because Microsoft, Google, and Amazon numbers are going up, 0:15:23.440 --> 0:15:25.560 but that's really only happening because of open AI and 0:15:25.600 --> 0:15:30.040 anthropic Most didn't see this coming because it has involved 0:15:30.080 --> 0:15:33.160 pushing back against one of the most gratuitous and excessive 0:15:33.200 --> 0:15:37.040 marketing campaigns in history, involving actual reporters that are at 0:15:37.080 --> 0:15:41.440 real outlets attacking and harassing people for daring to question 0:15:41.520 --> 0:15:44.680 the powerful, and those people will be held accountable, mark 0:15:44.720 --> 0:15:47.840 my words. But let's talk about who actually matters, and 0:15:47.840 --> 0:15:50.320 that's you, and a lot of you have been here 0:15:50.440 --> 0:15:53.680 with me for years, hearing me build these arguments, and 0:15:53.720 --> 0:15:55.880 I appreciate the amount of support I hear from you 0:15:56.000 --> 0:15:59.440 via a signal, the terminal, and on the subreddit an email. 0:15:59.480 --> 0:16:02.440 Of course, you're all awesome, and you keep me going 0:16:02.480 --> 0:16:04.920 when things are hard, when people are pressing back, when 0:16:04.960 --> 0:16:07.840 people are genuinely horrible to me. I don't want to complain. 0:16:07.840 --> 0:16:09.400 I get to do the best job in the world. 0:16:09.400 --> 0:16:11.640 It fucking rocks to record this podcast and write the 0:16:11.640 --> 0:16:14.480 news letter. It's the greatest job I've ever done. I 0:16:14.480 --> 0:16:17.160 couldn't be happy at doing it. It's stressful, but I 0:16:17.240 --> 0:16:19.080 keep going because of all of you, because I really 0:16:19.120 --> 0:16:21.680 truly enjoy it. I'll be back next week, and to 0:16:21.800 --> 0:16:24.320 quote Drill, I'll be calling out all the bullshit of society, 0:16:24.360 --> 0:16:26.640 the smart way and against all odds. I'll do it 0:16:26.680 --> 0:16:28.600 for free. Thank you for listening.