1
00:00:02,800 --> 00:00:08,840
Speaker 1: Zone media, I'm better off line and this is ed Zetron. Wait,

2
00:00:20,680 --> 00:00:23,480
Speaker 1: that's right, folks. We're here for part two of our

3
00:00:23,560 --> 00:00:27,040
Speaker 1: multi part series about the final casualty of the incertification

4
00:00:27,200 --> 00:00:30,560
Speaker 1: process the financial markets. You see, we're in a world

5
00:00:30,560 --> 00:00:33,800
Speaker 1: where because companies can no longer stimulate growth by just

6
00:00:33,880 --> 00:00:37,800
Speaker 1: script mining their individual and business users, they've now resorted

7
00:00:37,840 --> 00:00:42,080
Speaker 1: to just lying. And lying is an interesting word because

8
00:00:42,120 --> 00:00:44,800
Speaker 1: they will claim we're not lying, we used words to

9
00:00:44,880 --> 00:00:49,080
Speaker 1: vaguely avoid telling you the truth, or we were technically

10
00:00:49,120 --> 00:00:51,640
Speaker 1: true in what we say. There are various kinds of

11
00:00:51,680 --> 00:00:54,160
Speaker 1: lies that don't have to just be I told you

12
00:00:54,200 --> 00:00:58,240
Speaker 1: something that was not categorically yes or no force. I mean,

13
00:00:58,720 --> 00:01:01,440
Speaker 1: you've spoken to other people, but they will say this

14
00:01:01,440 --> 00:01:05,280
Speaker 1: for legal purposes. So fine, it's not technically lying, it's

15
00:01:05,440 --> 00:01:08,399
Speaker 1: just bullshit. But I think it's important to give you

16
00:01:08,440 --> 00:01:11,000
Speaker 1: some examples and to talk about how analyst in the

17
00:01:11,040 --> 00:01:15,399
Speaker 1: media simply allowed these lights to propagate unchecked. Let me

18
00:01:15,440 --> 00:01:18,800
Speaker 1: set the scene. Our story begins on September tenth, twenty

19
00:01:18,800 --> 00:01:21,560
Speaker 1: twenty five. I was for some reason in Los Angeles

20
00:01:21,560 --> 00:01:24,600
Speaker 1: that I can't remember. Maybe a concert, maybe an interview.

21
00:01:24,800 --> 00:01:29,280
Speaker 1: Truly don't know. Oracle announced its unfillable, unpayable, three hundred

22
00:01:29,319 --> 00:01:32,160
Speaker 1: billion dollar deal with open AI, which they didn't have

23
00:01:32,160 --> 00:01:34,360
Speaker 1: the capacity to serve, and open a I didn't have

24
00:01:34,400 --> 00:01:36,280
Speaker 1: the cash to pay them, and this led to a

25
00:01:36,319 --> 00:01:39,840
Speaker 1: thirty percent or more bump in stock price. Analysts who

26
00:01:40,000 --> 00:01:43,280
Speaker 1: should extensibly be able to count, called it momentous and

27
00:01:43,319 --> 00:01:46,640
Speaker 1: said they were in shock. On September twenty second, twenty

28
00:01:46,680 --> 00:01:52,240
Speaker 1: twenty five, CEO Saffracats stepped down. She stepped down from oraclen.

29
00:01:52,680 --> 00:01:57,280
Speaker 1: Nobody seemed to think that was weird or suspicious or worrying.

30
00:01:57,440 --> 00:01:59,920
Speaker 1: This was this was meant to be the beginning of

31
00:02:00,080 --> 00:02:03,480
Speaker 1: the AI like boom too. I mean, we kind of

32
00:02:03,520 --> 00:02:05,960
Speaker 1: really in the AI bool, but this was meant to

33
00:02:06,040 --> 00:02:09,000
Speaker 1: be the even more bigger one, right, Why would you

34
00:02:09,320 --> 00:02:11,720
Speaker 1: step down? Why would you step down? Why would you

35
00:02:11,760 --> 00:02:14,560
Speaker 1: step down? And then have two guys like two different

36
00:02:14,600 --> 00:02:18,040
Speaker 1: They have co CEOs. Very sorry to my friend Jim,

37
00:02:18,120 --> 00:02:21,800
Speaker 1: but co CEOs don't usually do well. Look at BlackBerry.

38
00:02:21,800 --> 00:02:25,000
Speaker 1: But anyway, two months after that, Oracle stock was down

39
00:02:25,040 --> 00:02:28,720
Speaker 1: forty percent, with investors worried about Oracle's growing capex, which

40
00:02:28,760 --> 00:02:31,200
Speaker 1: is surprising. I suppose if you didn't think about how

41
00:02:31,280 --> 00:02:34,120
Speaker 1: Oracle would build the data centers to serve the three

42
00:02:34,200 --> 00:02:39,320
Speaker 1: hundred billion dollar deal. A makes me feel insane when

43
00:02:39,360 --> 00:02:42,320
Speaker 1: I say these things out how loud, But just to

44
00:02:42,360 --> 00:02:47,000
Speaker 1: be clear, anybody who traded on that got burned. Now

45
00:02:47,080 --> 00:02:50,160
Speaker 1: another thing happened on September twenty second, twenty twenty five,

46
00:02:50,480 --> 00:02:53,560
Speaker 1: in Vidio announced what they called a strategic partnership to

47
00:02:53,600 --> 00:02:56,919
Speaker 1: invest up to one hundred billion dollars and build ten

48
00:02:57,040 --> 00:03:00,000
Speaker 1: gigawads of data centers with open Ai, with the first

49
00:03:00,200 --> 00:03:02,400
Speaker 1: to gigawatt to be deployed in the second half of

50
00:03:02,440 --> 00:03:06,480
Speaker 1: twenty twenty six. Just a few questions from me personally,

51
00:03:06,760 --> 00:03:09,480
Speaker 1: where would the data centers go, How would open Ai

52
00:03:10,000 --> 00:03:12,959
Speaker 1: afford to build them? How would open Ai build a

53
00:03:13,000 --> 00:03:17,080
Speaker 1: gigawa in data centers in less than a year? Don't

54
00:03:17,120 --> 00:03:21,000
Speaker 1: ask questions. Pig back in your cage now. In video

55
00:03:21,080 --> 00:03:23,680
Speaker 1: stock bumped from one hundred and seventy five point three

56
00:03:23,880 --> 00:03:25,720
Speaker 1: to one hundred and eighty one dollars in the space

57
00:03:25,760 --> 00:03:28,080
Speaker 1: of a day. Said they're very normally moving on and

58
00:03:28,120 --> 00:03:29,839
Speaker 1: the media wrote about the story as if the deal

59
00:03:29,960 --> 00:03:32,360
Speaker 1: was done, with CNBC claiming that and I quote the

60
00:03:32,400 --> 00:03:35,760
Speaker 1: initial ten billion dollar tranch was expected to close within

61
00:03:35,800 --> 00:03:38,960
Speaker 1: a month or so once the transaction had been finalized.

62
00:03:39,160 --> 00:03:41,760
Speaker 1: I read at least ten stories that said in Video

63
00:03:41,800 --> 00:03:45,640
Speaker 1: had invested one hundred billion dollars in open Ai. Analysts

64
00:03:45,640 --> 00:03:47,680
Speaker 1: would go on to say that in Video was locking

65
00:03:47,720 --> 00:03:50,400
Speaker 1: in open Ai and I quote again to remain the

66
00:03:50,400 --> 00:03:54,120
Speaker 1: backbone of the next gen AI infrastructure, and that demand

67
00:03:54,160 --> 00:03:57,240
Speaker 1: for in video GPUs is effectively baked into the development

68
00:03:57,280 --> 00:04:00,960
Speaker 1: of frontier AI models, and that the deal, the partnership

69
00:04:00,960 --> 00:04:04,000
Speaker 1: between the two companies and Validate did in Vidia's long

70
00:04:04,080 --> 00:04:07,280
Speaker 1: term growth numbers with so much volume and compute capacity.

71
00:04:07,800 --> 00:04:10,200
Speaker 1: Others would say that in Video was enabling open Ai

72
00:04:10,280 --> 00:04:14,320
Speaker 1: to meet surging demand. Three analysts, rask Gone at Bernstein Laurier,

73
00:04:14,480 --> 00:04:17,640
Speaker 1: DA Davidson, and Wagner at Aaptors Capital all raised circular

74
00:04:17,680 --> 00:04:19,920
Speaker 1: deal concerns, but they were in the minority, and those

75
00:04:19,920 --> 00:04:22,880
Speaker 1: concerns were still often buried under buoyant optimism around the

76
00:04:22,880 --> 00:04:26,640
Speaker 1: prospects of the company, and nobody just fucking saying the words, Hey,

77
00:04:26,760 --> 00:04:29,680
Speaker 1: no one could afford any of the Oracle can't afford

78
00:04:29,720 --> 00:04:32,000
Speaker 1: to build it, they don't have the capacity, they ust

79
00:04:32,040 --> 00:04:35,440
Speaker 1: can't get paid by open Ai. Open Ai doesn't have

80
00:04:35,480 --> 00:04:39,640
Speaker 1: the money. But in Vidia, this deal also doesn't make sense.

81
00:04:39,720 --> 00:04:42,159
Speaker 1: You know. It feels like open ai is just promising

82
00:04:42,200 --> 00:04:46,320
Speaker 1: everyone's stuff they can't afford. All of these things seemed

83
00:04:46,360 --> 00:04:48,679
Speaker 1: like very obvious questions, but people just kind of clapped

84
00:04:48,680 --> 00:04:50,840
Speaker 1: their hands and went, fuck, yeah, we're doing Ai even

85
00:04:50,880 --> 00:04:54,240
Speaker 1: more now. But then there was this other thing about

86
00:04:54,240 --> 00:04:58,359
Speaker 1: the Nvidia deal. It's this quincy teeny weeny quincy little

87
00:04:58,400 --> 00:05:02,800
Speaker 1: problem one little pro everyone about the Invidia deal. The

88
00:05:02,839 --> 00:05:05,440
Speaker 1: deal that everyone said was done. It was a letter

89
00:05:05,480 --> 00:05:08,560
Speaker 1: of intent. It said so in the announcement and on

90
00:05:08,600 --> 00:05:11,680
Speaker 1: Invidia's November runnings it said that it had entered into

91
00:05:11,680 --> 00:05:15,520
Speaker 1: a letter of intent with an opportunity to invest in

92
00:05:15,560 --> 00:05:18,880
Speaker 1: open Ai. It turns out the deal didn't exist, and

93
00:05:18,960 --> 00:05:22,800
Speaker 1: everybody fell for him. Invidia hasn't sent a diamond, likely won't.

94
00:05:23,120 --> 00:05:25,359
Speaker 1: A letter of intent is also known as the concept

95
00:05:25,400 --> 00:05:28,560
Speaker 1: of a plan. A little over two weeks later, on

96
00:05:28,600 --> 00:05:32,240
Speaker 1: October first, twenty twenty five, Reuters reported that Samsung and

97
00:05:32,400 --> 00:05:35,520
Speaker 1: sk Heinex had signed letters of intent to supply memory

98
00:05:35,600 --> 00:05:39,640
Speaker 1: chips for open AI's data centers, with South Korea's Presidential

99
00:05:39,640 --> 00:05:42,960
Speaker 1: Office saying that said chip demand was expected to reach

100
00:05:43,279 --> 00:05:46,520
Speaker 1: nine hundred thousand wafers a month, with and I quote

101
00:05:46,640 --> 00:05:49,560
Speaker 1: much of that from Samsung and esk Heinex, which was

102
00:05:49,640 --> 00:05:53,800
Speaker 1: quickly extrapolated to meet around forty percent of global dram

103
00:05:53,839 --> 00:05:57,800
Speaker 1: outpum stocks in both companies to quote Reuter's Sword, with

104
00:05:57,920 --> 00:06:00,680
Speaker 1: Samsung climbing four percent and sk Heinex more than twelve

105
00:06:00,680 --> 00:06:04,440
Speaker 1: percent to an all time high. Analyst Jeff Kim KB

106
00:06:04,560 --> 00:06:07,560
Speaker 1: Securities said that and I quote, there have been worries

107
00:06:07,600 --> 00:06:11,880
Speaker 1: about high bandwidth memory prices falling next year on intensifying competition,

108
00:06:11,960 --> 00:06:14,640
Speaker 1: but such worries will now be easily resolved by the

109
00:06:14,760 --> 00:06:18,480
Speaker 1: strategic partnership, adding that since Stargate is a key project

110
00:06:18,560 --> 00:06:21,680
Speaker 1: led by President Trump, wrong lie, there also is a

111
00:06:21,720 --> 00:06:24,920
Speaker 1: possibility that partnership will have a positive impact on South

112
00:06:25,000 --> 00:06:28,839
Speaker 1: Korea's trade negotiations with the US, which is just to

113
00:06:28,880 --> 00:06:33,839
Speaker 1: be clear, total and utter bollocks. It's wank, it's scrap,

114
00:06:34,040 --> 00:06:37,800
Speaker 1: it's nonsense, boop them above all the technical terms. Donald

115
00:06:37,839 --> 00:06:40,840
Speaker 1: Trump is not leading Stargate. Stargate is a name used

116
00:06:40,880 --> 00:06:43,440
Speaker 1: to refer to data centers built by open Ai, by

117
00:06:43,440 --> 00:06:45,719
Speaker 1: which I mean other people building them for open Ai.

118
00:06:46,279 --> 00:06:49,120
Speaker 1: KB Securities is around forty three billion dollars of assets

119
00:06:49,200 --> 00:06:52,360
Speaker 1: under management. This is the level of analysis you get

120
00:06:52,360 --> 00:06:56,440
Speaker 1: from these analysts. This is how much they know. There

121
00:06:56,480 --> 00:07:00,200
Speaker 1: are people that study JoJo's bizarre adventure that have more

122
00:07:00,240 --> 00:07:04,520
Speaker 1: economic analysis than any of these fucking people. Most Star

123
00:07:04,600 --> 00:07:07,240
Speaker 1: Wars fans who have just memorized all the names of

124
00:07:07,279 --> 00:07:09,800
Speaker 1: the glop shittos of the world have better analysis than this.

125
00:07:11,280 --> 00:07:14,280
Speaker 1: But nevertheless, you'd think, right that these companies have made

126
00:07:14,280 --> 00:07:16,600
Speaker 1: all this noise, and they've talked about this stuff, the

127
00:07:16,640 --> 00:07:19,120
Speaker 1: Presidential Office all involved, you think we'd be able to

128
00:07:19,240 --> 00:07:22,559
Speaker 1: get a whiff of that on their earning scores, right well.

129
00:07:23,400 --> 00:07:26,560
Speaker 1: On I sk Hihenex's October twenty ninth, twenty twenty five

130
00:07:26,560 --> 00:07:30,480
Speaker 1: earning score. Weeks after the announcement, it's CEO Kim Wu

131
00:07:30,640 --> 00:07:33,480
Speaker 1: Yun was asked a question about high bandwidth memory growth

132
00:07:33,480 --> 00:07:38,120
Speaker 1: by Eskkim from Daiwa Securities and I quote, so this

133
00:07:38,280 --> 00:07:41,120
Speaker 1: is this is SKKM asking the question. Thank you very

134
00:07:41,200 --> 00:07:43,720
Speaker 1: much for taking my question. It is on demand now.

135
00:07:43,760 --> 00:07:45,960
Speaker 1: There have been a series of announcements of GPU and

136
00:07:46,000 --> 00:07:49,240
Speaker 1: a six supply cooperation between big techs and AI companies,

137
00:07:49,480 --> 00:07:53,080
Speaker 1: fueling expectations of further AI market growth. Then, against this

138
00:07:53,160 --> 00:07:56,040
Speaker 1: back drop, what is the company's outlook on high bend

139
00:07:56,080 --> 00:07:59,360
Speaker 1: with memory demand as well as a broadening of the

140
00:07:59,360 --> 00:08:03,040
Speaker 1: customer base, es sk Heinex. Thank you for the question. Now,

141
00:08:03,080 --> 00:08:06,080
Speaker 1: with upward adjustment in big text CAPEX and increasing investment

142
00:08:06,280 --> 00:08:09,400
Speaker 1: by AI companies, the high bandwidth memory market, even by

143
00:08:09,400 --> 00:08:11,800
Speaker 1: a conservative vestmer, will keep growing at an average of

144
00:08:12,360 --> 00:08:14,960
Speaker 1: over thirty percent for the next five years. I will

145
00:08:15,000 --> 00:08:17,520
Speaker 1: point to our recent letter of intent with open Ai

146
00:08:17,680 --> 00:08:20,800
Speaker 1: for large scale DRAM supply as an example of the

147
00:08:20,920 --> 00:08:24,040
Speaker 1: very strong demand for AI as well as the need

148
00:08:24,080 --> 00:08:26,840
Speaker 1: to secure AI memory based on HBM more than anything

149
00:08:26,840 --> 00:08:31,480
Speaker 1: else went developing AI technology. That is it. That is

150
00:08:31,520 --> 00:08:35,760
Speaker 1: the only mention of open AI. That's it. No other

151
00:08:35,800 --> 00:08:39,720
Speaker 1: mentions otherwise, sk Heinex has not added any guidance that

152
00:08:39,760 --> 00:08:42,560
Speaker 1: would suggest that it's DRAM sales will spike beyond overall

153
00:08:42,600 --> 00:08:44,640
Speaker 1: growth where it's already grown because everyone's running out of

154
00:08:44,679 --> 00:08:47,640
Speaker 1: ramthanks to fucking AI. Other than mentioning that it had

155
00:08:47,720 --> 00:08:50,640
Speaker 1: an I quote completed year twenty twenty six supply discussions

156
00:08:50,679 --> 00:08:53,240
Speaker 1: with key customers, there is no mention of open AI

157
00:08:53,400 --> 00:08:56,600
Speaker 1: in any earning's presentation. Now You may think there were

158
00:08:56,600 --> 00:08:59,000
Speaker 1: two companies. You said two company names. There were Samsung

159
00:08:59,040 --> 00:09:02,200
Speaker 1: as well. Right well. On Samsung's October thirtieth, twenty twenty

160
00:09:02,240 --> 00:09:05,480
Speaker 1: five earning school Samsung mentioned the term d RAM eighteen

161
00:09:05,520 --> 00:09:08,400
Speaker 1: times and neither mentioned open AI nor any letters of

162
00:09:08,400 --> 00:09:11,400
Speaker 1: intent of any kind. In its Q three twenty twenty

163
00:09:11,400 --> 00:09:15,199
Speaker 1: five earnings presentation, Samsung mentions it will a quote prioritize

164
00:09:15,200 --> 00:09:18,520
Speaker 1: the expansion of high bandwidth memory four business with differentiate

165
00:09:18,559 --> 00:09:23,200
Speaker 1: performance to address increasing AI demand. Great place to mention

166
00:09:23,280 --> 00:09:26,360
Speaker 1: open AI. Right, yeah, that would be if the money

167
00:09:26,400 --> 00:09:29,480
Speaker 1: existed and they were gonna do anything Now, analysts do

168
00:09:29,520 --> 00:09:31,360
Speaker 1: not appear to have noticed the lack of revenue from

169
00:09:31,400 --> 00:09:33,439
Speaker 1: an apparent deal for forty percent of the world's RAM.

170
00:09:33,720 --> 00:09:37,320
Speaker 1: Oh well, Pobody's perfect right. My frustration, you hear in

171
00:09:37,320 --> 00:09:40,280
Speaker 1: my voice is that I'm a guy with a Google doc.

172
00:09:41,040 --> 00:09:44,120
Speaker 1: I'm not a financial analyst, but I appeared to be

173
00:09:44,120 --> 00:09:45,960
Speaker 1: able to do what they're doing because I'm able to

174
00:09:46,000 --> 00:09:50,480
Speaker 1: fucking read a Both Samsung and s k Heinez's stocks

175
00:09:50,520 --> 00:09:52,600
Speaker 1: have continued to write since, and you'd be forgiven for

176
00:09:52,640 --> 00:09:54,760
Speaker 1: thinking this deal was something to do with it, even

177
00:09:54,800 --> 00:09:59,360
Speaker 1: though it was not. But silly season was not yet over, however,

178
00:09:59,400 --> 00:10:02,120
Speaker 1: and on October fifth, twenty twenty five, AMD announced that

179
00:10:02,120 --> 00:10:04,560
Speaker 1: it had entered a multi year, multi generation agreement with

180
00:10:04,600 --> 00:10:07,760
Speaker 1: open Ai to build six goddamn gigawads of data centers,

181
00:10:07,840 --> 00:10:10,000
Speaker 1: with the first one gig what deployment said to begin

182
00:10:10,080 --> 00:10:12,440
Speaker 1: in the second half of twenty twenty six, calling the

183
00:10:12,480 --> 00:10:15,240
Speaker 1: agreement definitive with terms that allowed open ai to buy

184
00:10:15,480 --> 00:10:19,400
Speaker 1: up to ten percent of AMD stock, festing over specific

185
00:10:19,480 --> 00:10:21,640
Speaker 1: milestones that started with the first giga what of data

186
00:10:21,640 --> 00:10:24,600
Speaker 1: center development, said data centers would use AMDs yet to

187
00:10:24,600 --> 00:10:27,559
Speaker 1: be released four to fifty GPUs. The deal would per

188
00:10:27,640 --> 00:10:31,760
Speaker 1: reuters bring in tens of billions of dollars of revenue. Now,

189
00:10:32,440 --> 00:10:34,920
Speaker 1: I know, hate to be an asshole, right, I know,

190
00:10:35,080 --> 00:10:37,719
Speaker 1: I'm always on this shit, right Yeah? Where would those

191
00:10:37,800 --> 00:10:41,160
Speaker 1: data centers go? How much? How much would they cost?

192
00:10:41,160 --> 00:10:43,400
Speaker 1: How would open Ai pay for them? Would the jips

193
00:10:43,440 --> 00:10:47,040
Speaker 1: be ready in time? Silence? Well, how dare you ask questions?

194
00:10:47,040 --> 00:10:49,680
Speaker 1: How dare you? Why are you asking questions? Number? Go up?

195
00:10:50,120 --> 00:10:53,080
Speaker 1: Number go up? Look look at number? Go up. AMD

196
00:10:53,240 --> 00:10:56,560
Speaker 1: shares searched by thirty four percent, with the analyst Dan

197
00:10:56,600 --> 00:10:58,880
Speaker 1: Ives of Webbush, who appears to dress like the San

198
00:10:58,960 --> 00:11:02,880
Speaker 1: Diego Padres Didy connect Jerseys, saying that this was a

199
00:11:03,000 --> 00:11:06,480
Speaker 1: major evaluation moment for AMD. As in the side, I've

200
00:11:06,520 --> 00:11:08,400
Speaker 1: said that in video would benefit from the meta verse

201
00:11:08,440 --> 00:11:10,959
Speaker 1: in twenty twenty one, and told CBS News in November

202
00:11:11,320 --> 00:11:14,120
Speaker 1: on November twenty second, twenty twenty one, that and I quote,

203
00:11:14,160 --> 00:11:17,040
Speaker 1: the metaverse was real and that Wall Street was looking

204
00:11:17,080 --> 00:11:22,800
Speaker 1: for winners, were they dan? Now one more thing, that

205
00:11:22,840 --> 00:11:26,280
Speaker 1: AMD's November earnings a month after that announcement might be

206
00:11:26,280 --> 00:11:29,280
Speaker 1: a barn burner full of remaining performance obligations from Open AI.

207
00:11:29,679 --> 00:11:32,000
Speaker 1: Things that Open appaying them moost tens of billions of

208
00:11:32,000 --> 00:11:35,560
Speaker 1: dollars of revenues right. In fact, CEO Lisa Sun who

209
00:11:35,760 --> 00:11:39,400
Speaker 1: said that AMD expected this partnership will significantly accelerated Stata

210
00:11:39,440 --> 00:11:42,160
Speaker 1: Center AI business and with the potential to generate well

211
00:11:42,200 --> 00:11:44,160
Speaker 1: over one hundred billion dollars in revenue over the next

212
00:11:44,160 --> 00:11:47,559
Speaker 1: few years in the announcement, and here's how AMD's ten

213
00:11:47,640 --> 00:11:50,440
Speaker 1: Q filing referred to it, and I quote. As of

214
00:11:50,440 --> 00:11:53,960
Speaker 1: September twenty seventh, twenty twenty five, the Aggagut transaction price

215
00:11:54,200 --> 00:11:56,880
Speaker 1: allocated to remaining performance applications on the contracts with an

216
00:11:56,920 --> 00:12:00,120
Speaker 1: original expected duration of more than one year. It was

217
00:12:00,120 --> 00:12:02,160
Speaker 1: two hundred and seventy nine million dollars, of which one

218
00:12:02,240 --> 00:12:04,320
Speaker 1: hundred and thirty nine million dollars is expected to be

219
00:12:04,360 --> 00:12:07,160
Speaker 1: recognized in the next twelve months. The revenue allocated to

220
00:12:07,200 --> 00:12:11,000
Speaker 1: remaining performance applications does not include amounts whichever original expected

221
00:12:11,040 --> 00:12:15,560
Speaker 1: duration of one year or less. Now, I don't know

222
00:12:15,600 --> 00:12:18,720
Speaker 1: about you. I did not I didn't do economics in school,

223
00:12:18,800 --> 00:12:21,760
Speaker 1: didn't do anythin like that, didn't do well at maths

224
00:12:21,760 --> 00:12:25,559
Speaker 1: in school. But two hundred and seventy nine million dollars

225
00:12:25,720 --> 00:12:30,640
Speaker 1: seems lower than over one hundred billion dollars. So I

226
00:12:30,640 --> 00:12:34,880
Speaker 1: guess just just no revenue from open AI, no revenue

227
00:12:34,920 --> 00:12:39,079
Speaker 1: of any kind. I guess. Now, AMD did raise guidance

228
00:12:39,080 --> 00:12:41,920
Speaker 1: by thirty five percent of the next five years, and

229
00:12:42,200 --> 00:12:44,880
Speaker 1: they're trailing. Twelve month revenue is thirty two billion dollars.

230
00:12:45,040 --> 00:12:47,559
Speaker 1: Tens of billions of dollars would surely lead to more

231
00:12:47,559 --> 00:12:50,520
Speaker 1: than a thirty five percent boost, because that would be

232
00:12:50,600 --> 00:12:54,440
Speaker 1: like eleven billion. That's just eleven billion dollars. That's a

233
00:12:54,480 --> 00:12:58,280
Speaker 1: lot less. That's a lot less than one hundred billion dollars.

234
00:12:58,280 --> 00:13:02,280
Speaker 1: I don't know guess all that was for nothing. No

235
00:13:02,360 --> 00:13:04,520
Speaker 1: follow up from the media, no questions from analysts, just

236
00:13:04,520 --> 00:13:07,040
Speaker 1: the shrug and we all move on. They were at ces.

237
00:13:07,160 --> 00:13:11,160
Speaker 1: Greg Brockman came on AMD's Lisa Sue said something about

238
00:13:11,240 --> 00:13:14,960
Speaker 1: Yachta flops, which really just pissed me off to think about.

239
00:13:16,360 --> 00:13:19,520
Speaker 1: Still nothing anyway. AMD stock is now down from a

240
00:13:19,559 --> 00:13:21,160
Speaker 1: high of two hundred and fifty nine bucks at the

241
00:13:21,240 --> 00:13:24,880
Speaker 1: end of October to around let's see md stocks understand

242
00:13:25,000 --> 00:13:28,040
Speaker 1: about two hundred and thirty one bucks right now. Everybody

243
00:13:28,040 --> 00:13:30,160
Speaker 1: who traded in based on an analyst and media comments

244
00:13:30,200 --> 00:13:44,360
Speaker 1: got fucked. Now let's go to my favor them, and

245
00:13:44,360 --> 00:13:48,800
Speaker 1: that's a little company named Broadcom. So back on September fifth,

246
00:13:48,800 --> 00:13:50,840
Speaker 1: Broadcom send and it's earning school that it had a

247
00:13:50,880 --> 00:13:54,640
Speaker 1: ten billion dollar order from a mysterious customer, which analysts

248
00:13:54,679 --> 00:13:57,760
Speaker 1: quickly assumed was open Ai, leading to Broadcom stock probably

249
00:13:57,800 --> 00:13:59,640
Speaker 1: nine percent and gradually increasing to a high of a

250
00:13:59,679 --> 00:14:02,679
Speaker 1: three hundred and sixty nine dollars on September tenth, before

251
00:14:02,720 --> 00:14:06,080
Speaker 1: declining a little until October thirteenth, when Broadcom announced a

252
00:14:06,200 --> 00:14:09,240
Speaker 1: ridiculous ten gigawa deal with open Ai, claiming that it

253
00:14:09,280 --> 00:14:11,720
Speaker 1: would deploy ten gigawats of open Ai designed chips, with

254
00:14:11,760 --> 00:14:14,360
Speaker 1: the first ranks to deploy the second half of twenty

255
00:14:14,360 --> 00:14:16,880
Speaker 1: twenty six and the entire deployment completed by the end

256
00:14:16,880 --> 00:14:19,400
Speaker 1: of twenty twenty ninth. So three fucking years for ten

257
00:14:19,440 --> 00:14:22,640
Speaker 1: fucking gigawatts. Just to give you an idea, it's about

258
00:14:22,760 --> 00:14:24,880
Speaker 1: two and a half years a gigawat right now. And

259
00:14:24,920 --> 00:14:27,200
Speaker 1: that's for open Ai slush fun projects, the ones that

260
00:14:27,240 --> 00:14:29,920
Speaker 1: have all the money and none of the oversight. Now.

261
00:14:29,960 --> 00:14:33,520
Speaker 1: The same day, President of Semiconductor Solutions Charlie Cowas added

262
00:14:33,520 --> 00:14:38,240
Speaker 1: that said mystery customers actually not open Ai, and I quote,

263
00:14:38,440 --> 00:14:40,640
Speaker 1: I would love to take a ten billion dollar purchase

264
00:14:40,760 --> 00:14:43,880
Speaker 1: order from my good friend Greg Brockman, CEO of open Ai,

265
00:14:44,040 --> 00:14:48,520
Speaker 1: Cowas said he has not given me that purchase order yet. Nevertheless,

266
00:14:48,520 --> 00:14:50,960
Speaker 1: Broadcom stock popped nine percent on the news about the

267
00:14:50,960 --> 00:14:53,440
Speaker 1: ten gigawa deal with CMBC, adding that the companies have

268
00:14:53,560 --> 00:14:56,440
Speaker 1: been working together for eighteen months because it's open Ai.

269
00:14:56,560 --> 00:15:00,000
Speaker 1: Nobody sat and thought about whether somebody at Broadcom was saying, wow,

270
00:15:00,000 --> 00:15:02,720
Speaker 1: well open has yet to ward these chips, yeah, and

271
00:15:02,760 --> 00:15:04,960
Speaker 1: whether that was a problem. In fact, the answer to

272
00:15:05,200 --> 00:15:08,360
Speaker 1: how does open Ai thought this appeared to be they'd

273
00:15:08,360 --> 00:15:11,720
Speaker 1: afford it when it came to analysts. Now I'm gonna

274
00:15:11,720 --> 00:15:14,360
Speaker 1: I'm about to read you a quote that is so

275
00:15:14,560 --> 00:15:17,160
Speaker 1: funny when you realize that this is someone who makes

276
00:15:17,240 --> 00:15:20,680
Speaker 1: like a shit ton of money. Ahem. The twenty twenty

277
00:15:20,720 --> 00:15:22,960
Speaker 1: six timelines set out by open Ai for the buildout

278
00:15:23,000 --> 00:15:25,640
Speaker 1: is aggressive, but the startup is also best position to

279
00:15:25,720 --> 00:15:28,600
Speaker 1: raise funds required for the project given the heights of

280
00:15:28,640 --> 00:15:33,680
Speaker 1: investor confidence, said gad Joe Sevilla, an analyst eMarketer. Financing

281
00:15:33,720 --> 00:15:36,160
Speaker 1: such a large chip deal will likely require a combination

282
00:15:36,240 --> 00:15:40,360
Speaker 1: of funding grounds pre orders, strategic investments and support from Microsoft,

283
00:15:40,480 --> 00:15:46,640
Speaker 1: as well as leveraging future revenue streams and potential credit facilities. Hogwash,

284
00:15:47,200 --> 00:15:53,880
Speaker 1: just complete, complete bollocks. Insane that people pay these analysts bullshit. Okay,

285
00:15:53,920 --> 00:15:56,800
Speaker 1: So just let's just break this down financing such a

286
00:15:56,840 --> 00:15:58,840
Speaker 1: large chip deal. So ten gigawats, you're going to be

287
00:15:58,880 --> 00:16:01,080
Speaker 1: looking at fifty bill dollars a gig was you look

288
00:16:01,120 --> 00:16:03,640
Speaker 1: at that half a trillion dollars. So to do this

289
00:16:03,960 --> 00:16:08,520
Speaker 1: for this aggressive buildout, which was ten giga Watson three years,

290
00:16:08,600 --> 00:16:12,560
Speaker 1: just not possible. They would fund this by pre orders,

291
00:16:13,240 --> 00:16:17,200
Speaker 1: pre orders for what exactly, because pre orders would be

292
00:16:17,200 --> 00:16:20,640
Speaker 1: then pre ordering something from Broadcom. Not really sure how

293
00:16:20,720 --> 00:16:25,200
Speaker 1: that fixed the problem. Strategic investments from whom just funding

294
00:16:25,240 --> 00:16:28,360
Speaker 1: I guess, support from Microsoft funding again, I guess, And

295
00:16:28,400 --> 00:16:30,680
Speaker 1: I don't know why Microsoft be paying for that, as

296
00:16:30,680 --> 00:16:36,560
Speaker 1: well as leveraging future revenue streams and potential credit for

297
00:16:36,640 --> 00:16:42,160
Speaker 1: say what what does that mean? Saving money doing like

298
00:16:42,560 --> 00:16:45,920
Speaker 1: a revenue trade deal? Like what are you talking about? Anyway,

299
00:16:45,960 --> 00:16:48,240
Speaker 1: you don't need to worry because open ai solution to

300
00:16:48,280 --> 00:16:51,560
Speaker 1: this was far simpler. It didn't order anything. During Broadcom's

301
00:16:51,640 --> 00:16:54,400
Speaker 1: November earning school, where Broadcom revealed that the ten billion

302
00:16:54,440 --> 00:16:57,040
Speaker 1: dollar order was actually from Anthropic, another l of them

303
00:16:57,040 --> 00:17:00,240
Speaker 1: started that burnst billions of dollars, which was actually buying

304
00:17:00,280 --> 00:17:05,880
Speaker 1: Google's TPUs from Broadcom, and he also added the Anthropic

305
00:17:05,920 --> 00:17:08,679
Speaker 1: had made a second eleven billion dollar deal. Alice somehow

306
00:17:08,680 --> 00:17:12,320
Speaker 1: believed that Anthropic is positioned to spend heavily despite being

307
00:17:12,359 --> 00:17:15,280
Speaker 1: yet another venture backed welfare recipient in the same flavor

308
00:17:15,280 --> 00:17:21,119
Speaker 1: as open Ai. Oh right, right, sorry, okay, Broadcom. You

309
00:17:21,119 --> 00:17:23,000
Speaker 1: may be wondering about the ten gig what deal though,

310
00:17:23,000 --> 00:17:26,240
Speaker 1: because those earnings right well, Broadcom CEO hoc Tan said

311
00:17:26,240 --> 00:17:28,560
Speaker 1: that he did not expect much in twenty twenty six

312
00:17:28,600 --> 00:17:32,080
Speaker 1: from the deal, and guidance did not change to reflect it. Now,

313
00:17:32,119 --> 00:17:34,119
Speaker 1: what was great about this as well? And I linked

314
00:17:34,119 --> 00:17:37,040
Speaker 1: to this in the newsletter. I really need you to

315
00:17:37,080 --> 00:17:39,760
Speaker 1: go and look up how hoc Tan had the conversation

316
00:17:39,840 --> 00:17:42,560
Speaker 1: because someone asked him for more detail and he just

317
00:17:42,560 --> 00:17:44,840
Speaker 1: straight up said, like, I don't want to talk about it.

318
00:17:44,840 --> 00:17:48,080
Speaker 1: It rocks. I honestly fucking love that. Anyway, Broadcom climbed

319
00:17:48,080 --> 00:17:50,920
Speaker 1: to a higher four hundred and twelve dollars a share

320
00:17:51,040 --> 00:17:52,840
Speaker 1: leading up to its earnings, and I imagine it did

321
00:17:52,840 --> 00:17:54,680
Speaker 1: so based on people trading on the belief that open

322
00:17:54,720 --> 00:17:57,080
Speaker 1: Air and Broadcom were doing a deal of some sort together,

323
00:17:57,200 --> 00:18:00,439
Speaker 1: which does not appear to be happening. While there is

324
00:18:00,440 --> 00:18:04,440
Speaker 1: an alleged seventy three billion dollar backlog for Broadcom, every

325
00:18:04,480 --> 00:18:07,600
Speaker 1: dollar from Anthropic is utterly questionable. Now, some of you

326
00:18:07,720 --> 00:18:11,800
Speaker 1: might say, ed, we can't just automatically distrust public companies. Actually,

327
00:18:11,920 --> 00:18:15,280
Speaker 1: yes we can. I'm doing it right now. Whenever a

328
00:18:15,359 --> 00:18:17,480
Speaker 1: company says a litter of intent, as in Video and

329
00:18:17,600 --> 00:18:20,800
Speaker 1: sk Heinez and Samsung did, it's important to immediately stop

330
00:18:20,800 --> 00:18:23,359
Speaker 1: taking the deal. Seriously until you get to a specific

331
00:18:23,400 --> 00:18:28,400
Speaker 1: word contract, not agreement or deal or announcement, but contract,

332
00:18:28,400 --> 00:18:33,040
Speaker 1: because contracts are the only thing that actually matters. A

333
00:18:33,080 --> 00:18:37,200
Speaker 1: great example of this is in in Video's last earnings,

334
00:18:37,200 --> 00:18:39,760
Speaker 1: they said they would be investing up to ten billion

335
00:18:39,800 --> 00:18:42,879
Speaker 1: dollars in Anthropic. Right, you'll notice that that was in

336
00:18:42,920 --> 00:18:45,480
Speaker 1: the middle of November. It's now January and we haven't

337
00:18:45,520 --> 00:18:49,280
Speaker 1: heard boo about it. It's probably because nothing's fucking happening.

338
00:18:49,520 --> 00:18:52,399
Speaker 1: And also that phrase up to as well means that

339
00:18:52,600 --> 00:18:55,000
Speaker 1: it will be However much in Video wants to invest,

340
00:18:55,320 --> 00:18:59,880
Speaker 1: if at all. They only just did their Intel investment. Anyway,

341
00:19:00,080 --> 00:19:03,359
Speaker 1: whenever they say these terms, you need to just get suspicious.

342
00:19:03,400 --> 00:19:06,280
Speaker 1: It's time. It's time to stop trusting them, because they've

343
00:19:06,280 --> 00:19:09,639
Speaker 1: played enough hide the sausage for one EPOC. I think

344
00:19:09,960 --> 00:19:13,160
Speaker 1: it's time for everybody, analysts, the media, members of Congress,

345
00:19:13,160 --> 00:19:15,560
Speaker 1: the fucking pope. I don't care, to start treating these

346
00:19:15,600 --> 00:19:20,440
Speaker 1: companies with actual suspicion and to start demanding timelines like

347
00:19:20,480 --> 00:19:23,080
Speaker 1: I said. In Video and Microsoft announced that fifteen billion

348
00:19:23,119 --> 00:19:26,479
Speaker 1: dollar investment, and in Anthropic, where's the money? Why does

349
00:19:26,520 --> 00:19:29,600
Speaker 1: the agreement say up to ten billion for in Video

350
00:19:29,680 --> 00:19:33,359
Speaker 1: and up to five billion for Microsoft. Where's the money?

351
00:19:34,720 --> 00:19:37,480
Speaker 1: Where is it? We haven't heard shit about it since.

352
00:19:37,800 --> 00:19:40,000
Speaker 1: What we have heard is that Anthropic is raising either

353
00:19:40,040 --> 00:19:42,080
Speaker 1: ten billion dollars or twenty five billion dollars at a

354
00:19:42,080 --> 00:19:45,000
Speaker 1: three hundred and fifty billion dollar valuation. Where'd the money go?

355
00:19:45,440 --> 00:19:49,200
Speaker 1: Where the money go from? Those fuck nuts? But they go, Sorry,

356
00:19:49,480 --> 00:19:51,880
Speaker 1: Jensen won for calling you a fuck n All right,

357
00:19:52,320 --> 00:19:56,800
Speaker 1: you're acting like one though, mate. Anyway, these deals are

358
00:19:56,840 --> 00:19:59,600
Speaker 1: announced with the intention of suggesting there is more revenue

359
00:19:59,760 --> 00:20:04,000
Speaker 1: and money and generative AI that actually exists. Furthermore, it

360
00:20:04,080 --> 00:20:07,640
Speaker 1: is irresponsible and actively harmful for analysts in the media

361
00:20:07,680 --> 00:20:10,360
Speaker 1: to continually act as if these deals will actually get

362
00:20:10,400 --> 00:20:13,639
Speaker 1: paid when you consider the financial conditions of these companies.

363
00:20:14,440 --> 00:20:16,919
Speaker 1: As part of its alleged funding announcement within Video and

364
00:20:16,960 --> 00:20:20,080
Speaker 1: Microsoft and Thropic also agreed to purchase thirty billion dollars

365
00:20:20,240 --> 00:20:22,640
Speaker 1: as your Compute, which is, for those of you who

366
00:20:22,680 --> 00:20:26,159
Speaker 1: don't know, the compute cloud platform, the competitors to Amazon

367
00:20:26,200 --> 00:20:29,760
Speaker 1: Web Services and Google Cloud that Microsoft runs. It also

368
00:20:29,800 --> 00:20:32,879
Speaker 1: agreed to Anthropic agreed to spend tens of billions of

369
00:20:32,880 --> 00:20:35,639
Speaker 1: dollars with Google Cloud. They ordered ten billion dollars of

370
00:20:35,720 --> 00:20:38,240
Speaker 1: chips from broad Coom earlier in twenty twenty five, and

371
00:20:38,280 --> 00:20:41,320
Speaker 1: apparently another eleven billion dollars of them in the last

372
00:20:41,320 --> 00:20:47,159
Speaker 1: fiscal quarter. How does Anthropic pay for them? How Anthropic

373
00:20:47,200 --> 00:20:50,359
Speaker 1: allegedly burned two point eight billion dollars last year twenty

374
00:20:50,400 --> 00:20:52,800
Speaker 1: twenty five, though I think they burn much more and

375
00:20:52,880 --> 00:20:56,280
Speaker 1: raise sixteen point five billion dollars in funding before Microsoft

376
00:20:56,320 --> 00:21:00,520
Speaker 1: and then Videa's imaginary money. How are investors tolerating Broadcom

377
00:21:00,520 --> 00:21:03,479
Speaker 1: not directly stating the future financial condition of this company

378
00:21:03,520 --> 00:21:06,640
Speaker 1: is questionable? Has Broadcom created a reserve for this deal?

379
00:21:07,000 --> 00:21:09,639
Speaker 1: If not, why not? Anthropic will make no more than

380
00:21:09,680 --> 00:21:11,840
Speaker 1: five billion dollars in twenty twenty five, I'm sure of it,

381
00:21:12,080 --> 00:21:15,040
Speaker 1: and raised about seventeen point five billion dollars with another

382
00:21:15,119 --> 00:21:18,000
Speaker 1: two point five billion dollars coming in debt. How can

383
00:21:18,000 --> 00:21:20,680
Speaker 1: it foreseeably afford to pay ten billion, eleven billion or

384
00:21:20,720 --> 00:21:24,119
Speaker 1: twenty one billion dollars considering it's already massive losses and

385
00:21:24,160 --> 00:21:27,480
Speaker 1: all those other obligations I mentioned. Will Jensen one hand

386
00:21:27,480 --> 00:21:29,600
Speaker 1: over ten billion dollars so that Anthropic can hand it

387
00:21:29,640 --> 00:21:33,600
Speaker 1: straight to Broadcom? I don't think so. I realize the

388
00:21:33,640 --> 00:21:35,840
Speaker 1: counter argument is that the companies aren't responsible for the

389
00:21:35,880 --> 00:21:38,560
Speaker 1: counterparty's financial health. But my argument is that it's the

390
00:21:38,600 --> 00:21:41,680
Speaker 1: responsibility of any public company to give a realistic view

391
00:21:41,680 --> 00:21:44,600
Speaker 1: of its financial health and revenue sources, which includes not

392
00:21:44,720 --> 00:21:46,560
Speaker 1: to give a chunk of its revenues from a startup

393
00:21:46,600 --> 00:21:48,760
Speaker 1: that can't afford to pay for its orders. There's no

394
00:21:48,840 --> 00:21:51,960
Speaker 1: counter for that. Anthropic cannot afford to pay broad Com

395
00:21:51,960 --> 00:21:54,600
Speaker 1: ten billion dollars right now or eleven billion dollars in

396
00:21:54,600 --> 00:21:57,240
Speaker 1: a few months. Where is the money coming from. They're

397
00:21:57,240 --> 00:22:00,360
Speaker 1: allegedly raising twenty five billion dollars. Great that ac shud

398
00:22:00,400 --> 00:22:04,080
Speaker 1: lead them with four billion dollars to pay for what

399
00:22:04,520 --> 00:22:09,119
Speaker 1: like thirty forty fifty billion dollars of compute costs. I

400
00:22:09,160 --> 00:22:13,120
Speaker 1: don't know. Again, I ain't no mathematician, but I can

401
00:22:13,160 --> 00:22:16,080
Speaker 1: put those numbers together and I can say that's a

402
00:22:16,119 --> 00:22:18,760
Speaker 1: pretty big minus. Again, I'm not even trying to be

403
00:22:18,880 --> 00:22:23,320
Speaker 1: sarcastic here. The things I'm saying are actually very rational.

404
00:22:24,240 --> 00:22:27,440
Speaker 1: They're very reasonable. Anthropic even if they raised twenty five

405
00:22:27,480 --> 00:22:30,399
Speaker 1: billion dollars, cannot pay even like a quarter of the

406
00:22:30,440 --> 00:22:34,040
Speaker 1: things they've agreed to their existent revenue sources. Even if

407
00:22:34,080 --> 00:22:36,359
Speaker 1: they made twenty billion dollars in twenty twenty six, which

408
00:22:36,359 --> 00:22:39,719
Speaker 1: they will not. By the way, they are absolutely I

409
00:22:39,840 --> 00:22:42,600
Speaker 1: just don't believe them about their revenues at all or

410
00:22:42,600 --> 00:22:47,359
Speaker 1: any of the estimates. But the problem is that in

411
00:22:47,400 --> 00:22:51,159
Speaker 1: any bubble, being really stupid and ignorant works right up

412
00:22:51,240 --> 00:22:53,960
Speaker 1: until it doesn't. And however harsh the dot com bubble

413
00:22:54,040 --> 00:22:56,680
Speaker 1: might have been, it wasn't harsh enough, and those who

414
00:22:56,720 --> 00:23:00,920
Speaker 1: were responsible were left unpunished and unashamed, and that guaranteed

415
00:23:00,920 --> 00:23:03,600
Speaker 1: that the cycle would happen again, and then it would

416
00:23:03,600 --> 00:23:08,800
Speaker 1: be worse. I want to be really abundantly clear about

417
00:23:08,800 --> 00:23:14,800
Speaker 1: what's happening. Every single stock you see growing because of AI,

418
00:23:15,040 --> 00:23:18,600
Speaker 1: outside of those selling RAM and GPUs, is actually growing

419
00:23:18,640 --> 00:23:23,880
Speaker 1: because of something else. Microsoft, Amazon, Google, and Meta all

420
00:23:23,920 --> 00:23:26,480
Speaker 1: have other products that are making the money and are

421
00:23:26,520 --> 00:23:30,680
Speaker 1: still growing. AI is not growing them. And because analysts

422
00:23:30,680 --> 00:23:32,880
Speaker 1: and investors in the media do not think about things

423
00:23:32,880 --> 00:23:35,480
Speaker 1: for two seconds, they have allowed themselves to be beaten

424
00:23:35,520 --> 00:23:39,159
Speaker 1: down and turned into supplicants for future public stock growth.

425
00:23:40,080 --> 00:23:42,880
Speaker 1: Investors have allowed themselves to be played, and the results

426
00:23:42,960 --> 00:23:46,040
Speaker 1: will be worse than the dot com bubble bursting by

427
00:23:46,040 --> 00:23:50,760
Speaker 1: several echelons and perhaps the biggest victims will be venture capital,

428
00:23:50,840 --> 00:23:53,679
Speaker 1: which is existentially threatened by the AI bubble, But the

429
00:23:53,720 --> 00:23:56,679
Speaker 1: actual victims will be regular people who made the mistake

430
00:23:56,720 --> 00:23:59,200
Speaker 1: of trusting analysts or anyone who told them to invest

431
00:23:59,200 --> 00:24:02,480
Speaker 1: in the markets. But I'm going to be really simplistic

432
00:24:02,560 --> 00:24:06,840
Speaker 1: for a second. I am skeptical of AI because everybody

433
00:24:06,880 --> 00:24:11,080
Speaker 1: loses money. I believe every AI company is unprofitable with

434
00:24:11,200 --> 00:24:14,240
Speaker 1: margins that are getting increasingly worse as they scale, and

435
00:24:14,280 --> 00:24:17,080
Speaker 1: as a result, none of them will be able to

436
00:24:17,080 --> 00:24:20,879
Speaker 1: get acquired or go public. This means that venture capitalists

437
00:24:20,880 --> 00:24:24,000
Speaker 1: that have sunk money into AI stocks and private AI stocks,

438
00:24:24,000 --> 00:24:25,800
Speaker 1: I should be clear, are going to be sitting on

439
00:24:25,840 --> 00:24:29,040
Speaker 1: a bunch of assets under management AEM the same assets

440
00:24:29,080 --> 00:24:31,560
Speaker 1: they collect fees on because that's how venture capitalists make

441
00:24:31,600 --> 00:24:34,520
Speaker 1: money that will eventually creter or go to zero because

442
00:24:34,520 --> 00:24:36,840
Speaker 1: there will be no way for any liquidity events so

443
00:24:37,000 --> 00:24:41,080
Speaker 1: taken in public or selling someone to occur. This is

444
00:24:41,119 --> 00:24:44,400
Speaker 1: at a time of historically low liquidity for venture capitalists,

445
00:24:44,440 --> 00:24:47,359
Speaker 1: with Pitchbook estimating that there would be only one hundred

446
00:24:47,440 --> 00:24:50,520
Speaker 1: point eight billion dollars in venture capital funds available at

447
00:24:50,560 --> 00:24:54,080
Speaker 1: the end of last year. Venture capitalists raise money from

448
00:24:54,119 --> 00:24:56,840
Speaker 1: limited partners who invest in venture capital with the hope

449
00:24:56,840 --> 00:25:00,440
Speaker 1: of returns that outpace investing in the public markets. Venture

450
00:25:00,480 --> 00:25:04,080
Speaker 1: capital vastly overinvested in twenty twenty one and twenty twenty

451
00:25:04,080 --> 00:25:06,600
Speaker 1: two as well, and this was also a problem with

452
00:25:06,640 --> 00:25:09,960
Speaker 1: private equity. In simple terms, this means that these funds

453
00:25:09,960 --> 00:25:12,400
Speaker 1: are sitting on tons of stock that they cannot shift,

454
00:25:12,600 --> 00:25:14,159
Speaker 1: and the longer it takes for a company to w

455
00:25:14,160 --> 00:25:16,600
Speaker 1: either go public or required, the more likely it is

456
00:25:16,640 --> 00:25:18,359
Speaker 1: the VC or the P firm will have to mark

457
00:25:18,400 --> 00:25:22,240
Speaker 1: down its value to something more realistic. This is so

458
00:25:22,440 --> 00:25:25,920
Speaker 1: bad that, according to Carter, as of August twenty twenty four,

459
00:25:25,960 --> 00:25:29,680
Speaker 1: and that CARTA some of you can't understand how I talk.

460
00:25:30,040 --> 00:25:32,520
Speaker 1: As of August twenty twenty four, less than ten percent

461
00:25:32,520 --> 00:25:35,600
Speaker 1: of VC funds raised in twenty twenty one have made

462
00:25:35,640 --> 00:25:39,040
Speaker 1: any distributions to their investors, that's handing any kind of

463
00:25:39,080 --> 00:25:42,760
Speaker 1: cash out. In a piece from September twenty twenty five,

464
00:25:43,080 --> 00:25:45,639
Speaker 1: Carter also revealed that about fifteen percent of funds from

465
00:25:45,640 --> 00:25:49,719
Speaker 1: twenty twenty three have generated any disbursements as of Q two,

466
00:25:49,760 --> 00:25:52,480
Speaker 1: twenty twenty five, and the media net internal rate of

467
00:25:52,480 --> 00:25:55,160
Speaker 1: return was a median zero point one percent, meaning that

468
00:25:55,440 --> 00:25:59,439
Speaker 1: at best most investors got their money back and absolutely

469
00:25:59,440 --> 00:26:03,000
Speaker 1: nothing else. And just to be clear, you don't invest

470
00:26:03,640 --> 00:26:06,280
Speaker 1: so you can get an exact amount. You don't give

471
00:26:06,320 --> 00:26:08,600
Speaker 1: someone a dollar, wait four years and then get a

472
00:26:08,640 --> 00:26:11,960
Speaker 1: dollar back. That's not investing, that's just putting money in

473
00:26:11,960 --> 00:26:14,439
Speaker 1: a box. And in fact, investing in venture capital has

474
00:26:14,520 --> 00:26:17,320
Speaker 1: kind of fucking sugged for a while. According to Carter,

475
00:26:17,680 --> 00:26:20,160
Speaker 1: as of the end of Q two, most VC funds

476
00:26:20,160 --> 00:26:23,359
Speaker 1: across all recent vintages that's after twenty eighteen had a

477
00:26:23,400 --> 00:26:26,520
Speaker 1: TVPI somewhere between zero point eight x and two x

478
00:26:26,880 --> 00:26:29,320
Speaker 1: TVPI I'll get to in a second, but there are

479
00:26:29,320 --> 00:26:33,320
Speaker 1: some areas where standout tvpis are surfacing. TVPI means total

480
00:26:33,359 --> 00:26:35,280
Speaker 1: value to paid in capital, or the amount of money

481
00:26:35,320 --> 00:26:38,600
Speaker 1: you made for each dollar invested. So if you invested

482
00:26:38,640 --> 00:26:42,280
Speaker 1: a dollar point eight x would mean you got eighty

483
00:26:42,320 --> 00:26:44,240
Speaker 1: cents back, and if you invested a dollar you get

484
00:26:44,240 --> 00:26:46,880
Speaker 1: two bucks on two X. I hope that makes sense. Now,

485
00:26:46,920 --> 00:26:49,000
Speaker 1: there's a chart in the newsletter version of this episode,

486
00:26:49,000 --> 00:26:50,800
Speaker 1: which I encourage you to check out, but it tells

487
00:26:50,880 --> 00:26:53,200
Speaker 1: you that for the most part, vcs have struggled to

488
00:26:53,200 --> 00:26:57,840
Speaker 1: provide even money returns. Since twenty seventeen, decent TVPI is

489
00:26:57,880 --> 00:27:00,359
Speaker 1: two point five x, and as you'll see from the chart,

490
00:27:01,359 --> 00:27:05,520
Speaker 1: things basically collapsed since twenty twenty one. Companies are not

491
00:27:05,560 --> 00:27:07,960
Speaker 1: going public or being acquired at the same rate, meaning

492
00:27:08,040 --> 00:27:11,040
Speaker 1: that investor capital is increasingly locked up, meaning that limited

493
00:27:11,080 --> 00:27:14,480
Speaker 1: partners are still wing for a payoff from the last bubbles.

494
00:27:14,920 --> 00:27:18,160
Speaker 1: Let learn this one now. Carter would update the chart

495
00:27:18,160 --> 00:27:21,960
Speaker 1: in December twenty twenty five, and things would somehow get worse.

496
00:27:22,280 --> 00:27:25,600
Speaker 1: TVPI soured further, suggesting a further lack of exits across

497
00:27:25,640 --> 00:27:28,600
Speaker 1: the board. The only slight improvement was the median IRR,

498
00:27:28,920 --> 00:27:31,479
Speaker 1: which rose from zero point five percent from funds from

499
00:27:31,520 --> 00:27:35,000
Speaker 1: twenty twenty one and zero point one percent for funds

500
00:27:35,040 --> 00:27:39,960
Speaker 1: from twenty twenty two. Those are still real, fucking terrible.

501
00:27:40,080 --> 00:27:46,679
Speaker 1: Those are absolutely astonishing. I mean, a good IRR is

502
00:27:46,720 --> 00:27:53,760
Speaker 1: ideally like fifteen to twenty percent, not zero point five percent.

503
00:27:53,840 --> 00:27:56,600
Speaker 1: I don't know, but in simpler terms, we are looking

504
00:27:56,640 --> 00:27:59,520
Speaker 1: at years of locked up capital, even venture capital cash starved,

505
00:27:59,520 --> 00:28:16,399
Speaker 1: and a little ESPRA. The worst part is that all

506
00:28:16,440 --> 00:28:18,720
Speaker 1: of this is happening during a generational increase in the

507
00:28:18,760 --> 00:28:21,680
Speaker 1: amounts that startups need to raise thanks to the ruinous

508
00:28:21,680 --> 00:28:24,920
Speaker 1: costs of generative AI and the negative margins of AI

509
00:28:25,000 --> 00:28:27,480
Speaker 1: powered services. Now, I'm going to quote myself from a

510
00:28:27,480 --> 00:28:31,800
Speaker 1: December fifth newsletter called the Ways the AI Bubble Might Burst. Cursor,

511
00:28:31,880 --> 00:28:34,520
Speaker 1: Anthropic's largest customer and now it's biggest competitor in the

512
00:28:34,560 --> 00:28:37,159
Speaker 1: AAI coding sphere, raised two point three billion dollars in

513
00:28:37,200 --> 00:28:40,360
Speaker 1: November after raising nine hundred million dollars in June. Now

514
00:28:40,920 --> 00:28:43,360
Speaker 1: for twenty twenty five, and I should add these are

515
00:28:43,440 --> 00:28:46,840
Speaker 1: on revenues of eighty three million dollars in the month

516
00:28:46,920 --> 00:28:51,120
Speaker 1: of I think November October, because they said annualized revenues

517
00:28:51,160 --> 00:28:53,720
Speaker 1: of one billion dollars. Is dogship perplexity one of the

518
00:28:53,720 --> 00:28:57,960
Speaker 1: most And I say this presitantly popular AI companies raised

519
00:28:57,960 --> 00:29:00,600
Speaker 1: two hundred million dollars in September after using one hundred

520
00:29:00,640 --> 00:29:02,800
Speaker 1: million dollars in July, after seeming to fail to raise

521
00:29:02,840 --> 00:29:05,680
Speaker 1: five hundred million dollars in May. After raising five hundred

522
00:29:05,760 --> 00:29:08,640
Speaker 1: million dollars in December twenty twenty four for a fucking

523
00:29:08,800 --> 00:29:13,000
Speaker 1: search engine, Cognition raised four hundred million dollars in September

524
00:29:13,080 --> 00:29:15,680
Speaker 1: after raising three hundred million dollars In March, Coher raised

525
00:29:15,680 --> 00:29:18,040
Speaker 1: one hundred million dollars in said temper, a month after

526
00:29:18,080 --> 00:29:22,160
Speaker 1: it raised five hundred million dollars. That's a lot of money,

527
00:29:22,440 --> 00:29:24,320
Speaker 1: and that's a lot of money in a very small

528
00:29:24,440 --> 00:29:28,000
Speaker 1: amount of time, and that's taking money away from everyone

529
00:29:28,000 --> 00:29:32,320
Speaker 1: else because they need that money. You need the money

530
00:29:32,360 --> 00:29:34,640
Speaker 1: to invest in the stops. But none of these companies

531
00:29:34,640 --> 00:29:36,440
Speaker 1: are profitable, nor do they have a path to end

532
00:29:36,560 --> 00:29:40,320
Speaker 1: acquisition or IBO. And why do you say, why why

533
00:29:40,360 --> 00:29:42,880
Speaker 1: do you ask? Ede? You're being so horrible? What's because

534
00:29:42,960 --> 00:29:46,360
Speaker 1: even the most advanced AI software companies ultimately prompting open

535
00:29:46,440 --> 00:29:49,880
Speaker 1: AI or Anthropics models, meaning that their only real intellectual

536
00:29:49,880 --> 00:29:52,720
Speaker 1: property is those prompts and their staff and whatever they

537
00:29:52,760 --> 00:29:54,920
Speaker 1: can build around the models they don't control, which has

538
00:29:54,960 --> 00:29:58,560
Speaker 1: been obvious from the meager acquisitions we've seen so far.

539
00:29:59,000 --> 00:30:03,400
Speaker 1: Another fun fact, Anthropic ended up cutting off people that

540
00:30:03,480 --> 00:30:06,760
Speaker 1: worked to Xai that were using Cursor from using their models. So,

541
00:30:06,960 --> 00:30:10,840
Speaker 1: just to be clear, Cursor completely different company Anthropic. If

542
00:30:10,880 --> 00:30:13,120
Speaker 1: you work at a company at Anthropic doesn't like and

543
00:30:13,160 --> 00:30:15,000
Speaker 1: I know we don't, we don't like elon mask, I

544
00:30:15,120 --> 00:30:18,800
Speaker 1: get that. But if Anthropic can arbitrarily cut off their

545
00:30:18,800 --> 00:30:24,040
Speaker 1: models from somebody's completely independent product. Just because they're deciding

546
00:30:24,080 --> 00:30:27,600
Speaker 1: to cut off a competitor, they're just going to broaden

547
00:30:27,680 --> 00:30:30,840
Speaker 1: the term competitor to mean anyone they don't like, anyone

548
00:30:30,880 --> 00:30:35,200
Speaker 1: that gets in the way. Now, Windsurf, which was allegedly

549
00:30:35,240 --> 00:30:37,840
Speaker 1: being sold to open Ai last year, ended up selling

550
00:30:37,840 --> 00:30:40,680
Speaker 1: its assets to Cognition in July twenty twenty five, with

551
00:30:40,760 --> 00:30:43,280
Speaker 1: Google paying two point four billion dollars just to hire

552
00:30:43,280 --> 00:30:46,960
Speaker 1: its co founders. And this nebulous licensing agreement similar to

553
00:30:46,960 --> 00:30:49,160
Speaker 1: the thing it did with Character to AI, where it

554
00:30:49,160 --> 00:30:51,800
Speaker 1: paid two point seven billion dollars to reheigh Nome Shazir,

555
00:30:52,160 --> 00:30:54,479
Speaker 1: who he was one of the authors of the Attention

556
00:30:54,560 --> 00:30:57,440
Speaker 1: is All You Need paper that started this spars licenses

557
00:30:57,520 --> 00:30:59,959
Speaker 1: tech can pay off the stock of its remaining staff.

558
00:31:00,000 --> 00:31:02,480
Speaker 1: This is also exactly what Microsoft did with Inflection AI

559
00:31:02,520 --> 00:31:05,560
Speaker 1: and it's co founder, mister Suliman, who, and I quote

560
00:31:05,560 --> 00:31:10,640
Speaker 1: someone who messaged me really recently, is a huge fucking asshole. Now,

561
00:31:11,080 --> 00:31:13,520
Speaker 1: mister Sulliman, if you hear this, I'm just quoting someone

562
00:31:13,560 --> 00:31:16,240
Speaker 1: that works for you. If they're saying that about you,

563
00:31:16,240 --> 00:31:18,680
Speaker 1: perhaps it's worth asking why are people calling me a

564
00:31:18,680 --> 00:31:21,280
Speaker 1: big fucking asshole. I don't know. Must Far come on

565
00:31:21,280 --> 00:31:24,560
Speaker 1: the show and ask me yourself. Now open Aiyes, acquisitions

566
00:31:24,560 --> 00:31:27,760
Speaker 1: of Statsic one point one billion dollars, Ioproducts six point

567
00:31:27,760 --> 00:31:30,400
Speaker 1: five billion dollars, in Neptune four hundred million dollars. We're

568
00:31:30,440 --> 00:31:35,880
Speaker 1: all in stock. Every other major acquisition of this era, Wiz, Confluence, Confluent,

569
00:31:35,920 --> 00:31:39,280
Speaker 1: even Informatica and so on, is either someone trying to

570
00:31:39,320 --> 00:31:42,480
Speaker 1: pretend that something is related to AI, like Wiz, or

571
00:31:42,640 --> 00:31:45,240
Speaker 1: trying to say that a data streaming platform is heavily

572
00:31:45,320 --> 00:31:50,080
Speaker 1: AI related because AI needs data, which which may be true,

573
00:31:50,240 --> 00:31:53,880
Speaker 1: but it doesn't mean that AI companies are selling and

574
00:31:53,920 --> 00:31:56,040
Speaker 1: they're not, by the way, which is a problem. As

575
00:31:56,080 --> 00:31:58,480
Speaker 1: forty one percent of US venture dollars in twenty twenty

576
00:31:58,480 --> 00:32:01,600
Speaker 1: five went to AI as of August, and according to Axios,

577
00:32:01,600 --> 00:32:04,720
Speaker 1: the global number was around fifty one percent. A crisis

578
00:32:04,800 --> 00:32:07,680
Speaker 1: is brewing. NERD Lawyer back in October wrote about the

579
00:32:07,720 --> 00:32:11,560
Speaker 1: explosive growth of secondary markets, and I quote enter the

580
00:32:11,600 --> 00:32:14,400
Speaker 1: secondary market a once niche corner of venture capital that

581
00:32:14,480 --> 00:32:19,280
Speaker 1: has transferred into a primary liquidity mechanism. What's remarkable is

582
00:32:19,320 --> 00:32:22,480
Speaker 1: how quickly this market is matured at least five major

583
00:32:22,560 --> 00:32:25,760
Speaker 1: venture funds have hired full time staff dedicated to manufacturing

584
00:32:25,800 --> 00:32:30,680
Speaker 1: non traditional exits. As Hans Swilden's CEO of Industry Ventures explained,

585
00:32:31,160 --> 00:32:33,480
Speaker 1: all the brand named funds are all staffing and thinking

586
00:32:33,480 --> 00:32:38,160
Speaker 1: through liquidity structures and professional buyers have flooded in. Megafunds

587
00:32:38,200 --> 00:32:42,800
Speaker 1: specializing in secondaries have raised unprecedented amounts. Lexington raised a

588
00:32:42,840 --> 00:32:46,160
Speaker 1: record twenty three billion dollars, or harbor Vest, Ardient and

589
00:32:46,200 --> 00:32:49,040
Speaker 1: Collar Capital have raised funds in the ten to twenty

590
00:32:49,080 --> 00:32:53,120
Speaker 1: billion dollar range. In simpler terms, there are now hot

591
00:32:53,120 --> 00:32:57,400
Speaker 1: potato funds where either another limited partner buys and other's allocation,

592
00:32:57,840 --> 00:33:00,480
Speaker 1: the companies themselves by back their stock, or the stock

593
00:33:00,560 --> 00:33:03,719
Speaker 1: is resold to other private investors who are building venture

594
00:33:03,720 --> 00:33:08,560
Speaker 1: capital firms out of shit. The venture capitalists can't sell.

595
00:33:09,120 --> 00:33:11,840
Speaker 1: You're seeing where the problem might be. Eventually, someone's going

596
00:33:11,880 --> 00:33:15,720
Speaker 1: to be. It's not Bomberman. It's not good now from

597
00:33:15,760 --> 00:33:18,720
Speaker 1: the same NERD lawyer piece and I quote again, and

598
00:33:18,760 --> 00:33:22,080
Speaker 1: they're not alone. The secondary market is projected to handle

599
00:33:22,080 --> 00:33:24,320
Speaker 1: one hundred and twenty two billion dollars in assets in

600
00:33:24,360 --> 00:33:27,400
Speaker 1: twenty twenty five, Yet that still represents just one point

601
00:33:27,480 --> 00:33:29,600
Speaker 1: nine percent of the total Unicorn value. As in the

602
00:33:29,680 --> 00:33:31,959
Speaker 1: side of Unicorn is a company worth a billion dollars,

603
00:33:32,360 --> 00:33:36,880
Speaker 1: there's six plus trillion dollars in untapped liquidity potential as

604
00:33:36,920 --> 00:33:39,920
Speaker 1: aside here. That's a really nice way of looking at this.

605
00:33:41,400 --> 00:33:44,480
Speaker 1: The transformation of the secondary market from emergency tool to

606
00:33:44,560 --> 00:33:48,800
Speaker 1: standard operating procedure represents the most significant structural shift in

607
00:33:48,880 --> 00:33:51,880
Speaker 1: venture capital since the rise of unicorns. It's not a

608
00:33:51,920 --> 00:33:55,880
Speaker 1: temporary fix. It's a permanent evolution driven by misaligned timeframes

609
00:33:56,000 --> 00:33:59,640
Speaker 1: between fund life cycles ten years and company maturation eleven

610
00:33:59,680 --> 00:34:03,240
Speaker 1: plus years. I read that it's not just a temporary

611
00:34:03,280 --> 00:34:05,840
Speaker 1: fix thing. I'm like, is this ai anyway back to

612
00:34:05,960 --> 00:34:08,680
Speaker 1: quoting for better or worse? This is the new reality

613
00:34:08,680 --> 00:34:11,760
Speaker 1: of startup funding. Vizs can no longer afford to simply

614
00:34:11,800 --> 00:34:14,440
Speaker 1: spray and pray and wait for exits. They need active

615
00:34:14,480 --> 00:34:18,160
Speaker 1: liquidity management strategies, and that fundamentally changes what kind of

616
00:34:18,200 --> 00:34:20,759
Speaker 1: companies are getting funded and how. I'm just going to

617
00:34:20,840 --> 00:34:24,200
Speaker 1: be honest. That last parts bullshit. I wrote this in

618
00:34:24,239 --> 00:34:26,759
Speaker 1: the newsletter. That last part is whang. This is not

619
00:34:26,880 --> 00:34:29,839
Speaker 1: changing anything about what companies get funded or how they're

620
00:34:29,840 --> 00:34:34,120
Speaker 1: still funding these unprofitable monstrosities. Nothing has changed, and I

621
00:34:34,160 --> 00:34:36,960
Speaker 1: would argue this piece frames that as a positive when

622
00:34:37,000 --> 00:34:40,000
Speaker 1: the reality is far grimmer. Venture capitalists are sitting on

623
00:34:40,040 --> 00:34:42,560
Speaker 1: piles of immovable equity and companies worth far less than

624
00:34:42,600 --> 00:34:44,840
Speaker 1: they invested in. And the answer, it appears, is to

625
00:34:44,840 --> 00:34:47,520
Speaker 1: find somebody else who is able to buy the dead weight.

626
00:34:48,000 --> 00:34:52,719
Speaker 1: I assume because they are stupid, Like I mean, I

627
00:34:52,760 --> 00:34:55,560
Speaker 1: realize someone is going to say, well, actually, it means

628
00:34:55,600 --> 00:34:57,319
Speaker 1: maybe they see maybe they can afford to hold it

629
00:34:57,360 --> 00:35:00,760
Speaker 1: for longer. No. Now, if you're buying anything since twenty

630
00:35:00,760 --> 00:35:04,879
Speaker 1: twenty one, you're probably get swindled. And according to Newcomer Only,

631
00:35:04,920 --> 00:35:08,120
Speaker 1: oney one hundred and seventeen venture funds closed in twenty

632
00:35:08,160 --> 00:35:11,080
Speaker 1: twenty five, down from twenty one hundred and twenty twenty four,

633
00:35:11,280 --> 00:35:13,640
Speaker 1: and forty three percent of dollars raised went to the

634
00:35:13,719 --> 00:35:16,360
Speaker 1: largest venture funds, per The New York Times and Pitchbook,

635
00:35:16,600 --> 00:35:20,080
Speaker 1: suggesting limited partners are becoming less interested in pumping cash

636
00:35:20,120 --> 00:35:22,560
Speaker 1: into the system at a time when AI startups are

637
00:35:22,600 --> 00:35:26,239
Speaker 1: demanding more capital than has ever been raised. How long

638
00:35:26,239 --> 00:35:29,160
Speaker 1: can the venture capital industry keep handing out one hundred

639
00:35:29,200 --> 00:35:32,120
Speaker 1: million to five hundred million dollars to multiple startups a year,

640
00:35:32,520 --> 00:35:34,680
Speaker 1: because all the signs suggest that the current pace of

641
00:35:34,680 --> 00:35:37,839
Speaker 1: funding must continue in perpetuity as nobody appears to have

642
00:35:37,840 --> 00:35:41,160
Speaker 1: worked out the generative AI is inherently unprofitable, and thus

643
00:35:41,239 --> 00:35:44,440
Speaker 1: every single company is on the Silicon Valley welfare system

644
00:35:44,480 --> 00:35:47,680
Speaker 1: until somebody or everybody gives up or the system itself

645
00:35:47,680 --> 00:35:51,560
Speaker 1: cannot sustain the pressure. I've read too many people make

646
00:35:51,600 --> 00:35:54,760
Speaker 1: offhanded comments about this being like the dot com boom,

647
00:35:54,760 --> 00:35:57,359
Speaker 1: and saying that lots of startups might die, but what's

648
00:35:57,440 --> 00:35:59,680
Speaker 1: left over will be good. And I hate them. I

649
00:35:59,719 --> 00:36:02,880
Speaker 1: hate so much. I hate them both for their flippancy

650
00:36:03,200 --> 00:36:06,440
Speaker 1: and for their ignorance. None of the current stack of

651
00:36:06,480 --> 00:36:09,080
Speaker 1: AI companies can survive on their own, meaning that the

652
00:36:09,160 --> 00:36:12,160
Speaker 1: venture capital industry is holding them up. If even one

653
00:36:12,200 --> 00:36:15,960
Speaker 1: of these companies faulters and dies, the entire narrative dies

654
00:36:16,000 --> 00:36:18,440
Speaker 1: with it. If that happens, it will be harder for

655
00:36:18,560 --> 00:36:21,560
Speaker 1: other AI companies to raise and even harder to sell

656
00:36:21,600 --> 00:36:24,040
Speaker 1: an AI company to somebody else. And if you can,

657
00:36:24,320 --> 00:36:27,799
Speaker 1: you'll be selling it for less. This is a punishment

658
00:36:27,880 --> 00:36:31,080
Speaker 1: for a decade plus of hubris, where companies were invested

659
00:36:31,080 --> 00:36:34,960
Speaker 1: in without ever considering a path to profitability. Venture capital

660
00:36:35,000 --> 00:36:37,640
Speaker 1: has made the same mistake again and again and again,

661
00:36:38,000 --> 00:36:41,239
Speaker 1: believing that because Uber or Facebook or Airbnb or any

662
00:36:41,320 --> 00:36:43,640
Speaker 1: number of other companies found in nearly twenty years ago

663
00:36:43,680 --> 00:36:47,239
Speaker 1: were unprofitable, at some point with past the profitability, I

664
00:36:47,320 --> 00:36:51,000
Speaker 1: might add, it was totally okay to keep pumping, pumpaning, pumping,

665
00:36:51,040 --> 00:36:53,440
Speaker 1: I'm just saying it, pumping up companies that have had

666
00:36:53,560 --> 00:36:57,000
Speaker 1: no path to profitability, which eventually became had no apparent

667
00:36:57,040 --> 00:37:00,520
Speaker 1: business models see the metaversal Web three, which eventually have

668
00:37:00,640 --> 00:37:03,920
Speaker 1: negative margins so severe, evaluation so high that we will

669
00:37:03,960 --> 00:37:06,480
Speaker 1: need an IPO at the a market cap higher than

670
00:37:06,520 --> 00:37:11,320
Speaker 1: Netflix at minimum. This is Silicon Valley's rot economy, the

671
00:37:11,440 --> 00:37:14,360
Speaker 1: desperate growth at all costs attachment the startups where you

672
00:37:14,440 --> 00:37:16,920
Speaker 1: and I quote really liked the founder, where and I

673
00:37:17,120 --> 00:37:19,239
Speaker 1: quote again the market could be huge who knows if

674
00:37:19,239 --> 00:37:21,480
Speaker 1: it is, and where you just don't need to worry

675
00:37:21,520 --> 00:37:26,279
Speaker 1: about profitability because IPOs and exits were easy. Yeah, you see,

676
00:37:26,360 --> 00:37:29,120
Speaker 1: venture capital used to be real easy because we were

677
00:37:29,120 --> 00:37:31,120
Speaker 1: still in an era of hypergrowth. You could be a

678
00:37:31,200 --> 00:37:33,279
Speaker 1: stupid asshole who doesn't know anything, but there were so

679
00:37:33,280 --> 00:37:35,080
Speaker 1: many good deals, and the more well known you were

680
00:37:35,200 --> 00:37:38,400
Speaker 1: the more likely be brought them first, guaranteeing a bigger payout,

681
00:37:38,600 --> 00:37:42,040
Speaker 1: guaranteeing more LP capital, guaranteeing more opportunities that were of

682
00:37:42,040 --> 00:37:44,560
Speaker 1: a higher quality. Because you were a big name. It

683
00:37:44,600 --> 00:37:47,160
Speaker 1: wasn't about being smart and knowing the fundamentals. It was

684
00:37:47,200 --> 00:37:49,719
Speaker 1: about being handed things. It was easier to make a

685
00:37:49,800 --> 00:37:52,440
Speaker 1: valuable company too, easier to get funded, and easier to

686
00:37:52,440 --> 00:37:54,520
Speaker 1: sell because the goal was always get funded, grow as

687
00:37:54,600 --> 00:37:58,400
Speaker 1: large as possible, or well go public. And here's the

688
00:37:58,440 --> 00:38:01,400
Speaker 1: thing about that as well. They were just more ideas.

689
00:38:01,600 --> 00:38:03,680
Speaker 1: They want more ideas to do. Like I said with

690
00:38:03,719 --> 00:38:06,359
Speaker 1: the rock com bubble, they were just there were more

691
00:38:06,360 --> 00:38:11,040
Speaker 1: things that people could create. That's the thing that ideas

692
00:38:11,040 --> 00:38:15,279
Speaker 1: are a finite source. And when you incentivize this kind

693
00:38:15,360 --> 00:38:19,879
Speaker 1: of thinking, you eventually stop incentivizing good ideas. You incentivize

694
00:38:19,920 --> 00:38:23,279
Speaker 1: the growth. And as a result, venture capital encouraged growth

695
00:38:23,280 --> 00:38:26,799
Speaker 1: at all cost, thinking above everything else. In twenty ten,

696
00:38:26,880 --> 00:38:29,319
Speaker 1: ben Horowitz said and I quote that the only thing

697
00:38:29,400 --> 00:38:32,480
Speaker 1: worse for an entrepreneur than startup hell, which was bankruptcy,

698
00:38:32,719 --> 00:38:36,800
Speaker 1: is startup purgatory. And I quote when you don't go bankrupt,

699
00:38:36,880 --> 00:38:38,520
Speaker 1: but you fail to build the Number one product in

700
00:38:38,560 --> 00:38:42,400
Speaker 1: the space that's purgatory. You have enough money with your

701
00:38:42,400 --> 00:38:44,759
Speaker 1: conservative burn rate to last for many years. You may

702
00:38:44,800 --> 00:38:47,640
Speaker 1: even be cash flow positive. However, you have zero chance

703
00:38:47,680 --> 00:38:50,000
Speaker 1: to becoming a high growth company. You have zero chance

704
00:38:50,040 --> 00:38:53,480
Speaker 1: of being anything but a very small technology business. From

705
00:38:53,480 --> 00:38:55,759
Speaker 1: the entrepreneur's point of view, this can be worse than

706
00:38:55,800 --> 00:38:59,160
Speaker 1: start up hell, since you're stuck with the small company.

707
00:39:00,080 --> 00:39:05,120
Speaker 1: What a fucking, noxious, ugly, horrifying thing to say. Let

708
00:39:05,120 --> 00:39:06,680
Speaker 1: me put it like this. If you run a good

709
00:39:06,680 --> 00:39:08,920
Speaker 1: business that people like and it's profitable, and you have

710
00:39:08,960 --> 00:39:12,399
Speaker 1: good your employees like it, your customers like it, that's

711
00:39:12,440 --> 00:39:14,279
Speaker 1: a good company. That isn't hell. And if you think

712
00:39:14,320 --> 00:39:17,440
Speaker 1: that way, you're a fucking psychopath. You're an actual nutter.

713
00:39:18,040 --> 00:39:20,560
Speaker 1: And this poisonous theory sadly paid off for him in

714
00:39:20,600 --> 00:39:23,120
Speaker 1: that startups got used to building high growth, low margin

715
00:39:23,200 --> 00:39:25,160
Speaker 1: companies that would easily sell to other companies of the

716
00:39:25,160 --> 00:39:29,879
Speaker 1: markets themselves, right up until it didn't. Of course, per

717
00:39:29,920 --> 00:39:32,919
Speaker 1: nerd Lawyer, IPOs have collapsed as next sit route along

718
00:39:32,960 --> 00:39:35,600
Speaker 1: with easy to raise capital. Per pitch Book, since twenty

719
00:39:35,600 --> 00:39:38,760
Speaker 1: twenty two, seventy percent of VC backed exits were valued

720
00:39:38,760 --> 00:39:40,799
Speaker 1: at less than the capital put in, with more than

721
00:39:40,800 --> 00:39:43,360
Speaker 1: a third of them being startups buying other startups. In

722
00:39:43,400 --> 00:39:46,000
Speaker 1: twenty twenty four, the money is drying up as the

723
00:39:46,080 --> 00:39:48,839
Speaker 1: value of VC's assets is decreasing at a time when

724
00:39:48,920 --> 00:39:51,360
Speaker 1: vcs need more money than ever because everybody is heavily

725
00:39:51,400 --> 00:39:54,400
Speaker 1: leveraged in the single most expensive funding climate in history,

726
00:39:55,120 --> 00:39:57,839
Speaker 1: and as we hit this historical liquidity crisis, the two

727
00:39:57,920 --> 00:40:01,120
Speaker 1: largest companies, open aron Anthropic are becoming drains on the

728
00:40:01,160 --> 00:40:04,160
Speaker 1: system that in a very real sense, are participating in

729
00:40:04,160 --> 00:40:07,480
Speaker 1: a massive redistribution of capital reserved for startups to one

730
00:40:07,520 --> 00:40:10,279
Speaker 1: of a few public companies. No really hear me out.

731
00:40:11,080 --> 00:40:12,920
Speaker 1: Open Ai is trying to raise as much as one

732
00:40:12,960 --> 00:40:15,680
Speaker 1: hundred billion dollars in funding so it can continue to

733
00:40:15,719 --> 00:40:18,320
Speaker 1: pass money to one of a few public companies, thirty

734
00:40:18,320 --> 00:40:21,280
Speaker 1: eight billion dollars to Amazon Web Services over seven years,

735
00:40:21,440 --> 00:40:23,919
Speaker 1: twenty two point four billion dollars to Core We've over

736
00:40:24,000 --> 00:40:26,440
Speaker 1: five years, and two hundred and fifty billion dollars over

737
00:40:26,480 --> 00:40:30,720
Speaker 1: an indeterminate period on Microsoft Azure. If successful, open AI's

738
00:40:30,800 --> 00:40:32,839
Speaker 1: venture Telethon will raise more money than it has ever

739
00:40:32,880 --> 00:40:35,480
Speaker 1: been raised in a single round, draining funds that actual

740
00:40:35,520 --> 00:40:38,919
Speaker 1: startups need. Anthropic has agreed, as I mentioned, to over

741
00:40:39,000 --> 00:40:42,480
Speaker 1: seventy billion dollars in compute and chips deal across Google, Amazon,

742
00:40:42,480 --> 00:40:45,600
Speaker 1: and Broadcom. And that's not including this hut eight compute

743
00:40:45,600 --> 00:40:47,719
Speaker 1: deal at Google is backing that I don't even really

744
00:40:47,760 --> 00:40:52,200
Speaker 1: want to think about. Think about it, just pre or

745
00:40:52,280 --> 00:40:56,560
Speaker 1: simple for a second. Instead of venture capital going to startups,

746
00:40:56,800 --> 00:41:00,000
Speaker 1: you know, early stage companies taking a risk that much

747
00:41:00,320 --> 00:41:02,600
Speaker 1: is going to Open AI and Anthropic so they can

748
00:41:02,600 --> 00:41:05,680
Speaker 1: hand it to big tech. This is big tech stealing

749
00:41:05,680 --> 00:41:08,080
Speaker 1: from Silicon Valley and to see it as anything else

750
00:41:08,320 --> 00:41:11,800
Speaker 1: is naive. And this money will come from what remains

751
00:41:11,840 --> 00:41:15,439
Speaker 1: of venture capital, private equity, and whatever hyperscalers will hand

752
00:41:15,480 --> 00:41:18,760
Speaker 1: over yet elsewhere, even the money that goes to regular

753
00:41:18,800 --> 00:41:22,440
Speaker 1: startups is ultimately being sent to those hyperscalers. That AI

754
00:41:22,560 --> 00:41:24,919
Speaker 1: startup that needs to keep raising one hundred billion dollars

755
00:41:24,960 --> 00:41:27,600
Speaker 1: in a single round isn't sending that cash to other startups.

756
00:41:27,680 --> 00:41:31,960
Speaker 1: It's mostly going to Open Ai, who sends it to Microsoft, Amazon, Core, Evil, Google,

757
00:41:32,400 --> 00:41:35,040
Speaker 1: Anthropic who sends it to Google, Microsoft or Amazon or

758
00:41:35,040 --> 00:41:38,520
Speaker 1: one of the large hyperscalers such as Amazon, Microsoft or Google.

759
00:41:39,840 --> 00:41:42,640
Speaker 1: Silicon Valley didn't birth the next big tech firm. It

760
00:41:42,719 --> 00:41:47,080
Speaker 1: incubated yet another hyperscalar level parasite, except instead of just

761
00:41:47,160 --> 00:41:50,319
Speaker 1: spending money on hyperscalar services and raising money to do so,

762
00:41:50,560 --> 00:41:53,600
Speaker 1: both Anthropic and open ai actively drain the venture capital

763
00:41:53,680 --> 00:41:56,920
Speaker 1: system as well as they both burn billions of dollars

764
00:41:57,080 --> 00:41:59,760
Speaker 1: and need those billions of dollars to keep running the models.

765
00:42:00,080 --> 00:42:03,879
Speaker 1: The AI startups can pay them by creating something that's

766
00:42:03,920 --> 00:42:07,640
Speaker 1: incredibly expensive to run, so destructively expensive to run, they

767
00:42:07,640 --> 00:42:11,120
Speaker 1: can naturally create startups more dependent on the venture capital system,

768
00:42:11,200 --> 00:42:13,440
Speaker 1: and the venture capital system has no idea what to

769
00:42:13,480 --> 00:42:17,120
Speaker 1: do other than say, just grow baby. Both open ai

770
00:42:17,200 --> 00:42:19,279
Speaker 1: and anthropics models might be getting cheaper on a per

771
00:42:19,360 --> 00:42:23,120
Speaker 1: million token basis, but they use more tokens, which increases

772
00:42:23,160 --> 00:42:25,680
Speaker 1: the cost of inference, which in turn increases the cost

773
00:42:25,760 --> 00:42:28,719
Speaker 1: of startups doing business, which in turn means Open Ai,

774
00:42:28,840 --> 00:42:32,080
Speaker 1: Anthropic and all connected startups lose more money, which increases

775
00:42:32,120 --> 00:42:35,800
Speaker 1: the burn on venture capital. This is a doom spiral,

776
00:42:35,960 --> 00:42:37,920
Speaker 1: one that can only be reversed through the most magical

777
00:42:37,960 --> 00:42:40,640
Speaker 1: and aggressive turnaround we will see in history, and it

778
00:42:40,680 --> 00:42:42,680
Speaker 1: will have to happen in the next year without fail,

779
00:42:44,640 --> 00:42:47,279
Speaker 1: and it won't. It's not going to happen and to

780
00:42:47,280 --> 00:42:50,000
Speaker 1: find out why. You just have to join me tomorrow

781
00:42:50,120 --> 00:42:53,839
Speaker 1: for part three of the in shitter financial Crisis. I'm

782
00:42:53,880 --> 00:43:06,000
Speaker 1: Better Offline. This is ed z trn fuck. Thank you

783
00:43:06,080 --> 00:43:07,440
Speaker 1: for listening to Better Offline.

784
00:43:07,560 --> 00:43:10,000
Speaker 2: The editor and composer of the Better Offline theme song

785
00:43:10,080 --> 00:43:12,719
Speaker 2: is Matasowski. You can check out more of his music

786
00:43:12,760 --> 00:43:16,399
Speaker 2: and audio projects at Matasowski dot com, M A T

787
00:43:16,400 --> 00:43:20,880
Speaker 2: T O S O W s ki dot com. You

788
00:43:20,880 --> 00:43:23,400
Speaker 2: can email me at easy at Better offline dot com

789
00:43:23,480 --> 00:43:25,800
Speaker 2: or visit Better Offline dot com to find more podcast

790
00:43:25,840 --> 00:43:29,160
Speaker 2: links and of course, my newsletter. I also really recommend

791
00:43:29,200 --> 00:43:31,120
Speaker 2: you go to chat dot Where's youreaed dot at to

792
00:43:31,239 --> 00:43:33,680
Speaker 2: visit the discord, and go to our slash Better.

793
00:43:33,560 --> 00:43:36,719
Speaker 1: Offline to check out our reddit. Thank you so much

794
00:43:36,760 --> 00:43:37,280
Speaker 1: for listening.

795
00:43:38,120 --> 00:43:40,799
Speaker 3: Better Offline is a production of cool Zone Media. For

796
00:43:40,920 --> 00:43:44,080
Speaker 3: more from cool Zone Media, visit our website cool Zonemedia

797
00:43:44,160 --> 00:43:47,000
Speaker 3: dot com, or check us out on the iHeartRadio app,

798
00:43:47,040 --> 00:44:02,680
Speaker 3: Apple Podcasts, or wherever you get your podcasts, film

799
00:44:05,840 --> 00:44:06,000
Speaker 1: Scho