1 00:00:02,400 --> 00:00:07,320 Speaker 1: Zone Media. Hello on welcomes a better offline. I'm at Zeitron, 2 00:00:07,360 --> 00:00:21,840 Speaker 1: and I'm nothing like Enron. Now. This is the second 3 00:00:21,880 --> 00:00:24,479 Speaker 1: episode in the Nvidia Enron series. And to give you 4 00:00:24,520 --> 00:00:27,000 Speaker 1: a recap, a few weeks ago a document from Nvidia 5 00:00:27,120 --> 00:00:29,319 Speaker 1: leaked where the company denied that it was anything like 6 00:00:29,480 --> 00:00:34,400 Speaker 1: Enron at all. But what was Enron? Now? Some of 7 00:00:34,440 --> 00:00:36,600 Speaker 1: you might know this, some of you might not, but 8 00:00:36,680 --> 00:00:39,239 Speaker 1: I think it's time to remind everyone what Enron was 9 00:00:39,280 --> 00:00:43,040 Speaker 1: and what Enron did and why Enron was So I 10 00:00:43,080 --> 00:00:46,680 Speaker 1: don't know if we should say special, but it sure 11 00:00:46,800 --> 00:00:50,040 Speaker 1: was Enron. Now, the collapse of Enron wasn't just in retrospect, 12 00:00:50,040 --> 00:00:53,199 Speaker 1: a large business that ultimately failed. If that's all it was, 13 00:00:53,240 --> 00:00:55,080 Speaker 1: it wouldn't come on the same space in our heads 14 00:00:55,120 --> 00:00:58,240 Speaker 1: as other failures like WorldCom or Nortel, which I'll actually 15 00:00:58,240 --> 00:01:01,279 Speaker 1: get to later talked about WorldCom line episode, both of 16 00:01:01,320 --> 00:01:04,000 Speaker 1: whom were similarly considered giants in their fields. By the way, 17 00:01:04,800 --> 00:01:07,200 Speaker 1: it's also not just about the fact that Enron failed 18 00:01:07,200 --> 00:01:10,480 Speaker 1: because had proven business and accounting fraud. World comm entered 19 00:01:10,480 --> 00:01:14,039 Speaker 1: bankruptcy due to similar circumstances, though, rather than being liquidated, 20 00:01:14,319 --> 00:01:16,200 Speaker 1: and this is kind of a mind fuck, so forgive me. 21 00:01:16,480 --> 00:01:19,360 Speaker 1: It was the quietest part of Verizon's acquisition of MCI, 22 00:01:19,560 --> 00:01:22,400 Speaker 1: the name of a company that had previously merged with WorldCom, 23 00:01:22,480 --> 00:01:26,840 Speaker 1: that WorldCom renamed itself to after bankruptcy. This era sucked, 24 00:01:27,760 --> 00:01:30,880 Speaker 1: and unlike Enron, WorldCom isn't the subject of multiple films 25 00:01:30,880 --> 00:01:34,120 Speaker 1: and even a Broadway production. And my editor saw the 26 00:01:34,240 --> 00:01:36,520 Speaker 1: UK tour of that and apparently it was pretty good, 27 00:01:36,640 --> 00:01:38,360 Speaker 1: you asked Matt Hughes, if you want to know more. 28 00:01:38,760 --> 00:01:41,039 Speaker 1: And it's also not the size of Enron that made 29 00:01:41,080 --> 00:01:43,959 Speaker 1: its downfall so intriguing, nor, for that matter, is it 30 00:01:44,000 --> 00:01:45,880 Speaker 1: the fact that Enron did a lot of legally and 31 00:01:45,880 --> 00:01:49,320 Speaker 1: ethically dubious stuff to bring about its downfall. No, what 32 00:01:49,360 --> 00:01:52,560 Speaker 1: makes Enron special is the sheer gravity of its fuckery, 33 00:01:52,800 --> 00:01:54,680 Speaker 1: the rotten culture at the heart of the company that 34 00:01:54,800 --> 00:01:58,280 Speaker 1: encouraged said malfeasance, and the creative ways Enron's leaders crafted 35 00:01:58,320 --> 00:02:00,960 Speaker 1: an image of success around what was at its heart 36 00:02:01,160 --> 00:02:04,440 Speaker 1: a big, fat dog of a company. Enron was born 37 00:02:04,480 --> 00:02:07,080 Speaker 1: in nineteen eighty five out on the foundations of two older, 38 00:02:07,200 --> 00:02:11,440 Speaker 1: much less interesting businesses. The first, Houston Natural Gas HG, 39 00:02:11,960 --> 00:02:15,120 Speaker 1: started life as a utility provider bumping natural gas from 40 00:02:15,160 --> 00:02:17,800 Speaker 1: the oil fields of Texas to customers throughout the region, 41 00:02:18,040 --> 00:02:21,239 Speaker 1: before later exiting the industry to focus on other opportunities. 42 00:02:21,600 --> 00:02:24,240 Speaker 1: The other Inter North, was based in Omaha, Nebraska, and 43 00:02:24,320 --> 00:02:28,120 Speaker 1: was in the same business pipelines. In the mid nineteen eighties, 44 00:02:28,440 --> 00:02:30,799 Speaker 1: H ANDNG was the subject of a hostile takeoff from 45 00:02:30,880 --> 00:02:34,440 Speaker 1: Coastal Corporation, which until two thousand and one operated a 46 00:02:34,520 --> 00:02:36,800 Speaker 1: chain of refineries and gas stations throughout much of the 47 00:02:36,880 --> 00:02:40,560 Speaker 1: US mainland. Unable to fend it off by itself, h 48 00:02:40,639 --> 00:02:44,880 Speaker 1: ANDNG merged with inter North, with the combined corporation renamed Enron. 49 00:02:45,120 --> 00:02:47,519 Speaker 1: The CEO of this new entity was called Ken Lay, 50 00:02:47,760 --> 00:02:50,040 Speaker 1: an economist by trade who spent most of his career 51 00:02:50,040 --> 00:02:53,240 Speaker 1: in the energy sector, who also enjoyed deep political connections 52 00:02:53,240 --> 00:02:55,880 Speaker 1: with the Bush family. He co chaired George H. W. 53 00:02:56,080 --> 00:02:59,280 Speaker 1: Bush's failed nineteen ninety two re election campaign and allowed 54 00:02:59,360 --> 00:03:02,080 Speaker 1: Enron's corporate jet to Ferry Bush Senior and Barbara Bush 55 00:03:02,120 --> 00:03:05,600 Speaker 1: back and forth. To Washington Center for Public Integrity director 56 00:03:05,639 --> 00:03:08,280 Speaker 1: Charles Lewis said that there was and I quote no 57 00:03:08,440 --> 00:03:11,800 Speaker 1: company in America closer to George W. Bush than Enron. 58 00:03:12,960 --> 00:03:15,600 Speaker 1: George W. Bush, and I mean the second one even 59 00:03:15,639 --> 00:03:19,400 Speaker 1: had a nickname for Lay Kenny Boy. Anyway, in nineteen 60 00:03:19,400 --> 00:03:22,840 Speaker 1: eighty seven, Enron hired McKinsey, the world's most evil management 61 00:03:22,919 --> 00:03:25,880 Speaker 1: consultancy firm, to help the company create a futures market 62 00:03:25,880 --> 00:03:28,799 Speaker 1: for natural gas. Well that means isn't particularly important to 63 00:03:28,840 --> 00:03:31,160 Speaker 1: the story, but essentially, a futures contract is where a 64 00:03:31,200 --> 00:03:33,480 Speaker 1: company agrees to buy or sell an assair in the 65 00:03:33,520 --> 00:03:36,560 Speaker 1: future at a fixed price. It's a way of hedging 66 00:03:36,640 --> 00:03:39,360 Speaker 1: against risk, whether that be from something like price or 67 00:03:39,400 --> 00:03:43,360 Speaker 1: currency fluctuations, or from default. If you're buying oil in dollars, 68 00:03:43,400 --> 00:03:46,000 Speaker 1: for example, buying a futures contract for oil to be 69 00:03:46,040 --> 00:03:48,840 Speaker 1: delivered in six months time at a predetermined price means 70 00:03:48,880 --> 00:03:51,040 Speaker 1: that if your currency weakens against the dollar, your costs 71 00:03:51,080 --> 00:03:54,920 Speaker 1: won't spiral. That bit isn't terribly important. What does matter 72 00:03:54,960 --> 00:03:58,360 Speaker 1: is while working with McKinsey, Lay met someone named Jess Skilling, 73 00:03:58,480 --> 00:04:01,840 Speaker 1: a young engineer turned consultant who impressed the company's CEO deeply, 74 00:04:02,240 --> 00:04:04,320 Speaker 1: so much so that Ley decided to poach him from 75 00:04:04,360 --> 00:04:06,600 Speaker 1: McKinsey in nineteen ninety and give him the role of 76 00:04:06,680 --> 00:04:10,880 Speaker 1: chairman and CEO of Enron. Finance Group. Oh. By the way, 77 00:04:11,000 --> 00:04:13,520 Speaker 1: Enron had a bunch of subsidiaries and some had their 78 00:04:13,520 --> 00:04:15,880 Speaker 1: own CEOs and boards. I mentioned this because he may 79 00:04:15,920 --> 00:04:18,120 Speaker 1: be a bit confused, as Lay was CEO of Nron 80 00:04:18,160 --> 00:04:22,039 Speaker 1: and Skilling was CEO of Nron Finance Group. In essence, 81 00:04:22,080 --> 00:04:24,159 Speaker 1: it's a bit like how Sam Altman is CEO of 82 00:04:24,200 --> 00:04:27,599 Speaker 1: open Ai and Fijisimo as the CEO of Applications. That 83 00:04:27,680 --> 00:04:29,520 Speaker 1: bit isn't important, but I just want to be as 84 00:04:29,560 --> 00:04:32,760 Speaker 1: explicit as possible. I don't think open Ai is Enron either, 85 00:04:33,160 --> 00:04:36,560 Speaker 1: at least I hope. Anyway, Skilling continued to impress Lay, 86 00:04:36,560 --> 00:04:40,320 Speaker 1: who gave him greater and greater responsibility, eventually crowning him 87 00:04:40,480 --> 00:04:44,200 Speaker 1: chief operating Officer of Enron itself. With Skilling in a 88 00:04:44,279 --> 00:04:47,600 Speaker 1: key leadership position who was able to shape the organization's culture, 89 00:04:47,880 --> 00:04:50,840 Speaker 1: He appreciated those who took risks, even if those risks, 90 00:04:50,880 --> 00:04:54,440 Speaker 1: when viewed with impartialized were deemed reckless or even criminal. 91 00:04:54,880 --> 00:04:57,359 Speaker 1: He introduced the practice of stack ranking, also known as 92 00:04:57,480 --> 00:04:59,960 Speaker 1: rank and yank, to Enron, which had previously been pioneer 93 00:05:00,200 --> 00:05:03,240 Speaker 1: by Jack Welcher General Electric. And you should listen to 94 00:05:03,279 --> 00:05:05,760 Speaker 1: the shareholder supremacy from last year. If you want to 95 00:05:05,800 --> 00:05:10,120 Speaker 1: hear what I think about that redacted guy here. Employees 96 00:05:10,120 --> 00:05:12,080 Speaker 1: were graded on a scale, and those at the bottom 97 00:05:12,120 --> 00:05:14,400 Speaker 1: of the scale were terminated. Managers had to place at 98 00:05:14,440 --> 00:05:17,119 Speaker 1: least ten percent. Other reports stay closer to fifteen percent 99 00:05:17,120 --> 00:05:19,760 Speaker 1: of employees in the lowest bracket, which created an almost 100 00:05:19,800 --> 00:05:23,320 Speaker 1: Darwinian culture and a drive to succeed that matched him. 101 00:05:23,800 --> 00:05:26,800 Speaker 1: Staffers were brutal hours, they cut corners, They did some 102 00:05:26,920 --> 00:05:30,760 Speaker 1: really really dodgy shit. None of this bothered skilling in 103 00:05:30,800 --> 00:05:34,120 Speaker 1: the slightest. Now you might be asking how dodgy ed well. 104 00:05:34,160 --> 00:05:36,480 Speaker 1: In two thousand and two thousand and one, California suffered 105 00:05:36,480 --> 00:05:39,839 Speaker 1: a series of electricity blackouts. This shouldn't have happened. California's 106 00:05:39,880 --> 00:05:42,080 Speaker 1: total energy demand at the time was twenty eight gigawats 107 00:05:42,080 --> 00:05:45,679 Speaker 1: and its production capacity was forty five gigawats. California also 108 00:05:45,680 --> 00:05:48,080 Speaker 1: shares a transmission grid with other states and for what 109 00:05:48,120 --> 00:05:51,279 Speaker 1: it's worth, the Canadian provinces of Alberta and British Columbia, 110 00:05:51,360 --> 00:05:54,400 Speaker 1: as well as part of Baja California in Mexico, meaning 111 00:05:54,400 --> 00:05:56,640 Speaker 1: that in the event of a shortage, it could simply 112 00:05:56,720 --> 00:05:59,960 Speaker 1: draw capacity from elsewhere. There's plenty to go around. Did 113 00:06:00,120 --> 00:06:04,640 Speaker 1: that happen? Well? Remember, Enron traded electricity like a commodity, 114 00:06:04,680 --> 00:06:06,520 Speaker 1: and as a result, it was incentivized to get the 115 00:06:06,560 --> 00:06:09,479 Speaker 1: highest possible price for that commodity, so it took power 116 00:06:09,480 --> 00:06:12,560 Speaker 1: plants offline during peak hours and exported power to other 117 00:06:12,600 --> 00:06:16,400 Speaker 1: states where there was real domestic demand. But how does 118 00:06:16,440 --> 00:06:20,680 Speaker 1: a company like Enron shut down a power station? Simple, 119 00:06:20,760 --> 00:06:24,599 Speaker 1: it just asked. In one taped phone conversation released after 120 00:06:24,640 --> 00:06:27,680 Speaker 1: the company's collapse, an Enron employee called Bill called an 121 00:06:27,680 --> 00:06:30,520 Speaker 1: official at a Las Vegas power company California shares the 122 00:06:30,560 --> 00:06:33,240 Speaker 1: same grip with Nevada, and asked him to and I quote, 123 00:06:33,400 --> 00:06:35,479 Speaker 1: get a little creative and come up with a reason 124 00:06:35,520 --> 00:06:38,360 Speaker 1: to go down. Anything you want to do over there, 125 00:06:38,520 --> 00:06:43,159 Speaker 1: any cleaning, anything like that. Very cool. The power crisis 126 00:06:43,160 --> 00:06:46,280 Speaker 1: had dramatic consequences for the people of California, who faced 127 00:06:46,320 --> 00:06:49,240 Speaker 1: power outages and price hikes, for Governor Gray Davis, who 128 00:06:49,240 --> 00:06:52,240 Speaker 1: was recalled by voters and later replaced by Arnold Schwarzenegger, 129 00:06:52,520 --> 00:06:55,719 Speaker 1: for PG and E, which entered Chapter eleven bankruptcy that year, 130 00:06:55,839 --> 00:06:58,800 Speaker 1: and for Southern California Edison, which was pushed at the 131 00:06:58,800 --> 00:07:02,440 Speaker 1: brink of bankkruptcies result. This kind of stuff could only 132 00:07:02,440 --> 00:07:05,560 Speaker 1: happen in an organization whose culture actively rewarded being a 133 00:07:05,600 --> 00:07:09,880 Speaker 1: fucking asshole. In fact, Skilling was seemingly determined to elevate 134 00:07:09,880 --> 00:07:12,880 Speaker 1: the dodgiest of characters to the highest positions within the company, 135 00:07:13,080 --> 00:07:16,400 Speaker 1: and few more more ethically dubious than Andy Fastau, who's 136 00:07:16,400 --> 00:07:19,240 Speaker 1: Skilling mentored like a protege, and who would later become 137 00:07:19,440 --> 00:07:23,560 Speaker 1: Enron's chief financial officer. And you gotta wonder how did 138 00:07:23,640 --> 00:07:28,360 Speaker 1: the CFO hired by Jeff Skilling do well? Even before 139 00:07:28,400 --> 00:07:31,000 Speaker 1: vaulting to the top of Enron's nasty little empire, fast 140 00:07:31,040 --> 00:07:33,600 Speaker 1: I was able to shape its accounting practices, with the 141 00:07:33,600 --> 00:07:37,960 Speaker 1: company adopting mark to market accounting practices in nineteen ninety one. Now, 142 00:07:38,000 --> 00:07:41,080 Speaker 1: mark to market sounds complicated, but it's actually real simple. 143 00:07:41,360 --> 00:07:43,640 Speaker 1: When listing assets on a balance sheet, you don't use 144 00:07:43,680 --> 00:07:46,400 Speaker 1: the acquisition cost like how you paid for him, but 145 00:07:46,520 --> 00:07:49,880 Speaker 1: rather the fair market value of an asset. So if 146 00:07:49,920 --> 00:07:52,000 Speaker 1: I buy a baseball card for a dollar and I 147 00:07:52,040 --> 00:07:54,280 Speaker 1: see that it's currently selling for ten bucks on eBay, 148 00:07:54,400 --> 00:07:56,240 Speaker 1: I'd say that asset is worth ten bucks, not the 149 00:07:56,240 --> 00:07:58,239 Speaker 1: dollar I paid for him, even though I haven't actually 150 00:07:58,320 --> 00:08:01,800 Speaker 1: sold it yet, nor do I not if anyone's interested. Now, 151 00:08:01,800 --> 00:08:05,280 Speaker 1: this sounds simple, maybe even reasonable, But the problem is 152 00:08:05,280 --> 00:08:07,920 Speaker 1: that the way you determine the value of an asset matters, 153 00:08:07,960 --> 00:08:10,920 Speaker 1: and marked to market accounting allows companies and individuals to 154 00:08:11,000 --> 00:08:16,400 Speaker 1: exercise some what's the word creativity. Sure, for publicly traded companies, 155 00:08:16,440 --> 00:08:19,040 Speaker 1: where the price of a share is verifiable and open knowledge, 156 00:08:19,040 --> 00:08:21,800 Speaker 1: it's not too bad. But for assets with limited liquidity, 157 00:08:22,000 --> 00:08:25,280 Speaker 1: limited buyers, or where the process to be engineered somehow, 158 00:08:25,520 --> 00:08:28,640 Speaker 1: you have a lot of latitude for fraud. Now, let's 159 00:08:28,640 --> 00:08:31,400 Speaker 1: go back to that baseball card example. How do you 160 00:08:31,400 --> 00:08:33,400 Speaker 1: know it's actually worth ten bucks and not one dollar? 161 00:08:33,679 --> 00:08:37,120 Speaker 1: What is the fair value of something you can't check 162 00:08:37,120 --> 00:08:40,720 Speaker 1: on any babe, what somebody told me in person it's worth. 163 00:08:40,920 --> 00:08:42,800 Speaker 1: What's to stop me from lying in saying that the 164 00:08:42,840 --> 00:08:45,839 Speaker 1: card is worth a one hundred dollars or one thousand dollars, Well, 165 00:08:46,880 --> 00:08:49,800 Speaker 1: other than the fact that I'd be committing fraud. Now, 166 00:08:49,840 --> 00:08:52,640 Speaker 1: what if I have ten one dollar baseball cards and 167 00:08:52,880 --> 00:08:55,080 Speaker 1: I give my friend ten dollars, and I tell him 168 00:08:55,160 --> 00:08:57,640 Speaker 1: by one of my baseball cards using that ten dollar 169 00:08:57,640 --> 00:09:00,640 Speaker 1: bill that I just handed him, and allows me to 170 00:09:00,640 --> 00:09:03,120 Speaker 1: say that I've now realized a nine dollar profit on 171 00:09:03,120 --> 00:09:05,240 Speaker 1: one of my one dollar baseball cards, and my other 172 00:09:05,320 --> 00:09:08,400 Speaker 1: cards are now because I just did that worth ninety 173 00:09:08,440 --> 00:09:11,439 Speaker 1: dollars and not nine dollars. Now, what if I use 174 00:09:11,520 --> 00:09:14,880 Speaker 1: that valuation, that phone evaluation of my remaining cards to 175 00:09:14,880 --> 00:09:17,679 Speaker 1: get a fifty dollar loan using those cards as collateral, 176 00:09:18,160 --> 00:09:21,199 Speaker 1: even though the collateral itself isn't worth even one fifth 177 00:09:21,240 --> 00:09:37,400 Speaker 1: of the value of the loan. You get the idea. Well, 178 00:09:37,440 --> 00:09:39,520 Speaker 1: A lot of things people can do to alter the 179 00:09:39,520 --> 00:09:42,000 Speaker 1: marked to market value of an asset ah illegal and 180 00:09:42,040 --> 00:09:44,959 Speaker 1: would be covered under generic fraud laws. It doesn't change 181 00:09:45,000 --> 00:09:47,240 Speaker 1: the fact that marked to market accounting allows for some 182 00:09:47,280 --> 00:09:51,400 Speaker 1: shenanigans to take place. Another trait of marked to market accounting, 183 00:09:51,400 --> 00:09:53,719 Speaker 1: as employed by Enron is that it would count all 184 00:09:53,760 --> 00:09:56,920 Speaker 1: the long term potential revenue for a deal as quarterly revenue, 185 00:09:57,000 --> 00:09:58,960 Speaker 1: even if that revenue would be delivered over the course 186 00:09:58,960 --> 00:10:02,240 Speaker 1: of a decades long contra or if the contract would 187 00:10:02,280 --> 00:10:06,440 Speaker 1: be terminated before its intended expiration DAME. It would also 188 00:10:06,600 --> 00:10:09,880 Speaker 1: realize potential revenue as actual revenue even before money changed 189 00:10:09,880 --> 00:10:13,559 Speaker 1: hands and when the conclusion of the deal wasn't a certainty. 190 00:10:13,600 --> 00:10:16,080 Speaker 1: For example, in nineteen ninety nine, Enron sold a stake 191 00:10:16,120 --> 00:10:20,079 Speaker 1: in four different electricity generating barges in Nigeria. They're basically 192 00:10:20,080 --> 00:10:23,520 Speaker 1: floating power stations, sold and sold it to Merrill Lynch, 193 00:10:23,559 --> 00:10:26,320 Speaker 1: which allowed the company to register twelve million dollars in profit. 194 00:10:27,120 --> 00:10:30,280 Speaker 1: One little problem that sale didn't actually happen, though that 195 00:10:30,360 --> 00:10:33,360 Speaker 1: didn't stop Enron from selling pieces to Meyal Lynch, which 196 00:10:33,400 --> 00:10:36,600 Speaker 1: I'm not kidding. Merril Lynch then sold to a special 197 00:10:36,600 --> 00:10:41,160 Speaker 1: purpose vehicle called LJM two. You'll never guess who owned that. 198 00:10:41,520 --> 00:10:44,720 Speaker 1: It was controlled by Andrew Fastau. Yep, remember that guy. 199 00:10:44,720 --> 00:10:47,800 Speaker 1: You're gonna hear that name again. Although the Meryl Lynch 200 00:10:47,840 --> 00:10:50,960 Speaker 1: bankers who participated in the deal were eventually convicted of 201 00:10:51,000 --> 00:10:54,360 Speaker 1: fraud and conspiracy charges long after the collapse of Enron. 202 00:10:54,400 --> 00:10:57,680 Speaker 1: Of course, the convictions were later questioned on appeal, Thank 203 00:10:57,760 --> 00:11:00,600 Speaker 1: you government, but still, for a moment, gave a jolt 204 00:11:00,640 --> 00:11:04,800 Speaker 1: to Enron's quarterly earnings. Anyway, Enron was incredibly creative when 205 00:11:04,840 --> 00:11:07,440 Speaker 1: it came to how it valued assets. Take, for example, 206 00:11:07,600 --> 00:11:11,280 Speaker 1: fiber optic cables. As the dot com bubble swelled, Enron 207 00:11:11,320 --> 00:11:13,600 Speaker 1: saw an opportunity and wanted to be able to trade 208 00:11:13,640 --> 00:11:15,960 Speaker 1: and control the supply of bandwidth, just like it did 209 00:11:15,960 --> 00:11:19,840 Speaker 1: with other more conventional commodities like oil and gas. It built, bought, 210 00:11:19,920 --> 00:11:22,520 Speaker 1: and leased fiber optic cables throughout the country, and then, 211 00:11:22,840 --> 00:11:26,280 Speaker 1: using exaggerated estimates of their value and potential long term revenue, 212 00:11:26,480 --> 00:11:29,280 Speaker 1: released glowing financial reports that made the company look a 213 00:11:29,280 --> 00:11:31,800 Speaker 1: lot more healthy and successful than it actually was. Whatever 214 00:11:31,840 --> 00:11:35,760 Speaker 1: would be quick digression here. One of the funniest ironies 215 00:11:35,760 --> 00:11:37,880 Speaker 1: of Enron is that it was in many ways ahead 216 00:11:37,880 --> 00:11:40,120 Speaker 1: of its time, when most people were still connecting to 217 00:11:40,200 --> 00:11:42,960 Speaker 1: the Internet through screeching fifty six K dial up modems. 218 00:11:42,960 --> 00:11:45,480 Speaker 1: It saw a future in edge and cloud computing, even 219 00:11:45,520 --> 00:11:48,440 Speaker 1: if said terms didn't exist at the time, and streaming 220 00:11:48,520 --> 00:11:51,680 Speaker 1: video here's a funny one. In two thousand, it entered 221 00:11:51,679 --> 00:11:54,080 Speaker 1: into a twenty year deal with Blockbuster Video to allow 222 00:11:54,160 --> 00:11:57,680 Speaker 1: customers to stream films and TV shows through Enron's fiber network, 223 00:11:58,000 --> 00:12:00,520 Speaker 1: something that would take at Netflix and other decade to 224 00:12:00,559 --> 00:12:03,760 Speaker 1: realize as a product now. There wasn't really much of 225 00:12:03,760 --> 00:12:06,199 Speaker 1: a market for at the time because broadbam was pretty 226 00:12:06,320 --> 00:12:10,200 Speaker 1: rare back then, nor was it necessarily technologically possible. But 227 00:12:10,240 --> 00:12:13,040 Speaker 1: it was such a fun idea. People really liked it. 228 00:12:13,040 --> 00:12:15,360 Speaker 1: It was a good idea. Everyone wanted it to happen. 229 00:12:15,440 --> 00:12:18,360 Speaker 1: Not really though, Anyway. The deal collapsed after a year, 230 00:12:18,840 --> 00:12:21,440 Speaker 1: but that didn't stop Enron's creative accountants from booking the 231 00:12:21,440 --> 00:12:24,280 Speaker 1: deal based on its projective future revenue as a profitable 232 00:12:24,400 --> 00:12:27,640 Speaker 1: venture mark to market accounting. Baby, you gotta love it. 233 00:12:27,679 --> 00:12:30,000 Speaker 1: We love it, don't we, folks? All right, I'll stop 234 00:12:30,000 --> 00:12:32,680 Speaker 1: doing the impression. Still, it's hilarious to think about a 235 00:12:32,679 --> 00:12:35,160 Speaker 1: future world in which Blockbuster and Enron stuck it out 236 00:12:35,200 --> 00:12:36,959 Speaker 1: and the former didn't collapse. Around the time of the 237 00:12:36,960 --> 00:12:42,040 Speaker 1: global financial crisis, Enron also loved making special purpose vehicles 238 00:12:42,040 --> 00:12:45,000 Speaker 1: that existed either to generate revenue that didn't exist or 239 00:12:45,040 --> 00:12:47,680 Speaker 1: to hold toxic assets that would otherwise need to be disclosed, 240 00:12:47,720 --> 00:12:50,720 Speaker 1: with Enron then using its holdings insid entities to boost 241 00:12:50,720 --> 00:12:54,480 Speaker 1: its balance sheet. One White Wing was created and capitalized 242 00:12:54,480 --> 00:12:57,680 Speaker 1: by Enron hand an outside investor, and pretty much exclusively 243 00:12:57,720 --> 00:13:00,520 Speaker 1: bought assets from Enron, which allowed the company to recognized 244 00:13:00,520 --> 00:13:02,640 Speaker 1: sales and profits on its balance sheets even when they 245 00:13:02,679 --> 00:13:07,160 Speaker 1: were fundamentally contrived. Another set of entities known as LJM, 246 00:13:07,280 --> 00:13:09,760 Speaker 1: named after the first initial of Andy Fasthou's wife and 247 00:13:09,800 --> 00:13:12,840 Speaker 1: two children, which I mentioned earlier, did the same thing, 248 00:13:13,160 --> 00:13:16,040 Speaker 1: allowing the company to hide risky or failing investments, to 249 00:13:16,120 --> 00:13:20,439 Speaker 1: limit its perceived debt, and to generate artificial profits and revenues. 250 00:13:20,800 --> 00:13:23,920 Speaker 1: LJM two was creatively the second version of this idea. 251 00:13:24,559 --> 00:13:28,000 Speaker 1: Even though the assets that LJM held were ultimately dog shared, 252 00:13:28,080 --> 00:13:31,640 Speaker 1: the distance that LJM provided, combined with Enron's use of 253 00:13:31,679 --> 00:13:34,040 Speaker 1: marked to market accounting, allowed the company to turn a 254 00:13:34,120 --> 00:13:37,040 Speaker 1: multi billion collective failure into a resounding and on paper 255 00:13:37,080 --> 00:13:40,640 Speaker 1: profitable triumph. So how did this happen? How did it 256 00:13:40,640 --> 00:13:43,840 Speaker 1: go on for so long? Well? First, Enron was that 257 00:13:43,880 --> 00:13:46,680 Speaker 1: it is peak worth seventy billion dollars. Its fail would 258 00:13:46,679 --> 00:13:49,240 Speaker 1: be a failure for its investors in shareholders and nobody 259 00:13:49,360 --> 00:13:52,120 Speaker 1: besides the press that is wanted to ask tough questions 260 00:13:52,520 --> 00:13:55,560 Speaker 1: remember when they did that. Anyway, he had auditors, but 261 00:13:55,559 --> 00:13:57,559 Speaker 1: they were paid handsomely, turning a blind eye to the 262 00:13:57,600 --> 00:14:01,120 Speaker 1: criminal malfeasans at the heart of Enron to Arthur Anderson 263 00:14:01,200 --> 00:14:03,880 Speaker 1: surrendered its license in two thousand and two, bringing an 264 00:14:03,960 --> 00:14:06,480 Speaker 1: end to the company and resulting in eighty five thousand 265 00:14:06,480 --> 00:14:12,560 Speaker 1: employees losing their jobs. Well, look, it's not so much. 266 00:14:12,600 --> 00:14:15,440 Speaker 1: It only turned a blind eye as much as it 267 00:14:15,480 --> 00:14:19,440 Speaker 1: turned on a big paper shredder treading tons. And I'm 268 00:14:19,560 --> 00:14:22,880 Speaker 1: actually being literal. That's a measure of weight and not 269 00:14:23,080 --> 00:14:26,440 Speaker 1: figuratively of documents as Enron started to implode, a crime 270 00:14:26,440 --> 00:14:29,920 Speaker 1: for which it was later convicted of obstruction of justice. 271 00:14:29,960 --> 00:14:33,000 Speaker 1: But I've already talked about Enron's culture. But I'd be 272 00:14:33,080 --> 00:14:35,520 Speaker 1: remiss if I didn't mention that Enron's highest performers and 273 00:14:35,560 --> 00:14:39,520 Speaker 1: its leadership received hefty bonuses in company equity, motivating them 274 00:14:39,520 --> 00:14:42,680 Speaker 1: to keep the charade going. Mron's pension scheme I had 275 00:14:42,760 --> 00:14:46,440 Speaker 1: was basically entirely Enron stock, and employees were regularly encouraged 276 00:14:46,480 --> 00:14:48,960 Speaker 1: to buy more, with Kenneth Lay telling employees weeks before 277 00:14:48,960 --> 00:14:53,120 Speaker 1: the company's collapse that and I quote the company is 278 00:14:53,200 --> 00:14:56,960 Speaker 1: fundamentally sound and that they should hang on to their stock. 279 00:14:58,240 --> 00:15:01,160 Speaker 1: VIDID did use the word fundamentally. They used that, but 280 00:15:01,280 --> 00:15:04,480 Speaker 1: it's not in videos, not Enron. It's not. It's really different. 281 00:15:04,480 --> 00:15:06,400 Speaker 1: They're not doing this. I swear to God, I just 282 00:15:06,680 --> 00:15:11,560 Speaker 1: don't love the language anyway. Additionally, by the terms of 283 00:15:11,560 --> 00:15:14,320 Speaker 1: the Enron pension plan, employees were prevented from shifting their 284 00:15:14,320 --> 00:15:17,200 Speaker 1: holdings into other pension funds or other investments until they 285 00:15:17,200 --> 00:15:21,000 Speaker 1: turned fifty. When the company collapsed, those people lost everything, 286 00:15:21,080 --> 00:15:24,480 Speaker 1: even those who didn't know about Enron's criminality. George Maddock's, 287 00:15:24,520 --> 00:15:27,760 Speaker 1: a retired former Enron employee, had his entire retirement tied 288 00:15:27,800 --> 00:15:30,160 Speaker 1: up in fourteen thousand Enron shares, worth at the time 289 00:15:30,200 --> 00:15:33,240 Speaker 1: more than one point three million dollars. He was forced 290 00:15:33,280 --> 00:15:35,520 Speaker 1: to spend his golden years making ends meet by bowing 291 00:15:35,560 --> 00:15:38,160 Speaker 1: pastures and living in a run down East Texas farmhouse. 292 00:15:39,080 --> 00:15:42,800 Speaker 1: The US government brought criminal charges against Enron's top leadership. 293 00:15:43,280 --> 00:15:45,440 Speaker 1: Ken Lay was convicted of four counts of fraud and 294 00:15:45,480 --> 00:15:48,200 Speaker 1: making false statements, but died on a skiing vacation to 295 00:15:48,280 --> 00:15:52,240 Speaker 1: Aspen before sentencing. May he burn in Hell? Maybe Ronald 296 00:15:52,280 --> 00:15:54,080 Speaker 1: Reagan and him could get in the hot tub together, 297 00:15:54,200 --> 00:15:56,520 Speaker 1: the hot tub scot lava in him not feeling particularly 298 00:15:56,520 --> 00:15:59,480 Speaker 1: creative on that one anyway. Jeff Skilling was convicted of 299 00:15:59,480 --> 00:16:02,240 Speaker 1: twenty four n fraud and conspiracy and sentenced to twenty 300 00:16:02,240 --> 00:16:05,160 Speaker 1: four years in jail. This was of course reduced in 301 00:16:05,200 --> 00:16:07,720 Speaker 1: twenty thirteen on appeal to fourteen years, and he was 302 00:16:07,760 --> 00:16:09,760 Speaker 1: released to a half way house in twenty eighteen and 303 00:16:09,800 --> 00:16:13,400 Speaker 1: then freed in twenty nineteen. He's since then tried to 304 00:16:13,560 --> 00:16:16,680 Speaker 1: re enter the energy sector with one venture combining energy 305 00:16:16,680 --> 00:16:20,240 Speaker 1: trading and I kid you not blockchain technology, although nothing 306 00:16:20,280 --> 00:16:23,480 Speaker 1: really came of him. I want to take this moment 307 00:16:23,840 --> 00:16:26,200 Speaker 1: to give a big shout out to Quartz, which, when 308 00:16:26,240 --> 00:16:29,240 Speaker 1: covering Skillings attempt to come back, had this opening line 309 00:16:29,280 --> 00:16:31,680 Speaker 1: and I quote Jeffrey Skilling, those are a thing or 310 00:16:31,720 --> 00:16:34,960 Speaker 1: two of our blocks and chains. Woo get his ass. 311 00:16:36,040 --> 00:16:39,000 Speaker 1: Andy Fastau pled guilty to two counts, one of manipulation 312 00:16:39,080 --> 00:16:41,880 Speaker 1: of financial statements and one of self dealing, and received 313 00:16:41,920 --> 00:16:45,600 Speaker 1: ten years in prison. This was has ever later reduced 314 00:16:45,600 --> 00:16:47,840 Speaker 1: to six years, including two years of probation, in part 315 00:16:47,840 --> 00:16:52,240 Speaker 1: Becsey cooperated with the investigations against other Enron executives. He's 316 00:16:52,280 --> 00:16:54,320 Speaker 1: now a public speaker and a tech investor in a 317 00:16:54,400 --> 00:16:59,080 Speaker 1: company and AI company called Keene Corp. His wife, Leah, 318 00:16:59,080 --> 00:17:01,440 Speaker 1: who also worked at en On, received twelve months for 319 00:17:01,480 --> 00:17:03,960 Speaker 1: conspiracy to commit wire fraud and money laundering and for 320 00:17:04,000 --> 00:17:07,200 Speaker 1: submitting false tax returns. She was released from custody in 321 00:17:07,240 --> 00:17:11,240 Speaker 1: two thousand and five. Enron's implosion was entirely self inflicted 322 00:17:11,280 --> 00:17:15,400 Speaker 1: and horrifyingly painfully criminal. Yet it had plenty of collateral 323 00:17:15,440 --> 00:17:18,159 Speaker 1: damage to the US economy, to those companies that lent 324 00:17:18,280 --> 00:17:20,359 Speaker 1: him money, to its employees who lost their jobs and 325 00:17:20,400 --> 00:17:23,120 Speaker 1: their life savings and their retirements, and to those employees 326 00:17:23,119 --> 00:17:25,320 Speaker 1: that the companies most intangled with Enron, like those that 327 00:17:25,400 --> 00:17:29,880 Speaker 1: ausering firm Arthur Anderson. This isn't unique among corporate failures. 328 00:17:30,000 --> 00:17:33,959 Speaker 1: WelCom had some dodgy accounting practices. Nortel too. Both companies failed, 329 00:17:34,000 --> 00:17:36,480 Speaker 1: Both companies wrecked the lives of their employees, and these 330 00:17:36,480 --> 00:17:41,000 Speaker 1: failures had systemic economic consequences, especially in Canada, where Nortel 331 00:17:41,080 --> 00:17:43,159 Speaker 1: had its peak accounted for one third of the market 332 00:17:43,160 --> 00:17:46,359 Speaker 1: cap of all companies in the Toronto Stock Exchange. The 333 00:17:46,400 --> 00:17:49,440 Speaker 1: reason why Enron remains captured in our imagination and why 334 00:17:49,560 --> 00:17:52,760 Speaker 1: Nvidia is so vociferously opposed to being compared to him, 335 00:17:53,119 --> 00:17:55,520 Speaker 1: is the extent to which Enron and manipulated reality to 336 00:17:55,520 --> 00:17:57,840 Speaker 1: appear stronger and more successful than it was, and how 337 00:17:57,840 --> 00:17:59,879 Speaker 1: long it was able to get away with him. We 338 00:18:00,119 --> 00:18:02,320 Speaker 1: may have forgotten the memory of Enron. It happened over 339 00:18:02,359 --> 00:18:05,400 Speaker 1: two decades ago. After all, we haven't forgotten the instincts 340 00:18:05,400 --> 00:18:07,240 Speaker 1: that it gave us. It's why on those is twitch 341 00:18:07,280 --> 00:18:10,159 Speaker 1: when we see special purpose vehicles being used to buy GPUs, 342 00:18:10,280 --> 00:18:13,280 Speaker 1: and while we gag when we see mark to market accounting. 343 00:18:14,280 --> 00:18:17,520 Speaker 1: It's entirely possible. I genuinely believe this, that everything in 344 00:18:17,640 --> 00:18:22,760 Speaker 1: Video is doing is above board. Really it's great. Doesn't 345 00:18:22,760 --> 00:18:24,879 Speaker 1: do anything to help the deep pit of dread in 346 00:18:24,960 --> 00:18:43,720 Speaker 1: my stomach, though. Now let's remind ourselves of watching video 347 00:18:43,880 --> 00:18:46,600 Speaker 1: Is It's a company that previously focused on gaming or 348 00:18:46,680 --> 00:18:50,040 Speaker 1: into GPUs graphics processing units, but thanks to a stroke 349 00:18:50,119 --> 00:18:52,359 Speaker 1: of forethought, managed to be the company that sells the 350 00:18:52,359 --> 00:18:56,000 Speaker 1: shovels for the current AI bubble. It's data center. GPUs 351 00:18:56,040 --> 00:18:58,639 Speaker 1: are expensive. They all rely on a technology that in 352 00:18:58,720 --> 00:19:01,760 Speaker 1: Video created two decades ago called Kuda, which no other 353 00:19:01,800 --> 00:19:04,080 Speaker 1: competitor has an alternative to, or at least one with 354 00:19:04,160 --> 00:19:07,600 Speaker 1: the maturity and depth and functionality, and no other company 355 00:19:07,600 --> 00:19:10,200 Speaker 1: shares its focus on this arena, at least not one 356 00:19:10,200 --> 00:19:13,280 Speaker 1: with its history. These gps are expensive, and thanks to 357 00:19:13,359 --> 00:19:16,640 Speaker 1: Nvidia's near monopoly status, it's seen its share price growth 358 00:19:16,640 --> 00:19:19,639 Speaker 1: by over three hundred and fifty percent at the time 359 00:19:19,720 --> 00:19:22,199 Speaker 1: of writing in the past five years. It's now the 360 00:19:22,200 --> 00:19:24,960 Speaker 1: most valuable company in the world by market capitalization, and 361 00:19:25,000 --> 00:19:28,760 Speaker 1: it's insanely, insanely profitable. If you're looking at this through 362 00:19:28,800 --> 00:19:31,879 Speaker 1: the cold, unthinking lens of late stage capitalism, this all 363 00:19:31,920 --> 00:19:34,919 Speaker 1: sounds really good. I've basically described a company that has 364 00:19:34,920 --> 00:19:37,080 Speaker 1: an essential monopoly and the one thing required for a 365 00:19:37,200 --> 00:19:40,840 Speaker 1: high growth if we're talking exclusively about capex spending industry 366 00:19:40,840 --> 00:19:44,320 Speaker 1: to ever exist. Moreover, that monopoly is all but assured 367 00:19:44,359 --> 00:19:47,560 Speaker 1: thanks to Invidia's Kuda mote, its first move for advantage, 368 00:19:47,560 --> 00:19:51,160 Speaker 1: and its actual capacities and capabilities of the products itself, 369 00:19:51,280 --> 00:19:54,120 Speaker 1: on top of the massive shipping and hardware operation at Bill, 370 00:19:54,400 --> 00:19:56,639 Speaker 1: thereby allowing the company to charge a pretty penny to 371 00:19:56,680 --> 00:20:01,040 Speaker 1: its customers. And those customers, well, if we've temporarily forget 372 00:20:01,040 --> 00:20:03,240 Speaker 1: about the likes of Nebuus and Core, we even how 373 00:20:03,320 --> 00:20:05,880 Speaker 1: I wish I could forget about Core, we've permanently we're 374 00:20:05,880 --> 00:20:08,119 Speaker 1: talking about the biggest companies on the planet, ones that 375 00:20:08,359 --> 00:20:11,520 Speaker 1: surely will have no problems paying their bills. But my 376 00:20:11,640 --> 00:20:17,320 Speaker 1: worry is a little simpler and more direct, because I'm 377 00:20:17,359 --> 00:20:21,560 Speaker 1: worried about the certainty that these purchases are being made 378 00:20:21,600 --> 00:20:24,800 Speaker 1: for the right or most prudent of reasons. Back in 379 00:20:24,840 --> 00:20:27,560 Speaker 1: February twenty twenty three, I put out the rot economy 380 00:20:27,600 --> 00:20:30,359 Speaker 1: and how everything in tech has become oriented around growth 381 00:20:30,400 --> 00:20:33,720 Speaker 1: at all cost, even if it meant making products harder 382 00:20:33,760 --> 00:20:36,240 Speaker 1: to use as a means of increasing user engagement or 383 00:20:36,280 --> 00:20:39,320 Speaker 1: funneling them towards more profitable parts of an app. Back 384 00:20:39,320 --> 00:20:41,280 Speaker 1: in June twenty twenty four, I wrote and did an 385 00:20:41,320 --> 00:20:43,800 Speaker 1: episode about the rock Com bubble and my greater theory 386 00:20:43,840 --> 00:20:46,640 Speaker 1: that the tech industry has run out of hypergrowth ideas. 387 00:20:47,080 --> 00:20:50,480 Speaker 1: In simple terms, big Tech, Amazon, Google, Microsoft, and Meta, 388 00:20:50,640 --> 00:20:53,159 Speaker 1: but also a number of other companies no longer have 389 00:20:53,280 --> 00:20:55,919 Speaker 1: a next big thing and jumped on AI out of 390 00:20:55,920 --> 00:20:59,800 Speaker 1: an abundance of desperation. Shit. Look at Oracle. This company's 391 00:20:59,800 --> 00:21:03,000 Speaker 1: start off by selling databases and ERP systems to big 392 00:21:03,040 --> 00:21:05,760 Speaker 1: companies and then trapping said companies by making it really 393 00:21:05,760 --> 00:21:08,119 Speaker 1: really difficult to migrate to cheaper and better solutions, and 394 00:21:08,119 --> 00:21:11,159 Speaker 1: then bleeding said companies with ownerous licensing terms, including some 395 00:21:11,240 --> 00:21:13,439 Speaker 1: where you're paid by the number of CPU cours that 396 00:21:13,440 --> 00:21:16,719 Speaker 1: you use the application. Fucking Oracle. It doesn't do anything 397 00:21:16,760 --> 00:21:20,240 Speaker 1: new or exciting or impressive, and even with when presented 398 00:21:20,280 --> 00:21:22,719 Speaker 1: with the opportunity to do things that are useful or 399 00:21:22,160 --> 00:21:25,640 Speaker 1: in innovative, like when it bought some microsystems, it turns away. 400 00:21:25,880 --> 00:21:28,840 Speaker 1: I imagine that deep down Oracle recognizes that its current 401 00:21:29,000 --> 00:21:32,280 Speaker 1: model just isn't viable long term, and so it needed 402 00:21:32,280 --> 00:21:37,360 Speaker 1: something else. When you haven't thought about innovation, ever, it's 403 00:21:37,400 --> 00:21:40,520 Speaker 1: kind of hard to start, and thus generative AI probably 404 00:21:40,560 --> 00:21:43,680 Speaker 1: seemed like a godsend to Larry Ellison, who founded Oracle 405 00:21:43,720 --> 00:21:46,719 Speaker 1: and despite not being CEO, still creeps around the building 406 00:21:46,800 --> 00:21:50,400 Speaker 1: like the Grinch. I'll get to Oracle in a bit. 407 00:21:50,480 --> 00:21:53,000 Speaker 1: But we also live in an era where nobody knows 408 00:21:53,000 --> 00:21:56,240 Speaker 1: what big tech CEOs actually do other than make nearly 409 00:21:56,240 --> 00:21:59,399 Speaker 1: one hundred million dollars a year, meaning that somebody like 410 00:21:59,480 --> 00:22:02,159 Speaker 1: Satching the Dell can get called a thoughtful leader with 411 00:22:02,240 --> 00:22:06,280 Speaker 1: striking humility for pushing co pilot AI up every asshole 412 00:22:06,280 --> 00:22:10,280 Speaker 1: in Microsoft's product experience, even notepad a place that no 413 00:22:10,440 --> 00:22:13,840 Speaker 1: human being would want it and accelerating capital expenditures and 414 00:22:13,880 --> 00:22:16,520 Speaker 1: twenty eight billion dollars across the entirety of fiscal year 415 00:22:16,560 --> 00:22:19,200 Speaker 1: twenty twenty three to thirty four point nine billion dollars 416 00:22:19,240 --> 00:22:23,400 Speaker 1: in its last quarter in simpler terms, spending money makes 417 00:22:23,440 --> 00:22:25,760 Speaker 1: the CEO look busy, and at a time when there 418 00:22:25,800 --> 00:22:28,960 Speaker 1: were no other potential growth avenues left, AI was a 419 00:22:29,000 --> 00:22:32,680 Speaker 1: convenient way to make everybody look busy. Every department can 420 00:22:32,720 --> 00:22:35,560 Speaker 1: have an AI strategy, and every useless fucking manager and 421 00:22:35,640 --> 00:22:39,439 Speaker 1: executive can yell a service. Now CEO Bill McDermott did 422 00:22:39,480 --> 00:22:41,840 Speaker 1: back in twenty twenty two. Let me make it clear 423 00:22:41,840 --> 00:22:46,919 Speaker 1: to everybody here, everything you do AIAIAIAI AI. That's a 424 00:22:47,119 --> 00:22:50,640 Speaker 1: real goddamn quote. I should also add that chat GPT 425 00:22:50,800 --> 00:22:53,360 Speaker 1: was the first real meaningful hit that the American tech 426 00:22:53,359 --> 00:22:56,040 Speaker 1: industry has produced in a long long time, the last 427 00:22:56,280 --> 00:22:59,000 Speaker 1: being if I'm honest, Uber, and that's if we allow 428 00:22:59,119 --> 00:23:02,439 Speaker 1: successful yet not particularly good businesses into the pile. I 429 00:23:02,480 --> 00:23:04,840 Speaker 1: guess you could say Instagram Stories, but that was ripped 430 00:23:04,840 --> 00:23:10,360 Speaker 1: off from TikTok. No sorry, Snapchat Christ they nicked reels 431 00:23:10,359 --> 00:23:15,040 Speaker 1: from TikTok. I hate this fucking industry. Now. Look, if 432 00:23:15,040 --> 00:23:17,199 Speaker 1: we're assholes about it and we insist on things like 433 00:23:17,240 --> 00:23:22,680 Speaker 1: profitability and sustainability, US tech hasn't done so great. Snowflake 434 00:23:22,760 --> 00:23:24,480 Speaker 1: runs at a loss, Snap runs at a loss, and 435 00:23:24,480 --> 00:23:27,159 Speaker 1: while Uber's turned things around, somewhat. It hardly created the 436 00:23:27,160 --> 00:23:31,800 Speaker 1: next cloud computing or smartphone industry. Putting aside finances, the 437 00:23:31,920 --> 00:23:34,840 Speaker 1: last major hit was probably Vemo or zell and maybe 438 00:23:34,840 --> 00:23:38,159 Speaker 1: if I'm feeling generous, smart speakers like Amazon Echo or 439 00:23:38,200 --> 00:23:41,359 Speaker 1: Apple home Pod, but those I think, I think Echo 440 00:23:41,400 --> 00:23:44,800 Speaker 1: and Alexa has lost Amazon billions much like Uber. None 441 00:23:44,840 --> 00:23:46,480 Speaker 1: of these were the next big thing, which would be 442 00:23:46,560 --> 00:23:49,160 Speaker 1: fine except big tech needs more growth forever right now, 443 00:23:49,200 --> 00:23:52,040 Speaker 1: you pig. To be clear, I'm not saying there's been 444 00:23:52,080 --> 00:23:55,200 Speaker 1: no innovation, just nothing at the scale of smartphones and 445 00:23:55,240 --> 00:23:59,080 Speaker 1: cloud computing. This is why Google, Amazon, and Meta all 446 00:23:59,080 --> 00:24:02,200 Speaker 1: do twenty different things, although rarely for any length of time, 447 00:24:02,280 --> 00:24:04,720 Speaker 1: with these things often having a shelf life shorter than 448 00:24:04,840 --> 00:24:07,920 Speaker 1: strawberries left on your counter. Because the rot economy's growth 449 00:24:07,960 --> 00:24:10,600 Speaker 1: at all costs mindset exists only to please the markets 450 00:24:10,760 --> 00:24:14,879 Speaker 1: and the market's demand growth. Chat GBT was different. Not 451 00:24:14,920 --> 00:24:17,280 Speaker 1: only did it do something new, it also did it 452 00:24:17,320 --> 00:24:19,160 Speaker 1: in a way that was relatively easy to get people 453 00:24:19,200 --> 00:24:21,920 Speaker 1: to try and see the potential of. It was also 454 00:24:21,960 --> 00:24:24,359 Speaker 1: really easy to convince people that would become something bigger 455 00:24:24,359 --> 00:24:28,159 Speaker 1: and better, because that's what tech does. To quote Bender 456 00:24:28,160 --> 00:24:31,040 Speaker 1: and Hannah Ai, is a marketing term, a squishy way 457 00:24:31,040 --> 00:24:34,119 Speaker 1: of evoking futuristic visions of autonomous computers that could do 458 00:24:34,160 --> 00:24:36,960 Speaker 1: anything and everything for us. And because both customers and 459 00:24:37,000 --> 00:24:39,480 Speaker 1: analysts have been primed to believe in trust the tech industry, 460 00:24:39,680 --> 00:24:42,359 Speaker 1: everybody believed that whatever chat GBT was would be the 461 00:24:42,400 --> 00:24:45,280 Speaker 1: next big thing, and said next big thing was powered 462 00:24:45,320 --> 00:24:48,200 Speaker 1: by large language models which required GPU sold by one 463 00:24:48,200 --> 00:24:52,600 Speaker 1: goddamn company in video. AI became a very useful thing 464 00:24:52,640 --> 00:24:55,119 Speaker 1: to do. If a company wanted to seem futuristic and 465 00:24:55,160 --> 00:24:58,639 Speaker 1: attract investors, it could now integrate AI. If a hyperscaler 466 00:24:58,680 --> 00:25:01,360 Speaker 1: wanted to seem enterprising and lae it was building the future, 467 00:25:01,520 --> 00:25:03,439 Speaker 1: it could buy a bunch of GPUs or invest in 468 00:25:03,440 --> 00:25:07,240 Speaker 1: its own silicone, as Google, Microsoft, Amazon and Meta have done, 469 00:25:07,440 --> 00:25:10,600 Speaker 1: though Meta to only some extent, and of course shove 470 00:25:10,640 --> 00:25:14,560 Speaker 1: AI in every imaginable crevice of an app. Investors could 471 00:25:14,600 --> 00:25:18,320 Speaker 1: invest in AI companies, Retail investors, regular people could invest 472 00:25:18,359 --> 00:25:21,159 Speaker 1: AAI stocks. Tech reporters could write about something new and 473 00:25:21,359 --> 00:25:24,840 Speaker 1: do access journalism. Even harder LinkedIn perverts could write long 474 00:25:24,880 --> 00:25:27,639 Speaker 1: screens about AI. The markets could become obsessed with AI, 475 00:25:27,760 --> 00:25:30,280 Speaker 1: and yeah, you can kind of see how things got 476 00:25:30,320 --> 00:25:33,160 Speaker 1: out of control. Everybody now had something to do, an 477 00:25:33,200 --> 00:25:36,919 Speaker 1: excuse to do an AI thing, regardless of whether it 478 00:25:36,920 --> 00:25:39,680 Speaker 1: made sense. Because everybody else was doing it, you didn't 479 00:25:39,680 --> 00:25:43,520 Speaker 1: need to justify it anymore. Everybody was doing AI chat 480 00:25:43,560 --> 00:25:46,040 Speaker 1: GPT quickly became one of the most popular websites on 481 00:25:46,040 --> 00:25:48,880 Speaker 1: the Internet, all while open AI burned billions of dollars, 482 00:25:48,960 --> 00:25:51,560 Speaker 1: and because the media effectively published every single thought that 483 00:25:51,600 --> 00:25:54,680 Speaker 1: Sam Mortman had, such as that GPT four would automate 484 00:25:54,720 --> 00:25:57,199 Speaker 1: away some jobs and create others, and that he was 485 00:25:57,280 --> 00:26:00,000 Speaker 1: a little bit scared of him. AI as a tech 486 00:26:00,240 --> 00:26:03,240 Speaker 1: knowlogy and an idea and a symbolic stock trope and 487 00:26:03,280 --> 00:26:05,800 Speaker 1: a marketing tool and a myth became so powerful that 488 00:26:05,800 --> 00:26:09,040 Speaker 1: it could do anything, replace anyone, be worth anything, even 489 00:26:09,080 --> 00:26:12,000 Speaker 1: the future of your company. Amongst the hype, there was 490 00:26:12,000 --> 00:26:15,120 Speaker 1: an assumption related to scaling laws, summarized well by Charlie Meyer, 491 00:26:15,119 --> 00:26:16,720 Speaker 1: who was on the podcast a few months ago. You 492 00:26:16,720 --> 00:26:19,960 Speaker 1: should go listen to his episode. In twenty twenty, one 493 00:26:19,960 --> 00:26:22,040 Speaker 1: of the most important papers in the development of AI 494 00:26:22,200 --> 00:26:25,119 Speaker 1: was published, Scaling Laws for Neural Language Models, which came 495 00:26:25,160 --> 00:26:28,359 Speaker 1: from a group a OpenAI. The papers showed with just 496 00:26:28,400 --> 00:26:31,440 Speaker 1: a few charts incredibly compelling evidence that increasing the size 497 00:26:31,440 --> 00:26:34,639 Speaker 1: of large language models would increase their performance. This paper 498 00:26:34,680 --> 00:26:36,680 Speaker 1: was a large driver in the creation of GPD three 499 00:26:36,840 --> 00:26:39,399 Speaker 1: and today's L ANDM revolution, of course, the movements of 500 00:26:39,480 --> 00:26:42,720 Speaker 1: trillions of dollars in the stock market. In simple terms, 501 00:26:42,760 --> 00:26:45,240 Speaker 1: the paper suggested that shoving more training data and using 502 00:26:45,240 --> 00:26:48,199 Speaker 1: more compute power would exponentially increase the ability of a 503 00:26:48,240 --> 00:26:50,399 Speaker 1: model to do stuff. And to make a model that 504 00:26:50,480 --> 00:26:53,520 Speaker 1: did more stuff, you needed more GPUs and more data centers. 505 00:26:53,960 --> 00:26:56,159 Speaker 1: Did it matter that there was compelling evidence in twenty 506 00:26:56,200 --> 00:26:58,480 Speaker 1: twenty two and god to say it, Gary Marcus was 507 00:26:58,560 --> 00:27:00,879 Speaker 1: right that there were limits to galing laws and that 508 00:27:00,920 --> 00:27:05,520 Speaker 1: they would hit the point of diminishing returns. Nah amidst 509 00:27:05,600 --> 00:27:07,639 Speaker 1: all of this and video has sold over two hundred 510 00:27:07,680 --> 00:27:10,240 Speaker 1: billion dollars of GPU since the beginning of twenty twenty three, 511 00:27:10,280 --> 00:27:12,960 Speaker 1: becoming the largest company on the stock market and trading 512 00:27:13,040 --> 00:27:15,880 Speaker 1: over one hundred and seventy dollars as of writing this sentence, 513 00:27:16,000 --> 00:27:18,320 Speaker 1: only a few years after being worth less than twenty 514 00:27:18,359 --> 00:27:22,560 Speaker 1: dollars a share. You see meta, Google, Amazon, Microsoft all 515 00:27:22,600 --> 00:27:24,400 Speaker 1: wanted to be part of the future, so they sunk 516 00:27:24,400 --> 00:27:26,240 Speaker 1: a lot of money into in video, making up forty 517 00:27:26,240 --> 00:27:28,600 Speaker 1: two percent of its revenue in its fiscal year twenty 518 00:27:28,600 --> 00:27:32,280 Speaker 1: twenty five. Though there are some arguments about how much 519 00:27:32,440 --> 00:27:36,120 Speaker 1: exactly big text billowing capital expenditures are spent on GPUs. 520 00:27:36,400 --> 00:27:38,919 Speaker 1: Some estimate somewhere between forty one percent and more than 521 00:27:38,960 --> 00:27:42,159 Speaker 1: fifty percent of a data center's capex is spent on GPUs. 522 00:27:42,480 --> 00:27:44,600 Speaker 1: If you're wondering what the payoff is, well, you're in 523 00:27:44,720 --> 00:27:47,840 Speaker 1: good company. I estimate that there's only around sixty one 524 00:27:47,920 --> 00:27:50,399 Speaker 1: billion dollars in total genera of AI revenue, and that 525 00:27:50,440 --> 00:27:54,440 Speaker 1: includes every HYPERSCALA and neocloud. Large language models are limited, 526 00:27:54,520 --> 00:27:57,240 Speaker 1: AI agents are a pipe dreene and simply do not work. 527 00:27:57,440 --> 00:28:01,280 Speaker 1: AI powered products are unreliable, and coding l make developers slower, 528 00:28:01,320 --> 00:28:02,960 Speaker 1: and the cost of influence the way in which a 529 00:28:03,000 --> 00:28:06,159 Speaker 1: model produces its output keeps going up. The thing is, 530 00:28:06,359 --> 00:28:09,480 Speaker 1: what comes up tends to have a habit of coming down. 531 00:28:09,520 --> 00:28:12,159 Speaker 1: And I'm increasingly if the opinion that a return to 532 00:28:12,200 --> 00:28:15,199 Speaker 1: a terror firmer is coming sooner rather than later. But 533 00:28:15,280 --> 00:28:26,480 Speaker 1: that's for the next episode. Join me, then, thank you 534 00:28:26,520 --> 00:28:29,200 Speaker 1: for listening to Better Offline, The editor and composer of 535 00:28:29,240 --> 00:28:32,200 Speaker 1: the Better Offline theme song is Matasowski. You can check 536 00:28:32,240 --> 00:28:35,160 Speaker 1: out more of his music and audio projects at Mattasowski 537 00:28:35,240 --> 00:28:38,719 Speaker 1: dot com m A T T O. S O w 538 00:28:39,040 --> 00:28:42,600 Speaker 1: Ski dot com. You can email me at easy at 539 00:28:42,600 --> 00:28:45,280 Speaker 1: Better Offline dot com or visit Better Offline dot com 540 00:28:45,280 --> 00:28:47,720 Speaker 1: to find more podcast links and of course, my newsletter. 541 00:28:48,120 --> 00:28:50,640 Speaker 1: I also really recommend you go to chat dot Where's 542 00:28:50,640 --> 00:28:53,120 Speaker 1: youreed dot at to visit the discord, and go to 543 00:28:53,160 --> 00:28:56,560 Speaker 1: our slash Better Offline to check out our reddit. Thank 544 00:28:56,600 --> 00:28:59,959 Speaker 1: you so much for listening. Better Offline is a production 545 00:29:00,160 --> 00:29:03,080 Speaker 1: cool Zone Media. For more from cool Zone Media, visit 546 00:29:03,120 --> 00:29:06,280 Speaker 1: our website Coolzonemedia dot com or check us out on 547 00:29:06,360 --> 00:29:09,320 Speaker 1: the iHeartRadio app, Apple Podcasts, or wherever you get your 548 00:29:09,360 --> 00:29:09,920 Speaker 1: podcasts