1 00:00:02,730 --> 00:00:05,590 Speaker 1: Bloomberg Audio Studios. Podcasts. 2 00:00:05,890 --> 00:00:07,270 Speaker 2: Radio. News. 3 00:00:18,630 --> 00:00:22,530 Speaker 3: Hello and welcome to another episode of the Odd Lots podcast. 4 00:00:22,670 --> 00:00:23,750 Speaker 3: I'm Joe Weisenthal. 5 00:00:23,989 --> 00:00:24,989 Speaker 1: And I'm Tracy Allaway. 6 00:00:25,090 --> 00:00:29,870 Speaker 3: Tracy, we're recording this Friday, November 13th. The NVIDIA story 7 00:00:30,010 --> 00:00:31,960 Speaker 3: is so fast moving that we have to sort of 8 00:00:32,000 --> 00:00:34,960 Speaker 3: be careful with the episodes we record because we don't 9 00:00:34,979 --> 00:00:36,860 Speaker 3: know what's going to happen by the time it comes out. 10 00:00:37,300 --> 00:00:39,750 Speaker 3: But one thing we at least do know is that 11 00:00:40,070 --> 00:00:44,250 Speaker 3: NVIDIA is scheduled to report earnings next week. And at 12 00:00:44,330 --> 00:00:47,900 Speaker 3: least that is the next sort of. thing that we know. 13 00:00:49,000 --> 00:00:53,630 Speaker 3: It's the one known unknown, so to speak, that we 14 00:00:53,670 --> 00:00:55,570 Speaker 3: can look forward to, right? It's the one thing we 15 00:00:55,610 --> 00:00:57,130 Speaker 3: do know is that there's something on the calendar. 16 00:00:57,350 --> 00:00:57,910 Speaker 1: That's right. 17 00:00:58,010 --> 00:00:59,890 Speaker 4: I feel like this is another one of those episodes 18 00:00:59,910 --> 00:01:02,590 Speaker 4: where we have to say exactly when we're recording. So 19 00:01:02,610 --> 00:01:06,500 Speaker 4: this is 1210 New York time. The stock is down 20 00:01:06,560 --> 00:01:10,280 Speaker 4: about 4% as we're recording. And obviously, you know, in 21 00:01:11,160 --> 00:01:14,880 Speaker 4: corporate analyst space, we hear people talk about key man 22 00:01:15,080 --> 00:01:17,959 Speaker 4: risk all the time, which is a hard thing to 23 00:01:18,060 --> 00:01:20,500 Speaker 4: actually measure and think about. But I think in the 24 00:01:20,520 --> 00:01:23,380 Speaker 4: case of Jensen, this is kind of the ultimate key 25 00:01:23,400 --> 00:01:24,360 Speaker 4: man risk, right? 26 00:01:24,400 --> 00:01:26,260 Speaker 1: This is the guy that built the company. 27 00:01:27,020 --> 00:01:30,280 Speaker 4: We know from various reports that he was intensely involved 28 00:01:30,330 --> 00:01:34,450 Speaker 4: in the company itself, something like 60 direct reports, no 29 00:01:34,550 --> 00:01:36,149 Speaker 4: succession plan, really. 30 00:01:36,590 --> 00:01:37,530 Speaker 1: It seems like. 31 00:01:38,280 --> 00:01:41,880 Speaker 4: He was very involved in individual, like, financing deals and 32 00:01:41,920 --> 00:01:44,680 Speaker 4: things like that. And so you can imagine if you're 33 00:01:44,700 --> 00:01:47,080 Speaker 4: an analyst covering NVIDIA stock right now, it must be 34 00:01:47,120 --> 00:01:47,860 Speaker 4: pretty challenging. 35 00:01:48,000 --> 00:01:48,720 Speaker 2: No, totally. 36 00:01:48,780 --> 00:01:51,490 Speaker 3: I've met a few people in my life who have 37 00:01:51,550 --> 00:01:54,610 Speaker 3: talked about, like, their companies. They're like, my company got 38 00:01:54,630 --> 00:01:56,950 Speaker 3: some investment from NVIDIA. I got to meet Jensen Wang. 39 00:01:56,970 --> 00:01:59,030 Speaker 3: I was like, oh, that's cool. But now I go 40 00:01:59,070 --> 00:02:03,530 Speaker 3: back and listen and remember hearing these stories, and it's like, wow. Wow, 41 00:02:03,710 --> 00:02:06,630 Speaker 3: he was super involved. Like that same story is like 42 00:02:06,970 --> 00:02:09,690 Speaker 3: on one. It's sort of like that meme is like, oh, 43 00:02:10,130 --> 00:02:12,230 Speaker 3: that's great. We have a super involved CEO. And then 44 00:02:12,290 --> 00:02:13,510 Speaker 3: it's like, oh, we have. 45 00:02:13,470 --> 00:02:16,560 Speaker 1: A super involved CEO. Is this good or bad? 46 00:02:16,660 --> 00:02:19,040 Speaker 3: Anyway, so if you're thinking if just thinking about it 47 00:02:19,080 --> 00:02:21,620 Speaker 3: from a stock perspective, you have the key man risk. 48 00:02:22,100 --> 00:02:24,600 Speaker 3: You have all of the legal questions that the company 49 00:02:24,639 --> 00:02:27,500 Speaker 3: is going to face and how, you know, just everything 50 00:02:27,560 --> 00:02:30,600 Speaker 3: related to the sort of mega speed and what it 51 00:02:30,650 --> 00:02:33,270 Speaker 3: knew when and all that stuff. And then there's just 52 00:02:33,310 --> 00:02:37,110 Speaker 3: the fundamentals of the business and there's chip demand, but 53 00:02:37,210 --> 00:02:41,010 Speaker 3: also its role in financing a big ecosystem of companies 54 00:02:41,360 --> 00:02:43,260 Speaker 3: that either use and buy video chips. 55 00:02:43,860 --> 00:02:45,520 Speaker 1: There's a lot to talk about, obviously. 56 00:02:45,639 --> 00:02:45,859 Speaker 2: All right. 57 00:02:46,120 --> 00:02:48,860 Speaker 3: Let's not waste more time. We really do have the 58 00:02:48,900 --> 00:02:51,700 Speaker 3: perfect guest, someone we haven't had him on in a while, 59 00:02:51,820 --> 00:02:54,840 Speaker 3: but he's who we talk to frequently when we want 60 00:02:54,860 --> 00:02:58,239 Speaker 3: to talk about NVIDIA specifically. We're going to be speaking 61 00:02:58,260 --> 00:03:01,370 Speaker 3: with Stacey Rasgan. He's a managing director and senior analyst 62 00:03:01,470 --> 00:03:06,810 Speaker 3: at Bernstein Research. He covers this company, covers semiconductor, semiconductor equipment, 63 00:03:07,190 --> 00:03:09,870 Speaker 3: knows the company very well. If you listen to earnings 64 00:03:09,910 --> 00:03:13,630 Speaker 3: conference calls, you've probably heard his voice before. So, Stacey, 65 00:03:14,030 --> 00:03:16,150 Speaker 3: thank you so much for coming back on OddLots. It's 66 00:03:16,169 --> 00:03:17,190 Speaker 3: been too long. We've missed you. 67 00:03:17,530 --> 00:03:19,669 Speaker 2: It has been. And it's my pleasure, my great pleasure 68 00:03:19,690 --> 00:03:21,929 Speaker 2: to be here. So thanks for having me on again. Absolutely. 69 00:03:22,389 --> 00:03:24,770 Speaker 3: How weird is this a moment as a stock analyst, 70 00:03:24,830 --> 00:03:27,850 Speaker 3: just figuring it out, just professionally, like your job is 71 00:03:27,889 --> 00:03:31,370 Speaker 3: not to figure it out. How many dimensions and variables 72 00:03:31,830 --> 00:03:32,730 Speaker 3: are you trying to work through? 73 00:03:33,130 --> 00:03:35,860 Speaker 2: This is a very weird moment as an analyst covering 74 00:03:35,880 --> 00:03:38,580 Speaker 2: this company and a sad moment. Let's be honest. I mean, 75 00:03:38,680 --> 00:03:42,700 Speaker 2: there's no good solutions or outcomes or anything else coming 76 00:03:42,840 --> 00:03:46,340 Speaker 2: out of this entire mess. Like it's just a disaster 77 00:03:46,380 --> 00:03:49,400 Speaker 2: of the highest proportions, frankly. And it's been sort of 78 00:03:49,440 --> 00:03:53,450 Speaker 2: surreal covering this as an analyst. It's not been fun. Clearly, 79 00:03:53,470 --> 00:03:55,630 Speaker 2: you know, Jensen's having a worse time than we are, 80 00:03:56,890 --> 00:03:59,410 Speaker 2: but it's been a long week for sure. 81 00:03:59,800 --> 00:04:04,130 Speaker 4: So you actually decided to downgrade NVIDIA earlier this week. 82 00:04:04,410 --> 00:04:07,130 Speaker 4: What was the catalyst for taking that move? Because that's 83 00:04:07,150 --> 00:04:09,750 Speaker 4: a pretty big call when the headlines are still flying, 84 00:04:09,770 --> 00:04:10,089 Speaker 4: you know. 85 00:04:10,250 --> 00:04:13,130 Speaker 2: Let me step back, actually. So the news flow around 86 00:04:13,250 --> 00:04:15,090 Speaker 2: this started, I mean, it had been coming up in 87 00:04:15,130 --> 00:04:17,480 Speaker 2: dribs and drabs, and it sort of crystallized on Friday 88 00:04:17,600 --> 00:04:22,159 Speaker 2: evening when the article started coming out that he had 89 00:04:22,200 --> 00:04:26,650 Speaker 2: had a relationship with this individual, Alice Wayne. It was 90 00:04:26,690 --> 00:04:29,490 Speaker 2: coming out that the US intelligence community was concerned that 91 00:04:29,529 --> 00:04:32,890 Speaker 2: she might be an asset for China. By the way, 92 00:04:32,910 --> 00:04:35,910 Speaker 2: I will say I was not necessarily convinced at that point. 93 00:04:35,950 --> 00:04:39,600 Speaker 2: I'm still not necessarily convinced that she was necessarily a spy. 94 00:04:40,310 --> 00:04:42,040 Speaker 2: May just be like a business woman who likes to 95 00:04:42,060 --> 00:04:44,520 Speaker 2: be close to power. Like, we'll see. But that was 96 00:04:44,540 --> 00:04:46,760 Speaker 2: the news that started coming out on Friday. We published 97 00:04:46,800 --> 00:04:50,960 Speaker 2: something on Sunday evening, sort of outlining those kinds of risks. 98 00:04:50,980 --> 00:04:53,690 Speaker 2: And at that point, My view was that he could 99 00:04:53,730 --> 00:04:57,210 Speaker 2: probably stick around. Again, I was less convinced and still 100 00:04:57,290 --> 00:05:00,990 Speaker 2: am not entirely convinced on the espionage aspect. And she 101 00:05:01,029 --> 00:05:04,140 Speaker 2: was a customer, which is kind of a gray area, 102 00:05:04,180 --> 00:05:06,080 Speaker 2: but maybe he can manage it. And to be clear, 103 00:05:06,100 --> 00:05:08,740 Speaker 2: I'm not making any judgments on any of his personal activities. 104 00:05:08,920 --> 00:05:12,380 Speaker 2: That's not my job to judge that. But my view 105 00:05:12,420 --> 00:05:15,000 Speaker 2: at the time on Sunday evening was nobody really wanted 106 00:05:15,020 --> 00:05:16,680 Speaker 2: to see him leave. That was always the risk. And 107 00:05:17,100 --> 00:05:18,690 Speaker 2: we sort of outlined in that piece, the key man 108 00:05:18,750 --> 00:05:21,490 Speaker 2: risk here was clearly the biggest piece of this. And 109 00:05:21,589 --> 00:05:23,690 Speaker 2: I assume we'll talk a little more in detail on that. 110 00:05:23,770 --> 00:05:26,750 Speaker 2: But my view on Sunday was that, look, shareholders didn't 111 00:05:26,770 --> 00:05:28,190 Speaker 2: want to see him on the board at that point, 112 00:05:28,230 --> 00:05:30,659 Speaker 2: didn't want to see him gone. And it wasn't anybody's 113 00:05:30,700 --> 00:05:32,100 Speaker 2: best interest to see him gone. So I thought he 114 00:05:32,140 --> 00:05:34,980 Speaker 2: could hang around. The catalyst for me was the secondary 115 00:05:35,020 --> 00:05:39,860 Speaker 2: disclosures about the relationships that he apparently had with his subordinates. 116 00:05:40,670 --> 00:05:43,989 Speaker 2: And unfortunately, that's just kind of the kiss of death. 117 00:05:44,029 --> 00:05:46,729 Speaker 2: Like you just You can't sleep with your subordinates and 118 00:05:46,750 --> 00:05:49,590 Speaker 2: then lie to the board about it. I've had other 119 00:05:49,670 --> 00:05:52,980 Speaker 2: CEOs in my coverage over the years fired for exactly 120 00:05:53,010 --> 00:05:55,760 Speaker 2: that and frankly, even for things that were less egregious 121 00:05:56,480 --> 00:05:59,000 Speaker 2: along those lines than what we saw here. And so 122 00:05:59,020 --> 00:06:00,740 Speaker 2: when that came out, we sort of had to make 123 00:06:00,779 --> 00:06:04,720 Speaker 2: a decision. Do we think that it's much more likely 124 00:06:04,760 --> 00:06:06,900 Speaker 2: now that he gets dismissed? And my conclusion at that 125 00:06:06,940 --> 00:06:09,469 Speaker 2: point was yes. And the next day we actually got 126 00:06:09,510 --> 00:06:12,270 Speaker 2: confirmation that they did in fact dismiss him. And so 127 00:06:12,290 --> 00:06:13,740 Speaker 2: then I had to decide, like, what do you do 128 00:06:13,750 --> 00:06:15,880 Speaker 2: with the stock? And this is where the discussion is. 129 00:06:15,960 --> 00:06:19,000 Speaker 2: And I was joking that we spent more time with 130 00:06:19,040 --> 00:06:21,840 Speaker 2: me and my team discussing what to do versus writing 131 00:06:21,880 --> 00:06:24,860 Speaker 2: the actual downgrade. And to be clear, we downgraded the 132 00:06:24,900 --> 00:06:27,549 Speaker 2: stock from outperforming the market before we cut our target 133 00:06:27,570 --> 00:06:29,910 Speaker 2: price in half. And it didn't take long to write it. 134 00:06:29,910 --> 00:06:32,370 Speaker 2: There's not a whole lot of analysis to do, right? 135 00:06:32,470 --> 00:06:35,210 Speaker 2: But you have to decide what to do. And so 136 00:06:35,270 --> 00:06:37,690 Speaker 2: it was probably the most difficult downgrade, though, from that 137 00:06:37,710 --> 00:06:40,970 Speaker 2: standpoint that I've ever had to write. Right. You never 138 00:06:41,010 --> 00:06:42,410 Speaker 2: want to do it. And like I said, I love 139 00:06:42,450 --> 00:06:44,810 Speaker 2: Jensen and I love NVIDIA and I love the story. 140 00:06:44,830 --> 00:06:46,600 Speaker 2: And I've been a very big data center bull and 141 00:06:46,620 --> 00:06:49,440 Speaker 2: I still am. But you sort of have to weigh 142 00:06:49,460 --> 00:06:51,820 Speaker 2: this stuff. Like the argument to keep the call, and 143 00:06:51,839 --> 00:06:54,440 Speaker 2: I have some sympathy for it, is the stock is 144 00:06:54,480 --> 00:06:57,540 Speaker 2: incredibly cheap. It was already cheap. It's gotten cheaper, clearly. 145 00:06:57,560 --> 00:07:00,160 Speaker 2: At this point, we didn't cut our numbers. I actually 146 00:07:00,180 --> 00:07:04,400 Speaker 2: don't think near-term trajectory for their business is really at 147 00:07:04,420 --> 00:07:06,500 Speaker 2: a ton of risk. I don't think customers at this 148 00:07:06,540 --> 00:07:08,880 Speaker 2: point are buying less data center stuff from them next 149 00:07:08,920 --> 00:07:12,720 Speaker 2: year because of this. And you can argue that this valuation, 150 00:07:12,920 --> 00:07:14,900 Speaker 2: that you're pricing in a lot of bad news. So 151 00:07:14,960 --> 00:07:18,500 Speaker 2: that's one argument. On the other hand, I think you're 152 00:07:18,520 --> 00:07:21,260 Speaker 2: going to see more and more news flow and tape 153 00:07:21,280 --> 00:07:23,490 Speaker 2: bombs coming up, both from inside the company as well 154 00:07:23,520 --> 00:07:26,350 Speaker 2: as externally. It's very clear that there seem to be 155 00:07:26,410 --> 00:07:28,790 Speaker 2: elements in the US government that want them gone. In fact, 156 00:07:28,810 --> 00:07:30,710 Speaker 2: the China hawks that I think are probably the ones 157 00:07:30,730 --> 00:07:32,930 Speaker 2: that kicked this off are probably not going to get 158 00:07:32,950 --> 00:07:35,210 Speaker 2: the scenario that they were hoping for. This is not 159 00:07:35,230 --> 00:07:38,020 Speaker 2: a good scenario, even if you're a China hawk. But 160 00:07:38,040 --> 00:07:40,840 Speaker 2: you're going to see more and more calls for investigation. 161 00:07:40,860 --> 00:07:44,460 Speaker 2: This is the problem is I think it's the board 162 00:07:44,580 --> 00:07:46,320 Speaker 2: and NVIDIA will find that it was a mistake to 163 00:07:46,360 --> 00:07:49,520 Speaker 2: let him go. But unfortunately, he made it impossible for 164 00:07:49,560 --> 00:07:51,660 Speaker 2: him to stay. It may have been even a bigger 165 00:07:51,680 --> 00:07:55,310 Speaker 2: mistake to keep him, right? And given the news flow, 166 00:07:55,320 --> 00:07:57,170 Speaker 2: we're likely to find they, I think they had to 167 00:07:57,250 --> 00:08:00,590 Speaker 2: sever that tie as quickly as possible. And now anything 168 00:08:00,610 --> 00:08:02,590 Speaker 2: that comes out of these investigations, at least you can 169 00:08:02,650 --> 00:08:05,860 Speaker 2: argue it's not NVIDIA, it was Jensen. I'm waiting to 170 00:08:05,900 --> 00:08:08,380 Speaker 2: see what happens when they, I don't know, they subpoena 171 00:08:08,400 --> 00:08:11,800 Speaker 2: his text and his emails. And you dig hard enough 172 00:08:11,840 --> 00:08:15,010 Speaker 2: into anybody's history, you're going to find something. And whatever 173 00:08:15,050 --> 00:08:16,610 Speaker 2: they find, my guess is going to be spun in 174 00:08:16,650 --> 00:08:19,310 Speaker 2: the absolute worst possible light. We've already seen that with 175 00:08:19,330 --> 00:08:21,770 Speaker 2: some of the headlines now. And I just find it 176 00:08:21,890 --> 00:08:25,560 Speaker 2: very difficult to advocate to my clients to own the 177 00:08:25,600 --> 00:08:28,600 Speaker 2: stock in front of all of that. And now I 178 00:08:28,620 --> 00:08:30,440 Speaker 2: fold on top of that, and certainly around the CEO 179 00:08:30,520 --> 00:08:33,510 Speaker 2: and What kind of a perspective is any new CEO 180 00:08:33,530 --> 00:08:36,620 Speaker 2: going to have, whether it's internal or external? And they've 181 00:08:36,630 --> 00:08:40,859 Speaker 2: made lots of decisions in the supply chain. And many 182 00:08:40,880 --> 00:08:44,100 Speaker 2: of them are based on Jensen's handshake and his personal relationships. 183 00:08:44,140 --> 00:08:46,480 Speaker 2: And how do you maintain that? I don't know. So 184 00:08:46,800 --> 00:08:49,230 Speaker 2: I came down on the side. If I can't sit 185 00:08:49,260 --> 00:08:51,689 Speaker 2: in front of a client and advocate, I think absolutely 186 00:08:51,710 --> 00:08:54,210 Speaker 2: you should be buying this dip. It's hard for me 187 00:08:54,230 --> 00:08:56,550 Speaker 2: to maintain the buy on it. So that's why we 188 00:08:56,570 --> 00:08:58,890 Speaker 2: downgraded it. But it was, and we discussed this for 189 00:08:58,929 --> 00:09:01,610 Speaker 2: hours among the team before writing it. Writing it took 190 00:09:01,610 --> 00:09:05,300 Speaker 2: 20 minutes, right? It's not a long thing to have 191 00:09:05,500 --> 00:09:08,080 Speaker 2: to go through on that standpoint. What a bummer. 192 00:09:08,440 --> 00:09:11,380 Speaker 3: Let's just take a few of these questions from the 193 00:09:11,400 --> 00:09:16,550 Speaker 3: stock perspective, sort of break it down. You're confident in 194 00:09:16,830 --> 00:09:18,329 Speaker 3: the numbers. You don't think there's going to be some 195 00:09:18,390 --> 00:09:22,300 Speaker 3: big cancellation of orders. But let's just talk about how 196 00:09:22,420 --> 00:09:28,240 Speaker 3: important are sales into Asia or Southeast Asia. When we 197 00:09:28,260 --> 00:09:31,630 Speaker 3: talk about valuations, we talk about the 2027 outlook or 198 00:09:31,670 --> 00:09:36,290 Speaker 3: the backwards looking 2026. And let's say there are meaningful 199 00:09:36,630 --> 00:09:41,579 Speaker 3: curbs here, or let's say there was some blind eye 200 00:09:41,640 --> 00:09:44,580 Speaker 3: towards smuggling or any questions around this, and something has 201 00:09:44,620 --> 00:09:47,559 Speaker 3: to happen, and there are meaningful changes in how NVIDIA 202 00:09:47,900 --> 00:09:51,400 Speaker 3: can sell into, I should say, the Asian market rather 203 00:09:51,440 --> 00:09:54,590 Speaker 3: than the Chinese market per se. How important is that 204 00:09:54,650 --> 00:09:57,850 Speaker 3: to the sort of lagging numbers? And how important is 205 00:09:57,870 --> 00:10:00,880 Speaker 3: this market to the forward numbers? And how much of 206 00:10:00,940 --> 00:10:04,679 Speaker 3: that is actually at material risk, either because some change 207 00:10:04,740 --> 00:10:08,500 Speaker 3: in policy or internal controls clamping down? 208 00:10:09,150 --> 00:10:12,050 Speaker 2: Yeah, you bet. So it's a tough question to answer, unfortunately. 209 00:10:12,090 --> 00:10:15,890 Speaker 2: So the company does report revenues by geography and they 210 00:10:15,910 --> 00:10:17,740 Speaker 2: do this now. They made a switch a little while 211 00:10:17,770 --> 00:10:20,040 Speaker 2: ago to report by the location of the headquarters of 212 00:10:20,059 --> 00:10:22,960 Speaker 2: the companies. So for example, if you look prior, they 213 00:10:22,980 --> 00:10:26,320 Speaker 2: had like very high sales going into Singapore and Malaysia. 214 00:10:26,980 --> 00:10:30,280 Speaker 2: And these were like the location of the shipments. But 215 00:10:30,320 --> 00:10:33,220 Speaker 2: the companies that were doing that ordering were located elsewhere. 216 00:10:33,820 --> 00:10:36,580 Speaker 2: So they've actually made a shift. And actually many semiconductor 217 00:10:36,620 --> 00:10:40,900 Speaker 2: companies recently have made this shift. to report geographical breakdown 218 00:10:40,920 --> 00:10:42,590 Speaker 2: by the location of the headquarters of the customer. So 219 00:10:42,670 --> 00:10:50,140 Speaker 2: on that basis, like Malaysian and Singapore, basically they report US, Taiwan, China, 220 00:10:50,220 --> 00:10:53,440 Speaker 2: including Hong Kong and others. and other would be Singapore 221 00:10:53,480 --> 00:10:54,500 Speaker 2: and Malaysia and all these other ones. 222 00:10:54,800 --> 00:10:58,130 Speaker 3: So an Oracle data center in Malaysia, and there was, 223 00:10:58,150 --> 00:11:00,870 Speaker 3: I don't know if, like, that would, under the current 224 00:11:00,929 --> 00:11:03,950 Speaker 3: reporting regime, that would actually show up as a sale 225 00:11:03,970 --> 00:11:07,710 Speaker 3: to the US, which from a geopolitical standpoint is arguably 226 00:11:07,750 --> 00:11:08,430 Speaker 3: very misleading. 227 00:11:08,630 --> 00:11:11,030 Speaker 2: Well, maybe or maybe not. I mean, I could argue 228 00:11:11,070 --> 00:11:13,390 Speaker 2: that whoever is actually owning and running the data center 229 00:11:13,429 --> 00:11:15,750 Speaker 2: is actually the most important thing, because I'm assuming Oracle's 230 00:11:16,890 --> 00:11:20,290 Speaker 2: I'm assuming Larry Ellison's not sneaking into Shenzhen with the 231 00:11:20,309 --> 00:11:22,050 Speaker 2: Vera Rubin shoved up as behind, right? 232 00:11:22,350 --> 00:11:23,210 Speaker 1: There's an image. 233 00:11:24,670 --> 00:11:26,540 Speaker 2: I don't necessarily object to this. And again, they're not 234 00:11:26,559 --> 00:11:29,120 Speaker 2: the only ones that are reporting revenues like this. As 235 00:11:29,160 --> 00:11:32,040 Speaker 2: an aside, I should step back. The geographical split by 236 00:11:32,140 --> 00:11:35,739 Speaker 2: shipments in many cases can be misleading for semis. So 237 00:11:35,820 --> 00:11:39,540 Speaker 2: for example, I'll make it up. Take a company like Qualcomm. 238 00:11:40,490 --> 00:11:43,400 Speaker 2: They will show very high sales into China. Why? Why? 239 00:11:44,160 --> 00:11:46,800 Speaker 2: Because where are the smartphones assembled? They're assembled in China. 240 00:11:47,390 --> 00:11:50,250 Speaker 2: So where are they actually shipping the parts? So even 241 00:11:50,290 --> 00:11:52,990 Speaker 2: a shipment to Apple from Qualcomm that's going to an iPhone, 242 00:11:53,670 --> 00:11:56,850 Speaker 2: in many cases under a strict geographical split would show 243 00:11:56,870 --> 00:11:59,620 Speaker 2: up as Chinese revenue, even though really it's not. So 244 00:11:59,910 --> 00:12:03,359 Speaker 2: I actually think the ownership and who's running these things 245 00:12:03,440 --> 00:12:07,380 Speaker 2: is important. And so I don't object to the split. 246 00:12:07,660 --> 00:12:10,579 Speaker 2: But on the current split, the Malaysian and say basically 247 00:12:10,600 --> 00:12:12,730 Speaker 2: the other block, it's like 1% of the revenue. It's 248 00:12:12,750 --> 00:12:16,750 Speaker 2: very small. The direct sales to Chinese customers, so China, 249 00:12:16,790 --> 00:12:20,220 Speaker 2: including Hong Kong, I think last quarter was roughly 8% 250 00:12:20,220 --> 00:12:22,960 Speaker 2: of their sales. And that will include a big chunk 251 00:12:23,000 --> 00:12:25,880 Speaker 2: of their gaming and auto revenues as well. In terms 252 00:12:25,920 --> 00:12:29,059 Speaker 2: of data center stuff, they're basically shipping effectively nothing directly 253 00:12:29,120 --> 00:12:32,850 Speaker 2: into China. They actually do have some license from the 254 00:12:32,920 --> 00:12:36,570 Speaker 2: US government to ship older parts. But in fact, the 255 00:12:36,630 --> 00:12:38,910 Speaker 2: Chinese government is not really allowing them to ship. And 256 00:12:39,210 --> 00:12:41,189 Speaker 2: this gets into a whole broader discussion we can have 257 00:12:41,230 --> 00:12:44,440 Speaker 2: around China policy that I think we'll get to. But 258 00:12:44,500 --> 00:12:47,179 Speaker 2: right now, they're not really shipping much data center stuff 259 00:12:47,200 --> 00:12:49,720 Speaker 2: directly into China. And it's mostly because of China. China 260 00:12:49,740 --> 00:12:52,610 Speaker 2: won't allow them, even where NVIDIA has the permission to 261 00:12:52,640 --> 00:12:55,010 Speaker 2: do it from the U.S. So the vast majority, I'm 262 00:12:55,309 --> 00:12:57,110 Speaker 2: going to make up the numbers, but I think these 263 00:12:57,150 --> 00:13:01,550 Speaker 2: are roughly correct. U.S. customers are over 60%. And then 264 00:13:01,570 --> 00:13:03,949 Speaker 2: the remainder, which I believe is low 20s, I could 265 00:13:03,970 --> 00:13:05,869 Speaker 2: look it up, but I think it's in the low 20s, 266 00:13:05,890 --> 00:13:09,339 Speaker 2: is Taiwanese customers. So my assumption would be that if 267 00:13:09,380 --> 00:13:12,939 Speaker 2: there's like stuff that is actually getting smuggled into China, 268 00:13:13,940 --> 00:13:18,209 Speaker 2: it's probably not from the U.S. customers, right? It's probably 269 00:13:18,250 --> 00:13:20,429 Speaker 2: not from the Chinese customers. You're not selling the stuff 270 00:13:20,470 --> 00:13:22,829 Speaker 2: to them anywhere. It's not very much. And the Malaysian 271 00:13:22,870 --> 00:13:25,170 Speaker 2: and Singaporean customers is very small. So it's got to 272 00:13:25,210 --> 00:13:27,959 Speaker 2: be the Taiwanese customers. I've seen a lot of numbers 273 00:13:28,000 --> 00:13:29,920 Speaker 2: floating around. Some people would say, oh, it's like up 274 00:13:30,120 --> 00:13:32,540 Speaker 2: to 20% of their revenue. I don't know how big 275 00:13:32,580 --> 00:13:34,260 Speaker 2: it is. I doubt it's that big. But like the 276 00:13:34,300 --> 00:13:37,809 Speaker 2: Taiwanese block is... in that ballpark in terms of a 277 00:13:37,850 --> 00:13:40,250 Speaker 2: percentage of their revenue. So maybe you could argue they're 278 00:13:40,270 --> 00:13:44,410 Speaker 2: not really Taiwanese customers. They're, you know, China fronts, right? 279 00:13:44,670 --> 00:13:47,840 Speaker 2: I don't know. My guess is it's probably not 20%. 280 00:13:47,840 --> 00:13:50,179 Speaker 2: It's probably not zero. I mean, you can never stop 281 00:13:50,540 --> 00:13:54,600 Speaker 2: smuggling entirely, but it's some amount. I don't think it's 20%, 282 00:13:54,600 --> 00:13:57,420 Speaker 2: but it's some amount. There's another piece of this, though, 283 00:13:58,440 --> 00:14:00,880 Speaker 2: which is, and this is now getting some traction, I think, 284 00:14:00,900 --> 00:14:03,170 Speaker 2: with the government, is They want to crack down on 285 00:14:03,390 --> 00:14:07,890 Speaker 2: China users within China using capacity remotely. Yeah. Right. 286 00:14:07,929 --> 00:14:09,340 Speaker 1: This is the rental issue. 287 00:14:09,429 --> 00:14:11,699 Speaker 2: So right now it is legal for a company in 288 00:14:11,780 --> 00:14:15,360 Speaker 2: China to access, you know, cloud services from outside of China. 289 00:14:15,820 --> 00:14:17,459 Speaker 2: And they want to put a stop to that. I 290 00:14:17,500 --> 00:14:19,700 Speaker 2: don't know how big that is. I don't know. I 291 00:14:19,700 --> 00:14:20,000 Speaker 2: don't know. 292 00:14:36,180 --> 00:14:41,170 Speaker 4: What's your impression of Jensen's overall involvement in the company? 293 00:14:41,250 --> 00:14:43,570 Speaker 4: And again, we hear these stories that a lot of 294 00:14:43,590 --> 00:14:47,060 Speaker 4: the deals are sort of personal handshake deals between him 295 00:14:47,680 --> 00:14:48,900 Speaker 4: and customers. 296 00:14:49,380 --> 00:14:51,660 Speaker 1: What was his day-to-day like at NVIDIA? 297 00:14:52,100 --> 00:14:53,600 Speaker 2: Well, I guess we know a little more about his 298 00:14:53,630 --> 00:14:58,620 Speaker 2: day-to-day now, unfortunately, than we wanted to. He's always running around. 299 00:14:58,680 --> 00:15:01,100 Speaker 2: He's the face. I mean, his job, it's kind of, 300 00:15:01,100 --> 00:15:03,960 Speaker 2: you know, he's sort of the chief technologist and the 301 00:15:04,080 --> 00:15:06,660 Speaker 2: chief strategist. And he's also almost like the chief sales 302 00:15:06,680 --> 00:15:10,320 Speaker 2: guy and the chief storyteller, right, in some sense. And 303 00:15:10,360 --> 00:15:14,840 Speaker 2: he's kind of been an evangelist for AI. Internally, I mean, look, 304 00:15:15,000 --> 00:15:17,860 Speaker 2: as you've said, he's got a very flat organization. Most 305 00:15:17,900 --> 00:15:20,720 Speaker 2: of the key folks report to him directly. He makes 306 00:15:20,800 --> 00:15:23,500 Speaker 2: all of the key decisions. And by the way, that's 307 00:15:23,540 --> 00:15:25,640 Speaker 2: a positive sense that they can move very, very quickly. 308 00:15:25,660 --> 00:15:27,260 Speaker 2: We're seeing all these folks come out of the woodwork 309 00:15:27,320 --> 00:15:30,450 Speaker 2: now about, oh, they had some huge problem and they 310 00:15:30,490 --> 00:15:32,990 Speaker 2: were in dire straits and they emailed Jensen and Jensen 311 00:15:33,270 --> 00:15:35,430 Speaker 2: fixed the problem in seven minutes or whatever it was. 312 00:15:36,030 --> 00:15:37,970 Speaker 2: They could move at a pace that most companies could 313 00:15:38,010 --> 00:15:41,320 Speaker 2: not because he had that degree of direct control. I 314 00:15:41,380 --> 00:15:43,680 Speaker 2: think a lot of these deals out there are handshakes 315 00:15:43,720 --> 00:15:46,400 Speaker 2: or if they're not, I mean, you've got customers and 316 00:15:46,440 --> 00:15:49,940 Speaker 2: suppliers that are presumably adding capacity because they trust Jensen. 317 00:15:51,130 --> 00:15:53,630 Speaker 2: And Jensen's convinced them to take that leap of faith 318 00:15:53,970 --> 00:15:55,810 Speaker 2: and build it. And I think those are going to 319 00:15:55,830 --> 00:15:58,850 Speaker 2: be very difficult things for a new CEO to come 320 00:15:58,870 --> 00:16:01,330 Speaker 2: in and fulfill. Like, I'm not really worried about the operations. 321 00:16:01,370 --> 00:16:03,070 Speaker 2: I mean, they've got very smart people there and they've 322 00:16:03,090 --> 00:16:05,510 Speaker 2: got great technologists. And I'm not really worried about that 323 00:16:05,550 --> 00:16:08,560 Speaker 2: at this point. But like the degree of direct control 324 00:16:08,600 --> 00:16:10,859 Speaker 2: that Jensen had, it worked very, very well while he 325 00:16:10,900 --> 00:16:12,580 Speaker 2: was there. It's going to cause problems, I think, once 326 00:16:12,600 --> 00:16:13,520 Speaker 2: he's gone, for sure. 327 00:16:13,780 --> 00:16:17,520 Speaker 3: On technology, like, OK, there's who knows how smooth the 328 00:16:17,560 --> 00:16:23,190 Speaker 3: sort of operations will continue. As a technologist, over the years, 329 00:16:23,210 --> 00:16:25,950 Speaker 3: you have to make some big calls, right? And so 330 00:16:25,970 --> 00:16:28,910 Speaker 3: when you think like, okay, maybe with the AI or 331 00:16:29,010 --> 00:16:31,600 Speaker 3: the tech roadmap for the next year or two is understood, 332 00:16:31,670 --> 00:16:34,380 Speaker 3: but presumably there are going to be big calls that 333 00:16:34,420 --> 00:16:37,760 Speaker 3: have to be made in the future. And he clearly 334 00:16:37,800 --> 00:16:41,520 Speaker 3: had an amazing track record of knowing where the puck 335 00:16:41,560 --> 00:16:44,800 Speaker 3: was going and getting some big calls right. How do 336 00:16:44,820 --> 00:16:48,479 Speaker 3: you think about that sort of medium-term risk and the 337 00:16:48,620 --> 00:16:50,100 Speaker 3: capacity of nvidia. 338 00:16:49,760 --> 00:16:52,540 Speaker 2: That's exactly what i'm worried about so again i'm not 339 00:16:52,580 --> 00:16:55,070 Speaker 2: worried about so much the revenue next year maybe even 340 00:16:55,090 --> 00:16:56,470 Speaker 2: in 20 but i mean you think about the roadmap 341 00:16:56,490 --> 00:16:59,810 Speaker 2: they're shipping ruben now ruben is baked ruben ultra sounds 342 00:16:59,830 --> 00:17:02,090 Speaker 2: like it's baked like fineman i guess will be next 343 00:17:02,330 --> 00:17:05,210 Speaker 2: i don't know how fully baked that that is and 344 00:17:05,230 --> 00:17:07,510 Speaker 2: then what comes after fineman and you're about what what 345 00:17:07,530 --> 00:17:09,650 Speaker 2: are the kind of calls that what's the architecture look 346 00:17:09,690 --> 00:17:12,150 Speaker 2: like yeah What kind of memory architecture are they using? 347 00:17:12,210 --> 00:17:13,770 Speaker 2: How much HBM is going to be there? Are they 348 00:17:13,790 --> 00:17:16,080 Speaker 2: going to use something like, who knows? These are big, 349 00:17:16,160 --> 00:17:19,080 Speaker 2: major inflections that will have to be called. And to 350 00:17:19,100 --> 00:17:20,740 Speaker 2: your point, he's had a very good track record. And 351 00:17:20,780 --> 00:17:23,669 Speaker 2: so that's my issue. Like, like he's the one that 352 00:17:23,730 --> 00:17:26,450 Speaker 2: sits there and thinks out like five years, 10 years. 353 00:17:26,490 --> 00:17:30,040 Speaker 2: He's the one that's seeing around those corners. The reason 354 00:17:30,060 --> 00:17:32,520 Speaker 2: they are so dominant today is because of decisions that 355 00:17:32,560 --> 00:17:34,690 Speaker 2: he made, you know, 10 years ago. Like what? 356 00:17:34,790 --> 00:17:36,950 Speaker 3: Like what was the decision he made 10 years ago 357 00:17:37,010 --> 00:17:38,050 Speaker 3: that you would say is painful? 358 00:17:38,090 --> 00:17:39,860 Speaker 2: I mean, just for, this has been plenty of, but 359 00:17:39,910 --> 00:17:45,930 Speaker 2: I mean, investing in CUDA, They weren't in dire straits 360 00:17:45,930 --> 00:17:48,540 Speaker 2: 10 years ago. 15 years ago, they were. He was 361 00:17:48,580 --> 00:17:51,520 Speaker 2: investing a ton of money into the software ecosystem. At 362 00:17:51,560 --> 00:17:54,180 Speaker 2: the time, I don't think anybody really wanted or cared about, 363 00:17:55,280 --> 00:17:57,960 Speaker 2: but look what it's got. It's built a massive moat 364 00:17:58,040 --> 00:17:59,360 Speaker 2: and he was in the right place at the right 365 00:17:59,380 --> 00:18:03,180 Speaker 2: time when these applications came to light. Plenty of other 366 00:18:03,240 --> 00:18:05,580 Speaker 2: CEOs out there probably would not have made that call. 367 00:18:05,619 --> 00:18:09,760 Speaker 2: If people forget, The stock's been, I mean, it's been amazing, 368 00:18:09,800 --> 00:18:12,600 Speaker 2: but I mean, it was a $ 10 to $ 20 stock 369 00:18:12,660 --> 00:18:14,320 Speaker 2: like for years and years. And by the way, on 370 00:18:14,340 --> 00:18:16,100 Speaker 2: the current share count, you got to divide both of 371 00:18:16,140 --> 00:18:18,939 Speaker 2: those numbers by 40. They've done a 40 for one 372 00:18:18,980 --> 00:18:20,800 Speaker 2: stock split since then. So I mean, what is that? 373 00:18:20,820 --> 00:18:23,800 Speaker 2: I can do my math, 25 to 50 cents on 374 00:18:23,840 --> 00:18:25,680 Speaker 2: today's share count basis. It would have been for years 375 00:18:25,740 --> 00:18:28,800 Speaker 2: and years and years while he was making these kinds 376 00:18:28,840 --> 00:18:31,820 Speaker 2: of investments when people were telling him to stop and 377 00:18:32,280 --> 00:18:34,560 Speaker 2: he didn't. And so I don't know what kind of 378 00:18:34,619 --> 00:18:36,920 Speaker 2: like pucks he, or where he sees the puck going 379 00:18:36,940 --> 00:18:39,220 Speaker 2: over the next 10 years. But you're going to have 380 00:18:39,240 --> 00:18:42,460 Speaker 2: somebody else making that call now, not him. I'd rather 381 00:18:42,500 --> 00:18:44,090 Speaker 2: have him making it if I could, you know? 382 00:18:44,750 --> 00:18:47,429 Speaker 4: So setting aside the sort of key man risk and 383 00:18:47,470 --> 00:18:49,810 Speaker 4: whether NVIDIA will continue to grow at the pace it's 384 00:18:49,830 --> 00:18:53,100 Speaker 4: been able to grow under Jensen, is one of the 385 00:18:53,160 --> 00:18:58,649 Speaker 4: risks that's emerging now, or I guess accelerating just the 386 00:18:58,710 --> 00:19:02,060 Speaker 4: China hawks even more on the rise when it comes 387 00:19:02,100 --> 00:19:04,080 Speaker 4: to chip policy. And I have to say, so again, 388 00:19:04,100 --> 00:19:06,820 Speaker 4: we're recording this on Friday, November 13th. One of the 389 00:19:06,859 --> 00:19:09,320 Speaker 4: big items of breaking news this morning was we saw 390 00:19:09,440 --> 00:19:13,740 Speaker 4: Treasury Secretary Scott Besant saying that, well, this is Letnick's 391 00:19:13,760 --> 00:19:16,340 Speaker 4: fault because Megaspeed should have been on the blacklist. 392 00:19:16,400 --> 00:19:18,699 Speaker 2: Yeah. So my guess is a lot of this may 393 00:19:18,720 --> 00:19:20,980 Speaker 2: have started with, these are the conspiracy theories, but I mean, 394 00:19:21,300 --> 00:19:23,080 Speaker 2: feels like a lot of this started with the China hawks. 395 00:19:23,359 --> 00:19:25,640 Speaker 2: And again, even with Alice Wang, like she doesn't need 396 00:19:25,660 --> 00:19:27,370 Speaker 2: to be a spy. She just needs to be able 397 00:19:27,380 --> 00:19:30,330 Speaker 2: to be credibly painted as a spy. And if none 398 00:19:30,350 --> 00:19:32,050 Speaker 2: of the other stuff had come out, like let's say 399 00:19:32,090 --> 00:19:33,730 Speaker 2: he had kept his job, like that would be great. 400 00:19:33,770 --> 00:19:37,090 Speaker 2: But could he ever be like a public voice on 401 00:19:37,250 --> 00:19:41,330 Speaker 2: China AI policy ever again? Probably not. So that might've 402 00:19:41,369 --> 00:19:43,790 Speaker 2: been the original goal of all of this was to get, 403 00:19:43,810 --> 00:19:45,970 Speaker 2: because I don't, there's plenty of China. You know, he 404 00:19:45,990 --> 00:19:49,030 Speaker 2: was called for a briefing. They wanted to come and 405 00:19:49,050 --> 00:19:51,450 Speaker 2: do a briefing. It was three Republicans. that called him, right? 406 00:19:51,730 --> 00:19:53,669 Speaker 2: I mean, for that. And then now I got the 407 00:19:53,710 --> 00:19:56,190 Speaker 2: Democrats coming in who, my guess is, are not super 408 00:19:56,230 --> 00:19:58,879 Speaker 2: happy with how close he's been to Trump. And so 409 00:19:58,890 --> 00:20:00,300 Speaker 2: you got a lot of folks in the government that 410 00:20:00,320 --> 00:20:02,300 Speaker 2: may have been aligning against him. And again, I think 411 00:20:02,380 --> 00:20:04,640 Speaker 2: this outcome is probably not what they would have wanted. 412 00:20:04,660 --> 00:20:06,240 Speaker 2: I would hope that this is not what they would 413 00:20:06,260 --> 00:20:08,219 Speaker 2: have wanted. Like, this is not a great outcome, but. 414 00:20:08,820 --> 00:20:10,399 Speaker 2: But yeah, I think he would have been muted in 415 00:20:10,440 --> 00:20:13,149 Speaker 2: terms of being advocating for China policy, because I do 416 00:20:13,190 --> 00:20:16,629 Speaker 2: think China policy is going to get tougher. Now, I 417 00:20:16,650 --> 00:20:18,930 Speaker 2: got to say, though, I'm more in Jensen's camp on this. 418 00:20:18,970 --> 00:20:21,520 Speaker 2: His whole view has been, and I'll put Megaspeed aside 419 00:20:21,560 --> 00:20:23,820 Speaker 2: for a moment. We can talk about Megaspeed. But his 420 00:20:23,900 --> 00:20:28,380 Speaker 2: general China policy has been, let's allow ourselves to sell 421 00:20:28,400 --> 00:20:30,180 Speaker 2: them the older parts. So there's been all this news 422 00:20:30,320 --> 00:20:32,590 Speaker 2: about how they're selling the most advanced stuff in China. Like, no, 423 00:20:32,609 --> 00:20:35,860 Speaker 2: they're not. These are H200s and H20s. This is like 424 00:20:35,920 --> 00:20:39,220 Speaker 2: three or four generations old. Hopper showed up in 2022 425 00:20:39,220 --> 00:20:41,439 Speaker 2: is actually when it showed up. So it's several generations old. 426 00:20:41,940 --> 00:20:44,060 Speaker 2: But the idea is, and even Lutnick has said this. 427 00:20:44,119 --> 00:20:45,860 Speaker 2: I think he said we wanted to get China addicted 428 00:20:45,900 --> 00:20:47,940 Speaker 2: to our technology, right? So I wouldn't have put it 429 00:20:47,980 --> 00:20:50,660 Speaker 2: that harshly, but it's kind of true. You'd like them 430 00:20:50,720 --> 00:20:54,739 Speaker 2: to build their local AI ecosystem around our technology so 431 00:20:54,790 --> 00:20:57,110 Speaker 2: we can know what's going on and keep it under control. 432 00:20:57,730 --> 00:21:00,870 Speaker 2: And if you limit those sales, you start to encourage 433 00:21:00,910 --> 00:21:03,210 Speaker 2: them to build their own, which is exactly what they're doing. 434 00:21:04,070 --> 00:21:08,050 Speaker 2: And today, like, again, Nvidia has licenses to sell some 435 00:21:08,109 --> 00:21:11,169 Speaker 2: H200s and everything, and they're not selling them because China 436 00:21:11,210 --> 00:21:13,520 Speaker 2: won't allow them to do it because China wants to 437 00:21:13,560 --> 00:21:16,060 Speaker 2: encourage their local players. And that's why I find a 438 00:21:16,080 --> 00:21:18,540 Speaker 2: lot of this stuff, you know, there were some allegations 439 00:21:18,580 --> 00:21:21,379 Speaker 2: like earlier in the week that supposedly like Alice Wang 440 00:21:21,420 --> 00:21:23,420 Speaker 2: was trying to get him to lobby for more sales 441 00:21:23,460 --> 00:21:26,000 Speaker 2: into China. So like, number one, he's doing that anyways. 442 00:21:27,320 --> 00:21:31,010 Speaker 2: Number two, China wants her to lobby for something that 443 00:21:31,050 --> 00:21:33,350 Speaker 2: they're already not allowing NVIDIA to do. It doesn't make 444 00:21:33,369 --> 00:21:36,720 Speaker 2: any sense. So I'm on Jensen's side on the China policy. 445 00:21:36,760 --> 00:21:39,639 Speaker 2: I don't think we'd be selling Blackwells or Rubens to China, 446 00:21:39,680 --> 00:21:41,520 Speaker 2: but older stuff, I think it would have been a 447 00:21:41,520 --> 00:21:42,920 Speaker 2: good idea. I think it's too late now, though. 448 00:21:43,000 --> 00:21:45,300 Speaker 3: Look, I have no opinion on this question, and I 449 00:21:45,340 --> 00:21:50,350 Speaker 3: don't know, but justify the status quo where you can't 450 00:21:50,440 --> 00:21:54,050 Speaker 3: sell the most, or steel man, as people like to say. 451 00:21:54,070 --> 00:21:54,830 Speaker 2: Steel man. 452 00:21:54,869 --> 00:21:57,330 Speaker 3: Maybe silicon man. Maybe we should say silicon man or 453 00:21:57,369 --> 00:22:01,690 Speaker 3: chip man. Chip man the argument that we shouldn't be 454 00:22:01,770 --> 00:22:06,720 Speaker 3: selling the most advanced GPUs into China, but that it's 455 00:22:06,840 --> 00:22:10,139 Speaker 3: okay to sell them to a Singaporean company with operations 456 00:22:10,200 --> 00:22:14,920 Speaker 3: in Malaysia that allows Chinese AI companies remote access. That 457 00:22:14,940 --> 00:22:16,540 Speaker 3: strikes to a lot of people as incoherent. 458 00:22:16,780 --> 00:22:19,540 Speaker 2: No, absolutely. And look, this is a whole other issue. 459 00:22:19,580 --> 00:22:21,060 Speaker 2: And so, like I said earlier, it looks like they're 460 00:22:21,080 --> 00:22:23,439 Speaker 2: going to be cracking down on remote access. And to 461 00:22:23,480 --> 00:22:25,899 Speaker 2: be fair, I do not know how much of the 462 00:22:26,000 --> 00:22:31,109 Speaker 2: current spending on cloud compute is being used by Chinese 463 00:22:31,150 --> 00:22:34,710 Speaker 2: customers in China. I don't know. We'll see. And yeah, 464 00:22:34,730 --> 00:22:38,129 Speaker 2: I don't think that that is necessarily, you know, a 465 00:22:38,170 --> 00:22:41,510 Speaker 2: consistent point of view, unless you're only allowing folks in 466 00:22:41,550 --> 00:22:45,230 Speaker 2: China to use, you know, hoppers remotely. You know, if 467 00:22:45,250 --> 00:22:48,210 Speaker 2: you're allowing folks in China to use, to remotely access 468 00:22:48,260 --> 00:22:51,840 Speaker 2: Blackwells and Rubens in size, then yeah, that's a hole 469 00:22:51,859 --> 00:22:54,940 Speaker 2: that maybe needs to be plugged and maybe should have been, right? 470 00:22:55,240 --> 00:22:58,199 Speaker 2: But that's a different debate than should we be allowing 471 00:22:58,280 --> 00:23:01,320 Speaker 2: sales of like lower end stuff into China? Should we 472 00:23:01,340 --> 00:23:04,869 Speaker 2: be allowing access whether it's direct or remote, frankly, of 473 00:23:04,910 --> 00:23:07,830 Speaker 2: lower end stuff into China. If the concern is we 474 00:23:07,850 --> 00:23:09,790 Speaker 2: don't want to have China access to the latest and greatest, 475 00:23:09,810 --> 00:23:12,250 Speaker 2: then clearly they shouldn't have access to the latest and greatest. 476 00:23:12,690 --> 00:23:14,270 Speaker 2: But I think two slightly different issues. 477 00:23:14,730 --> 00:23:17,190 Speaker 4: So as Joe said in the beginning, the next known 478 00:23:17,470 --> 00:23:21,730 Speaker 4: unknown coming up for NVIDIA is results next week. What 479 00:23:21,750 --> 00:23:24,350 Speaker 4: are you actually looking out for there? Like if the 480 00:23:24,369 --> 00:23:26,360 Speaker 4: big risk now is, you know, it's the key man 481 00:23:26,390 --> 00:23:30,320 Speaker 4: risk and it's the future trajectory of China policy in 482 00:23:30,359 --> 00:23:32,220 Speaker 4: the U.S., like what are we actually going to learn 483 00:23:32,280 --> 00:23:32,820 Speaker 4: from earnings? 484 00:23:33,210 --> 00:23:34,990 Speaker 2: Yeah, I mean, so like, first, I would say the 485 00:23:35,050 --> 00:23:40,609 Speaker 2: numbers better be technically perfect, right? God forbid something's off 486 00:23:40,650 --> 00:23:42,070 Speaker 2: in the numbers. Now, like I said, I don't think 487 00:23:42,090 --> 00:23:45,669 Speaker 2: you're going to see any material changes to the near-term 488 00:23:45,690 --> 00:23:47,950 Speaker 2: trajectory as a result of this. I don't think so. Again, 489 00:23:48,350 --> 00:23:50,750 Speaker 2: we can talk about the longer term. You know, there's 490 00:23:50,830 --> 00:23:52,590 Speaker 2: other names that have been running a lot more, like 491 00:23:52,609 --> 00:23:54,550 Speaker 2: some of their competitors have been running a lot on 492 00:23:54,570 --> 00:23:57,129 Speaker 2: the belief that a weaker NVIDIA roadmap over the long-term 493 00:23:57,150 --> 00:23:59,990 Speaker 2: strengthens it, which may be true. We can debate that. 494 00:24:01,119 --> 00:24:02,820 Speaker 2: I would be surprised if they were changing. If they 495 00:24:02,840 --> 00:24:05,020 Speaker 2: were changing in the near term, that would be very problematic. 496 00:24:05,100 --> 00:24:08,240 Speaker 2: I mean, clearly. Look, I guess they'll have to give 497 00:24:08,359 --> 00:24:12,320 Speaker 2: some certainty on that near-term trajectory, near to medium-term trajectory, 498 00:24:12,420 --> 00:24:15,780 Speaker 2: and give investors confidence and conviction that at least that 499 00:24:15,820 --> 00:24:19,020 Speaker 2: the current trajectory is not changing. I think they're going to, 500 00:24:19,100 --> 00:24:21,920 Speaker 2: I don't know what they're going to say around, you know, Kolecka, 501 00:24:21,940 --> 00:24:24,300 Speaker 2: who was the CFO and is now the interim CEO, 502 00:24:24,980 --> 00:24:27,300 Speaker 2: is going to have to give like some confidence that 503 00:24:27,340 --> 00:24:30,580 Speaker 2: like the ship is stable and still sailing, right? They're 504 00:24:30,600 --> 00:24:32,060 Speaker 2: going to get a lot of tough questions on how 505 00:24:32,100 --> 00:24:33,820 Speaker 2: this went down. They're going to get a lot of 506 00:24:33,840 --> 00:24:37,240 Speaker 2: questions on talent retention and things like that. And, you know, 507 00:24:37,540 --> 00:24:39,639 Speaker 2: it's funny, you know, I can't remember the number, but 508 00:24:39,660 --> 00:24:41,659 Speaker 2: a huge portion of their employees have net worths that 509 00:24:41,680 --> 00:24:43,679 Speaker 2: are in the tens or even hundreds of millions of dollars. 510 00:24:44,720 --> 00:24:46,850 Speaker 2: And they presumably don't need to work. And they're there 511 00:24:47,440 --> 00:24:50,290 Speaker 2: maybe because Jensen's there. And like, what happens to that? Like, 512 00:24:50,330 --> 00:24:53,130 Speaker 2: so how do we get confidence that the talent base 513 00:24:53,170 --> 00:24:54,510 Speaker 2: is still going to be good? She's going to get 514 00:24:54,530 --> 00:24:57,350 Speaker 2: all these kinds of questions. And we'll see what she 515 00:24:57,369 --> 00:24:59,450 Speaker 2: can say. It's going to be an interesting call for sure. 516 00:24:59,630 --> 00:25:03,470 Speaker 3: The other thing that's happening next week is the Anthropic S1. 517 00:25:03,510 --> 00:25:06,689 Speaker 3: And if you read that, S1, they make a point 518 00:25:07,250 --> 00:25:10,080 Speaker 3: about the fact that they use multi-chips and their chips 519 00:25:10,119 --> 00:25:13,760 Speaker 3: are homogenous to them. And the TPUs and tranium chips 520 00:25:14,240 --> 00:25:16,879 Speaker 3: and GPUs from AMD are all just part of this 521 00:25:16,920 --> 00:25:19,960 Speaker 3: big homogenized stack. I think it's interesting they say that 522 00:25:20,260 --> 00:25:22,860 Speaker 3: is a technical thing. And we also know that Anthropic 523 00:25:22,900 --> 00:25:26,700 Speaker 3: specifically is very hawkish on chips. chip export policy. From 524 00:25:26,740 --> 00:25:30,520 Speaker 3: a competitive position, whether we're talking about TPUs, whether we're 525 00:25:30,540 --> 00:25:34,660 Speaker 3: talking about others, set aside all the scandals and everything else. 526 00:25:35,000 --> 00:25:39,200 Speaker 3: How assailable is the NVIDIA moat looking right now, even 527 00:25:39,240 --> 00:25:41,150 Speaker 3: if we weren't talking about Jensen scandals? 528 00:25:41,510 --> 00:25:43,550 Speaker 2: I'm not going to say anything about Anthropic. I'm going 529 00:25:43,570 --> 00:25:46,950 Speaker 2: to put that aside. Yeah. Very clearly to my compliance people. Yeah, totally. 530 00:25:47,010 --> 00:25:49,659 Speaker 2: Not going to say anything. But In general, the idea 531 00:25:49,700 --> 00:25:53,939 Speaker 2: of customers using multiple sources, there are a number of them. 532 00:25:54,000 --> 00:25:57,280 Speaker 2: So for example, OpenAI, they've got a big deal with NVIDIA. 533 00:25:57,320 --> 00:25:59,140 Speaker 2: They also have a pretty big deal with Broadcom. They've 534 00:25:59,180 --> 00:26:02,050 Speaker 2: even got a big deal with AMD. Same thing with Meta. 535 00:26:02,810 --> 00:26:04,850 Speaker 2: And there are others. Google, clearly, I mean, they do 536 00:26:04,869 --> 00:26:07,430 Speaker 2: their own TV. They still buy GPUs, but they do TPUs. 537 00:26:07,950 --> 00:26:10,090 Speaker 2: You've got more and more customers now that are starting, 538 00:26:10,670 --> 00:26:13,030 Speaker 2: that Google is now starting to sell TPUs to like 539 00:26:13,430 --> 00:26:16,850 Speaker 2: on a secondary basis as well. Um, so that is 540 00:26:16,890 --> 00:26:21,260 Speaker 2: happening and like my, my general view for now has 541 00:26:21,280 --> 00:26:23,760 Speaker 2: been that it's not the more, the more important question 542 00:26:23,800 --> 00:26:25,940 Speaker 2: so far has been, is the opportunity bigger? Is there not, 543 00:26:26,820 --> 00:26:28,460 Speaker 2: you know, Nvidia has had like the bulk of this 544 00:26:28,540 --> 00:26:31,379 Speaker 2: market and presumably you'll get other players coming in and 545 00:26:31,420 --> 00:26:34,379 Speaker 2: that market share will go down, but is the pie bigger? 546 00:26:34,440 --> 00:26:36,900 Speaker 2: Is it not? If it's big, there's probably room for everybody. 547 00:26:36,920 --> 00:26:38,550 Speaker 2: And if not, everybody would be screwed. 548 00:26:38,570 --> 00:26:39,090 Speaker 1: Right. 549 00:26:39,660 --> 00:26:41,950 Speaker 2: And if you look at the numbers for next year, 550 00:26:42,230 --> 00:26:45,930 Speaker 2: just for coming on this for 2027, Assuming no changes, 551 00:26:45,970 --> 00:26:50,129 Speaker 2: NVIDIA currently is guided for revenue growth of at least 70%, 552 00:26:50,130 --> 00:26:52,270 Speaker 2: which would put them in the ballpark in spitting distance 553 00:26:52,310 --> 00:26:56,889 Speaker 2: of $ 700 billion for the company. Broadcom has suggested that 554 00:26:56,930 --> 00:27:01,280 Speaker 2: their data center revenues will double year over year in 555 00:27:01,280 --> 00:27:03,820 Speaker 2: 27 to like $ 115 billion and will double again in 556 00:27:03,820 --> 00:27:08,300 Speaker 2: 2028 to $ 230 billion and again with demand outlook to 557 00:27:08,320 --> 00:27:11,890 Speaker 2: do even more if they can get there. You look 558 00:27:11,910 --> 00:27:15,710 Speaker 2: at AMD. AMD is, if you Back into it, they're 559 00:27:15,770 --> 00:27:19,689 Speaker 2: kind of suggesting that their GPU sales will probably almost 560 00:27:19,770 --> 00:27:22,330 Speaker 2: triple year over year, and you're probably into the low 561 00:27:22,330 --> 00:27:26,560 Speaker 2: $ 40 billion for AMD. Even Qualcomm is getting into this market. 562 00:27:26,580 --> 00:27:29,700 Speaker 2: They said they're going to do $ 5 billion. You've got others, 563 00:27:29,760 --> 00:27:33,100 Speaker 2: MediaTek and Marvell and all these others that sell. Everybody's 564 00:27:33,119 --> 00:27:36,800 Speaker 2: doing fine right now. But in general, are there more 565 00:27:36,859 --> 00:27:37,899 Speaker 2: alternative options? 566 00:27:37,940 --> 00:27:38,340 Speaker 4: There are. 567 00:27:38,359 --> 00:27:42,889 Speaker 2: And look, AMD, their roadmap is... It takes time, but 568 00:27:42,910 --> 00:27:45,970 Speaker 2: it's getting better. And again, you're seeing more customers using TPUs. 569 00:27:46,950 --> 00:27:49,369 Speaker 2: And guys like Broadcom have had more customers for their 570 00:27:49,390 --> 00:27:53,710 Speaker 2: custom chips. AMD themselves, everybody talks about NVIDIA CUDA as 571 00:27:53,760 --> 00:27:55,440 Speaker 2: a moat. CUDA is their software ecosystem. I think it 572 00:27:55,500 --> 00:27:58,300 Speaker 2: is a moat. But AMD has talked and others have 573 00:27:58,320 --> 00:28:01,220 Speaker 2: talked about now AI is apparently getting good at writing 574 00:28:01,240 --> 00:28:05,380 Speaker 2: software from what we've heard and using AI actually to 575 00:28:05,420 --> 00:28:08,300 Speaker 2: help to port things over like companies are experimenting with that. 576 00:28:09,090 --> 00:28:11,730 Speaker 2: And again, this gets back to Jensen. Part of the 577 00:28:11,780 --> 00:28:13,859 Speaker 2: decisions he's been making are not just on the hardware side. 578 00:28:13,880 --> 00:28:17,560 Speaker 2: They've also been on the software side. And so making 579 00:28:17,619 --> 00:28:20,200 Speaker 2: sure that that CUDA mode stays wrong. I mean, they've 580 00:28:20,240 --> 00:28:22,690 Speaker 2: got other people there that are good. You've got Ian Buck, 581 00:28:23,210 --> 00:28:27,270 Speaker 2: who's been floated as a name for potential success. People 582 00:28:27,330 --> 00:28:30,090 Speaker 2: say that he invented CUDA. He was the driving force 583 00:28:30,130 --> 00:28:33,520 Speaker 2: behind CUDA, and he runs the hyperscale business over there. 584 00:28:34,080 --> 00:28:36,180 Speaker 2: They've got good technical folks over there who can keep 585 00:28:36,220 --> 00:28:38,480 Speaker 2: this going. But again, some of those big decisions on 586 00:28:38,520 --> 00:28:40,500 Speaker 2: where even the software platform is going still have to 587 00:28:40,520 --> 00:28:44,500 Speaker 2: be made. And if I'm a customer, again, not next 588 00:28:44,540 --> 00:28:47,260 Speaker 2: year or even the year after, but longer term, do 589 00:28:47,300 --> 00:28:49,700 Speaker 2: I maybe need to think a little harder about potentially 590 00:28:49,740 --> 00:28:53,180 Speaker 2: diversifying my footprint? Maybe. And this is why some of 591 00:28:53,220 --> 00:28:57,160 Speaker 2: their competitors have been... strong in the last week or 592 00:28:57,200 --> 00:28:59,000 Speaker 2: so in the wake of all of this news flow. 593 00:28:59,280 --> 00:29:02,280 Speaker 3: I have one last question, and this is just about 594 00:29:02,600 --> 00:29:05,140 Speaker 3: what people call the central bank of AI and the 595 00:29:05,180 --> 00:29:09,520 Speaker 3: degree to which demand for NVIDIA chips is partly as 596 00:29:09,580 --> 00:29:12,440 Speaker 3: high as it is because it's vendor financed in one 597 00:29:12,480 --> 00:29:15,830 Speaker 3: way or another from NVIDIA. How much risk is that 598 00:29:15,870 --> 00:29:16,790 Speaker 3: when you think about the stock? 599 00:29:17,110 --> 00:29:19,930 Speaker 2: Yeah, I don't know. But again, they do have this 600 00:29:19,990 --> 00:29:22,670 Speaker 2: reputation of being the bank of AI. And again, I 601 00:29:22,690 --> 00:29:25,000 Speaker 2: think a lot of this is being driven by the 602 00:29:25,040 --> 00:29:27,840 Speaker 2: force of Jensen's will. Yeah. And there's already been stories coming, 603 00:29:27,860 --> 00:29:30,020 Speaker 2: which we kind of heard that like potentially Collette was 604 00:29:30,080 --> 00:29:32,719 Speaker 2: not maybe as completely on board with some of the 605 00:29:32,800 --> 00:29:35,350 Speaker 2: magnitude of what they're doing as maybe Jensen was. And 606 00:29:35,370 --> 00:29:38,030 Speaker 2: by the way, that was reported yesterday. That's been floating 607 00:29:38,070 --> 00:29:39,830 Speaker 2: around for a while. And it's not Chucky. She was 608 00:29:39,870 --> 00:29:42,630 Speaker 2: a a CFO. That's her job, right? But one has 609 00:29:42,650 --> 00:29:44,580 Speaker 2: to wonder. It'll be very interesting. They're going to get 610 00:29:44,620 --> 00:29:47,080 Speaker 2: a new permanent CEO at some point. And what will 611 00:29:47,100 --> 00:29:50,080 Speaker 2: the appetite be? I don't know. But it's a good 612 00:29:50,120 --> 00:29:52,980 Speaker 2: question to ask. And these are, again, some of those 613 00:29:53,060 --> 00:29:55,340 Speaker 2: unknowns that are out there. I don't know. I'd say 614 00:29:55,700 --> 00:29:59,390 Speaker 2: Jensen was clearly on the more bullish side. of committing 615 00:29:59,470 --> 00:30:03,090 Speaker 2: company capital for this. Not just for, you know, backstops 616 00:30:03,150 --> 00:30:05,620 Speaker 2: and things, but I mean, also for supply. People throw 617 00:30:05,650 --> 00:30:07,220 Speaker 2: in all these numbers. I mean, they spent hundreds of 618 00:30:07,260 --> 00:30:08,780 Speaker 2: billions of dollars. You have to remember, most of the 619 00:30:08,880 --> 00:30:12,300 Speaker 2: actual physical cash outlay that's gone out from Medifa is 620 00:30:12,360 --> 00:30:15,380 Speaker 2: not backstops and financing. It's to secure supply. It's to 621 00:30:15,420 --> 00:30:20,150 Speaker 2: secure memory and wafers and co-ops and that kind of stuff. Right. 622 00:30:20,340 --> 00:30:21,840 Speaker 2: And I mean, a lot of that was I mean, 623 00:30:21,860 --> 00:30:24,420 Speaker 2: they're based on his own partnerships and relationships as well. 624 00:30:24,720 --> 00:30:25,020 Speaker 2: All right. 625 00:30:25,200 --> 00:30:28,920 Speaker 3: Stacey Raskin, we'll have to leave it there. Very fast 626 00:30:29,120 --> 00:30:31,920 Speaker 3: unfolding story, but glad we could get a preview in 627 00:30:32,280 --> 00:30:34,840 Speaker 3: before the earnings. Thank you so much for coming back 628 00:30:34,900 --> 00:30:35,340 Speaker 3: on Odd Law. 629 00:30:35,760 --> 00:30:37,020 Speaker 2: Oh, you bet. My pleasure. Anytime. 630 00:30:49,900 --> 00:30:53,180 Speaker 3: Lots of big questions, Tracy. One thing that Stacey mentioned 631 00:30:53,460 --> 00:30:56,500 Speaker 3: that I think is interesting is what happens now to 632 00:30:56,640 --> 00:31:00,040 Speaker 3: all of the really rich NVIDIA employees? Do they just like, 633 00:31:00,080 --> 00:31:02,360 Speaker 3: you know what? Maybe now is a good time to retire. 634 00:31:02,440 --> 00:31:05,280 Speaker 3: Maybe now is a good time to just enjoy life. 635 00:31:05,360 --> 00:31:07,740 Speaker 3: I don't think tech people have a reputation for being like, 636 00:31:08,120 --> 00:31:10,720 Speaker 3: you know what? I'm good. I'm ready. I'm ready to 637 00:31:10,840 --> 00:31:13,120 Speaker 3: just enjoy life. Maybe a few of them will take 638 00:31:13,160 --> 00:31:15,490 Speaker 3: a pause here and just Just lock out the door. 639 00:31:15,710 --> 00:31:17,910 Speaker 4: This has been a question that I've been asking about 640 00:31:17,950 --> 00:31:23,870 Speaker 4: the anthropic IPO, like this huge liquidity event for San Francisco, basically. 641 00:31:23,970 --> 00:31:27,470 Speaker 4: And what do people think that everyone who owns that 642 00:31:27,530 --> 00:31:29,730 Speaker 4: stock is actually going to do with the money? And 643 00:31:29,790 --> 00:31:33,430 Speaker 4: no one seems to know because no one really seems 644 00:31:33,470 --> 00:31:35,650 Speaker 4: to know what tech bros want to do with their 645 00:31:35,690 --> 00:31:37,760 Speaker 4: lives other than Develop tech. 646 00:31:37,820 --> 00:31:41,470 Speaker 3: Yeah, that's the thing. Like if you're like at Anthropic, there's, 647 00:31:41,590 --> 00:31:43,850 Speaker 3: you know, people, I'm not going to say the word, 648 00:31:43,910 --> 00:31:45,800 Speaker 3: but other people call it a cult. 649 00:31:45,820 --> 00:31:46,980 Speaker 1: So it's like, all right, it doesn't matter. 650 00:31:47,000 --> 00:31:49,840 Speaker 3: You become billionaires. You just keep working at cult in a. 651 00:31:49,840 --> 00:31:50,420 Speaker 2: Good way, maybe. 652 00:31:50,810 --> 00:31:53,229 Speaker 3: But like if there was a cult of Jensen and 653 00:31:53,270 --> 00:31:56,630 Speaker 3: he's gone, then it's like, yeah, like why stay if 654 00:31:56,650 --> 00:31:58,010 Speaker 3: you're worth $ 100 million? Maybe not. 655 00:31:58,250 --> 00:32:02,060 Speaker 4: Also, what happens to his shareholdings alone, right? Is he 656 00:32:02,080 --> 00:32:04,720 Speaker 4: going to sell those? He would probably have a decent 657 00:32:04,760 --> 00:32:06,860 Speaker 4: reason to at the moment. But I also think like 658 00:32:06,880 --> 00:32:10,600 Speaker 4: the one takeaway, sort of macro takeaway from all of 659 00:32:10,640 --> 00:32:13,620 Speaker 4: this for me is that it feels like the China 660 00:32:13,660 --> 00:32:17,420 Speaker 4: hawkishness is going to ratchet up. in DC, like just 661 00:32:17,800 --> 00:32:22,740 Speaker 4: based on the optics alone, even without like actual factual 662 00:32:22,780 --> 00:32:26,550 Speaker 4: verification of all these claims. feels like we're starting to 663 00:32:26,630 --> 00:32:29,270 Speaker 4: see that. And so that, like, it'll be interesting to 664 00:32:29,310 --> 00:32:33,110 Speaker 4: see how NVIDIA actually reacts to that, especially given that 665 00:32:33,170 --> 00:32:35,810 Speaker 4: Jensen is one of the chips guys who has been 666 00:32:35,850 --> 00:32:39,310 Speaker 4: making the argument that, like, actually diversification into the China 667 00:32:39,350 --> 00:32:41,690 Speaker 4: market would be good for us in the future. 668 00:32:41,990 --> 00:32:45,660 Speaker 3: No, this is going to be huge. And seeing, like, 669 00:32:45,760 --> 00:32:49,210 Speaker 3: Besant throw Lutnik under the bus this morning, It was 670 00:32:49,410 --> 00:32:54,290 Speaker 3: obviously this inter sort of internecine fight within the White 671 00:32:54,330 --> 00:32:58,460 Speaker 3: House on the sort of China chips question that's now 672 00:32:58,680 --> 00:33:01,140 Speaker 3: clearly spilled out into the public in. 673 00:33:01,060 --> 00:33:01,920 Speaker 2: A big way. 674 00:33:02,760 --> 00:33:06,220 Speaker 3: And I think the people who argue like, no, let's 675 00:33:06,280 --> 00:33:09,160 Speaker 3: sell into China. You have to remember, you know, one 676 00:33:09,180 --> 00:33:13,550 Speaker 3: thing I think that's important to remember is like People 677 00:33:13,630 --> 00:33:15,830 Speaker 3: may say, we don't want to sell chips into China 678 00:33:15,850 --> 00:33:18,050 Speaker 3: because we don't want their AI models to catch up 679 00:33:18,070 --> 00:33:21,460 Speaker 3: with ours. We don't have a ton of leverage. China 680 00:33:21,880 --> 00:33:24,660 Speaker 3: can say, all right, fine, no more rare earths, etc. 681 00:33:25,020 --> 00:33:29,720 Speaker 3: So it's like these are very precarious models. unstable. These 682 00:33:29,780 --> 00:33:33,040 Speaker 3: are relationships that are easily knocked off course because China 683 00:33:33,120 --> 00:33:36,210 Speaker 3: obviously has plenty of leverage over the U.S. and many 684 00:33:36,250 --> 00:33:40,930 Speaker 3: other areas. So like you knock that a little bit 685 00:33:41,010 --> 00:33:43,010 Speaker 3: and you don't really know how it's going to spill over. 686 00:33:43,050 --> 00:33:47,410 Speaker 3: But clearly, clearly the China hawks have like seized a 687 00:33:47,450 --> 00:33:50,080 Speaker 3: moment here and are trying to like get a big 688 00:33:50,260 --> 00:33:53,580 Speaker 3: change in the way the U.S. conducts export policy. 689 00:33:53,790 --> 00:33:54,230 Speaker 2: Absolutely. 690 00:33:54,290 --> 00:33:57,030 Speaker 1: And Besson didn't just, like, throw Lutnik under the bus. 691 00:33:57,050 --> 00:33:59,530 Speaker 4: He had the bus driver, like, back up and run 692 00:33:59,550 --> 00:34:02,010 Speaker 4: him over a few more times and then block the ambulance. 693 00:34:02,090 --> 00:34:04,530 Speaker 1: And, like, this was a throwing under the bus of 694 00:34:04,550 --> 00:34:05,030 Speaker 1: epic proportions. 695 00:34:05,050 --> 00:34:05,440 Speaker 2: Yeah, on TV. 696 00:34:05,590 --> 00:34:07,510 Speaker 1: Yeah, all right. Shall we leave it there? 697 00:34:07,750 --> 00:34:08,390 Speaker 2: Yeah, let's leave it there. 698 00:34:08,670 --> 00:34:11,110 Speaker 4: This has been another episode of the All Thoughts Podcast. 699 00:34:11,280 --> 00:34:13,900 Speaker 4: I'm Tracey Alloway. You can follow me at Tracey Alloway. 700 00:34:14,100 --> 00:34:16,919 Speaker 3: And I'm Jill Wiesenthal. You can follow me at TheStalwart. 701 00:34:17,219 --> 00:34:22,490 Speaker 3: Follow our producers, Carmen Rodriguez at CarmenArmond-O'Bennett at DashbotKaleBrooks at KaleBrooks, 702 00:34:22,840 --> 00:34:26,069 Speaker 3: and Kevin Lozano at KevinLloydLozano. And of course, you can 703 00:34:26,090 --> 00:34:29,290 Speaker 3: chat about all of these topics 24-7 in our Discord. 704 00:34:29,750 --> 00:34:32,290 Speaker 3: And Tracy and I write every day for the OddLots newsletter, 705 00:34:32,310 --> 00:34:36,060 Speaker 3: which you can find and subscribe to at Bloomberg.com slash OddLots. 706 00:34:36,440 --> 00:34:38,680 Speaker 4: And if you enjoyed this conversation, if you like it 707 00:34:38,739 --> 00:34:41,259 Speaker 4: when we talk about the future of NVIDIA from the 708 00:34:41,360 --> 00:34:43,790 Speaker 4: perspective of the analysts that cover it, then please leave 709 00:34:43,830 --> 00:34:46,810 Speaker 4: us a positive review on your favorite podcast platform. And 710 00:34:46,830 --> 00:34:49,489 Speaker 4: if you're watching on video, then please like or leave 711 00:34:49,530 --> 00:34:51,390 Speaker 4: a comment or better yet, subscribe. 712 00:34:51,770 --> 00:34:54,490 Speaker 3: And remember, if you're a Bloomberg subscriber, you can listen 713 00:34:54,530 --> 00:34:57,540 Speaker 3: to all of our episodes absolutely ad-free. All you need 714 00:34:57,560 --> 00:34:59,719 Speaker 3: to do is go to the Bloomberg channel on Apple 715 00:34:59,780 --> 00:35:02,080 Speaker 3: Podcasts and follow the instructions there. 716 00:35:02,360 --> 00:35:03,080 Speaker 1: Thanks for joining. 717 00:35:11,460 --> 00:35:11,960 Speaker 2: So so