1 00:00:00,040 --> 00:00:15,080 Speaker 1: Bloomberg Audio Studios, podcasts, radio news. Bloomberg Tech is live 2 00:00:15,200 --> 00:00:18,880 Speaker 1: from the heart of Silicon Valley with Ed Lavow in 3 00:00:19,079 --> 00:00:20,480 Speaker 1: San Francisco. 4 00:00:22,760 --> 00:00:25,319 Speaker 2: This is Bloomberg Tech coming up. Chip stocks are headed 5 00:00:25,320 --> 00:00:28,960 Speaker 2: toward their best quarter ever, but recent swings are making 6 00:00:28,960 --> 00:00:33,040 Speaker 2: investors concerned about the durability of the AI boom. Plus 7 00:00:33,080 --> 00:00:36,440 Speaker 2: Taiwan raids the offices of super Micro as part of 8 00:00:36,479 --> 00:00:41,280 Speaker 2: an investigation into the alleged smuggling of Nvidio chips into China, 9 00:00:41,840 --> 00:00:46,920 Speaker 2: and a billion dollar misunderstanding leaves Korean investors empty handed 10 00:00:47,080 --> 00:00:52,599 Speaker 2: in the blockbuster SpaceX ipo. Everything is about semiconductors at 11 00:00:52,600 --> 00:00:54,360 Speaker 2: the moment, in the here and now. The NASA one 12 00:00:54,400 --> 00:00:57,920 Speaker 2: hundred is higher, and there is outperformance in chipstocks on 13 00:00:58,040 --> 00:01:01,400 Speaker 2: the SOX today's session, but also so over the last 14 00:01:01,440 --> 00:01:04,000 Speaker 2: three months. We have to zoom out. The SOCKS is 15 00:01:04,000 --> 00:01:07,240 Speaker 2: now on track for its best quarter on record ever 16 00:01:07,640 --> 00:01:09,800 Speaker 2: with the final trading day of June and the second 17 00:01:09,880 --> 00:01:12,560 Speaker 2: quarter upon us. The SOCKS is up eighty six percent 18 00:01:12,920 --> 00:01:14,800 Speaker 2: over the last three months. It's also up almost one 19 00:01:15,360 --> 00:01:17,920 Speaker 2: hundred percent year today. I would make the point they're 20 00:01:17,959 --> 00:01:20,160 Speaker 2: only halfway through the year. We'll see where we net 21 00:01:20,160 --> 00:01:23,000 Speaker 2: out when the year end. Even after recent swings, this 22 00:01:23,040 --> 00:01:26,680 Speaker 2: has become a historic rally driven by one question, how 23 00:01:26,880 --> 00:01:29,800 Speaker 2: long can the AI infrastructure spending boomlast? This's gout out 24 00:01:29,800 --> 00:01:32,960 Speaker 2: its Bloomberg Tech equity reporter Rhyan Lasceellak. You've been writing 25 00:01:32,959 --> 00:01:35,000 Speaker 2: about this on the Bloomberg and the main point being 26 00:01:35,040 --> 00:01:38,480 Speaker 2: that we're on track for the biggest quarter ever on 27 00:01:38,560 --> 00:01:42,360 Speaker 2: the socks. But there have been swings, there has been volatility. 28 00:01:42,560 --> 00:01:43,280 Speaker 2: What do we need to know? 29 00:01:44,560 --> 00:01:47,360 Speaker 3: Hey, Yeah, so this has been a very violable quarter 30 00:01:47,480 --> 00:01:49,920 Speaker 3: four stars, although the general move has been not just 31 00:01:50,080 --> 00:01:52,400 Speaker 3: up but dramatically up. But just to give you a 32 00:01:52,440 --> 00:01:56,440 Speaker 3: sense of the volatility, just yesterday the starks at one 33 00:01:56,480 --> 00:01:59,120 Speaker 3: point fell three percent, then a reverse course and it 34 00:01:59,320 --> 00:02:02,240 Speaker 3: ended up three percent, So a six percentage point swing 35 00:02:02,320 --> 00:02:04,920 Speaker 3: within a day. And this isn't coming on any kind 36 00:02:04,920 --> 00:02:08,040 Speaker 3: of fundamental change in the outlook. People are really having, 37 00:02:08,680 --> 00:02:10,959 Speaker 3: you know, a lot of debate right now about how 38 00:02:11,000 --> 00:02:14,120 Speaker 3: long is this AI capex going to stay at the 39 00:02:14,200 --> 00:02:16,239 Speaker 3: level that it's been for the past few years, and 40 00:02:16,280 --> 00:02:18,680 Speaker 3: it's not just at this level, but it's been accelerating. 41 00:02:18,880 --> 00:02:22,320 Speaker 3: How much longer can we continue to expect that kind 42 00:02:22,320 --> 00:02:25,600 Speaker 3: of spending acceleration going forward, and if it starts to 43 00:02:25,639 --> 00:02:28,200 Speaker 3: slow down, does that mean that we're actually at peak 44 00:02:28,280 --> 00:02:29,280 Speaker 3: earnings right now? 45 00:02:29,320 --> 00:02:29,600 Speaker 4: And when? 46 00:02:29,600 --> 00:02:32,160 Speaker 3: If that's the case, what is the outlook for the stocks, 47 00:02:32,280 --> 00:02:34,280 Speaker 3: especially following the move that we've had. 48 00:02:35,600 --> 00:02:39,000 Speaker 2: Phase one of the AI wave of this decade at 49 00:02:39,120 --> 00:02:44,240 Speaker 2: least Swizz Nvidia right and if you expanded that GPUs 50 00:02:44,320 --> 00:02:49,079 Speaker 2: and compute right now, what's driving this is memory and storage? 51 00:02:49,160 --> 00:02:52,160 Speaker 2: Could you explain a little bit the biggest contributors to 52 00:02:52,240 --> 00:02:54,680 Speaker 2: the rally that we've seen in chip stocks overall. 53 00:02:55,360 --> 00:02:57,520 Speaker 3: Yeah, So, I'd say the real story this year has 54 00:02:57,600 --> 00:03:02,519 Speaker 3: been in particular Micron, but also so sand Disk, Western Digital, Seagate, 55 00:03:02,720 --> 00:03:05,520 Speaker 3: all these companies that are involved in memory and storage. 56 00:03:05,520 --> 00:03:08,880 Speaker 3: People are really starting to appreciate how central this is 57 00:03:09,080 --> 00:03:11,560 Speaker 3: when you're building out data centers, and these companies have 58 00:03:11,600 --> 00:03:15,440 Speaker 3: seen astronomical growth, including Micron with I believe it was 59 00:03:15,480 --> 00:03:17,519 Speaker 3: just last week that its report came out and it 60 00:03:17,639 --> 00:03:20,640 Speaker 3: talked of in a game extremely robust forecast, really pointing 61 00:03:20,639 --> 00:03:23,800 Speaker 3: to how they see visibility for a really extended period 62 00:03:23,800 --> 00:03:26,000 Speaker 3: of time. This is a real change from the way 63 00:03:26,120 --> 00:03:29,480 Speaker 3: Micron and other companies like this have historically operated, where 64 00:03:29,480 --> 00:03:33,480 Speaker 3: they were far more cyclical moving on consumer trends, including 65 00:03:33,520 --> 00:03:37,440 Speaker 3: like phones, PCs, video game consoles, that kind of thing. 66 00:03:37,480 --> 00:03:40,160 Speaker 3: What we've seen Apple come out and they're increasing their 67 00:03:40,560 --> 00:03:43,120 Speaker 3: prices for phones. 68 00:03:42,720 --> 00:03:43,800 Speaker 2: And iPads and so forth. 69 00:03:43,840 --> 00:03:45,840 Speaker 3: Microsoft came out and said it's going to raise the 70 00:03:45,840 --> 00:03:48,640 Speaker 3: price for the Xbox because of higher memory. This has 71 00:03:48,720 --> 00:03:50,839 Speaker 3: really been the story right now. This has really been 72 00:03:51,000 --> 00:03:54,760 Speaker 3: the bottleneck in the AI trade. There simply are not 73 00:03:55,080 --> 00:03:58,000 Speaker 3: enough memory chips being made to meet all this demand, 74 00:03:58,000 --> 00:04:00,520 Speaker 3: which means that demand is likely going to stay elevated 75 00:04:00,960 --> 00:04:04,040 Speaker 3: for really the foreseeable future, probably into twenty twenty seven, 76 00:04:04,120 --> 00:04:06,240 Speaker 3: maybe into twenty twenty eight. I've had some people say 77 00:04:06,920 --> 00:04:09,600 Speaker 3: maybe as early as twenty twenty thirty is when this 78 00:04:09,640 --> 00:04:11,000 Speaker 3: is really going to start to change. 79 00:04:12,200 --> 00:04:15,040 Speaker 2: Ryan, the biggest question in technology and in markets is 80 00:04:15,080 --> 00:04:19,120 Speaker 2: always what happens next. Is there any sense from strategists 81 00:04:19,120 --> 00:04:23,560 Speaker 2: and analysts of whether the socks continues on this like 82 00:04:24,600 --> 00:04:27,839 Speaker 2: amazing trajectory for the balance of twenty twenty six. 83 00:04:29,000 --> 00:04:31,120 Speaker 3: Well, I will say that most people are very confident 84 00:04:31,160 --> 00:04:34,000 Speaker 3: about the fundamental story, so they do expect the cabbecks 85 00:04:34,040 --> 00:04:37,400 Speaker 3: to continue. They expect a demand to remain extremely high. 86 00:04:37,560 --> 00:04:39,640 Speaker 3: The question of course, is really how much will that 87 00:04:39,720 --> 00:04:42,200 Speaker 3: continue to translate to the stocks the way they have been. 88 00:04:42,240 --> 00:04:44,680 Speaker 3: I think there's a little bit more skepticism about that. 89 00:04:45,080 --> 00:04:48,120 Speaker 3: Some of these stocks, including Intel, including arm, they are 90 00:04:48,240 --> 00:04:52,960 Speaker 3: very expensive on traditional valuation metrics. A company like Micron 91 00:04:53,120 --> 00:04:56,080 Speaker 3: looks pretty cheap, but historically that's been a bad time 92 00:04:56,160 --> 00:04:58,440 Speaker 3: to buy it because that signals it's more in your 93 00:04:58,480 --> 00:05:00,320 Speaker 3: peak earnings. But again you go back to the question 94 00:05:00,360 --> 00:05:02,880 Speaker 3: of is this actually peak earnings or is this cycle 95 00:05:03,040 --> 00:05:06,200 Speaker 3: very different than historical ones? And we're not anywhere near 96 00:05:06,320 --> 00:05:08,160 Speaker 3: that kind of peak yet. So as far as what's 97 00:05:08,200 --> 00:05:11,040 Speaker 3: going to happen, obviously nobody knows. It seems hard to 98 00:05:11,080 --> 00:05:14,200 Speaker 3: imagine that we'll see another what ninety percent move over 99 00:05:14,200 --> 00:05:15,960 Speaker 3: the next three months, but of course, you know all 100 00:05:15,960 --> 00:05:17,279 Speaker 3: that really means to be seen. But so really this 101 00:05:17,279 --> 00:05:19,120 Speaker 3: is where the more mintum is in the market right now. 102 00:05:19,760 --> 00:05:23,400 Speaker 2: Bloombo's around for SELCA, thank you very much. Indeed, AI 103 00:05:23,520 --> 00:05:27,000 Speaker 2: doesn't run on chips alone. As hyperscalers continue ramping up 104 00:05:27,040 --> 00:05:30,720 Speaker 2: capital spending, investors are increasingly looking beyond the likes of 105 00:05:30,880 --> 00:05:35,120 Speaker 2: video to the company's supplying yes, the memory, but also networking, cooling, 106 00:05:35,160 --> 00:05:37,880 Speaker 2: and power needed to support the AI build out. Join 107 00:05:38,000 --> 00:05:41,080 Speaker 2: US Now is Rob Fummel's senior portfolio manager. It taught 108 00:05:41,120 --> 00:05:45,080 Speaker 2: US Capital. Just to recap, the Philadelphia Semiconductor Index is 109 00:05:45,240 --> 00:05:48,239 Speaker 2: likely to have its best quarter ever eighty six percent 110 00:05:48,320 --> 00:05:51,680 Speaker 2: gain over three months. Rob, what does that signal to you? 111 00:05:53,160 --> 00:05:55,920 Speaker 5: Well, I think that the you know, the AI story 112 00:05:56,000 --> 00:05:57,719 Speaker 5: is still here. It's intact deed, and it's going to 113 00:05:57,720 --> 00:06:01,440 Speaker 5: continue growing. And we continue to see cap spending growing. 114 00:06:01,480 --> 00:06:05,400 Speaker 5: And we've got an AI Infrastructure Active ETF that we've 115 00:06:05,440 --> 00:06:08,680 Speaker 5: designed to capture exactly what you said. The infrastructure, that's 116 00:06:08,680 --> 00:06:11,279 Speaker 5: where there's the opportunity, and that includes things like rian 117 00:06:11,320 --> 00:06:14,400 Speaker 5: and you were talking about like memory, but the base 118 00:06:14,520 --> 00:06:18,800 Speaker 5: level of infrastructure is electricity, and that's where we're going 119 00:06:18,839 --> 00:06:22,080 Speaker 5: to see tremendous demand for electricity going forward. The US 120 00:06:22,120 --> 00:06:25,680 Speaker 5: has got a competitive advantage in terms of being able 121 00:06:25,680 --> 00:06:28,920 Speaker 5: to provide low cost electricity, and so we see lots 122 00:06:28,920 --> 00:06:31,640 Speaker 5: of opportunities in memory and storage, but also in a 123 00:06:31,800 --> 00:06:35,839 Speaker 5: really basic commodity, which is electricity. 124 00:06:36,360 --> 00:06:39,960 Speaker 2: I'm wanting to understand, at least from your perspective, Rob, 125 00:06:40,480 --> 00:06:46,760 Speaker 2: how direct the correlation is between growing capital expenditure and 126 00:06:46,880 --> 00:06:49,760 Speaker 2: the sort of outlook for let's say memory, just to 127 00:06:49,839 --> 00:06:53,080 Speaker 2: keep it more focused. If capital expenditures from the HYPERSCTUS 128 00:06:53,120 --> 00:06:56,039 Speaker 2: continues to go up, then the recipients of that CAPEX 129 00:06:56,200 --> 00:06:59,560 Speaker 2: memory names in particular, you would expect them to have 130 00:07:00,120 --> 00:07:02,040 Speaker 2: the sort of similar continued trajectory. 131 00:07:03,560 --> 00:07:03,760 Speaker 6: Yeah. 132 00:07:03,839 --> 00:07:05,120 Speaker 5: I think that's a simple way to look at it, 133 00:07:05,120 --> 00:07:06,880 Speaker 5: and I think it's an accurate way to look at it. 134 00:07:07,279 --> 00:07:10,640 Speaker 5: I think if capital spending continues to grow at twenty 135 00:07:10,760 --> 00:07:13,720 Speaker 5: thirty forty percent anualized growth rates, well, then I think 136 00:07:13,760 --> 00:07:16,080 Speaker 5: you can expect you know, where's that money going. Well, 137 00:07:16,120 --> 00:07:18,640 Speaker 5: it's going into the infrastructure. So I think you can 138 00:07:18,680 --> 00:07:22,240 Speaker 5: expect infrastructure memory, storage to grow, demand to grow twenty 139 00:07:22,320 --> 00:07:27,400 Speaker 5: thirty percent, liquid cooling, network switches, all all of that 140 00:07:27,440 --> 00:07:32,600 Speaker 5: infrastructure that's really essential for AI to continue to operate. Well, 141 00:07:33,680 --> 00:07:37,200 Speaker 5: the growth rate will mirror really what capital spending does. 142 00:07:38,480 --> 00:07:38,600 Speaker 4: Rob. 143 00:07:38,680 --> 00:07:41,239 Speaker 2: One of the one of the many reasons we've enjoyed 144 00:07:41,280 --> 00:07:43,520 Speaker 2: having you on Bloomberg Tech is you're looking at this 145 00:07:43,640 --> 00:07:47,520 Speaker 2: much more holistically. Right what goes into a data center? 146 00:07:47,920 --> 00:07:49,880 Speaker 2: And right at the top of the conversation we said 147 00:07:49,920 --> 00:07:54,440 Speaker 2: it's more than memory, networking, power cooling. Right now, you know, 148 00:07:54,560 --> 00:07:56,760 Speaker 2: is you plan for the balance of twenty six where 149 00:07:56,760 --> 00:08:00,520 Speaker 2: are you positioning yourself based on your belief of of 150 00:08:00,360 --> 00:08:02,920 Speaker 2: the main beneficiaries of this data cent a build out 151 00:08:04,280 --> 00:08:06,040 Speaker 2: well inside the data center. 152 00:08:06,120 --> 00:08:09,040 Speaker 5: It's more we're probably more positioned on the memory side 153 00:08:09,480 --> 00:08:11,720 Speaker 5: because of what what you were talking about with Ryan. 154 00:08:11,760 --> 00:08:13,840 Speaker 5: I mean, if you look at the Microns, what the 155 00:08:13,880 --> 00:08:17,480 Speaker 5: earnings Micron posted, it's very clear that the market is 156 00:08:17,720 --> 00:08:21,480 Speaker 5: uh is undersupplied from memory, right, I mean, Micron posted 157 00:08:21,520 --> 00:08:24,640 Speaker 5: a gross margin of eighty five percent. I think that's 158 00:08:24,640 --> 00:08:27,440 Speaker 5: ten percent, ten points higher than what Nvidia is. I mean, 159 00:08:27,440 --> 00:08:29,680 Speaker 5: I mean that's that's just phenomenal, right, So and and 160 00:08:29,680 --> 00:08:34,640 Speaker 5: and the memory market, as Ryan highlighted, just continues to grow. 161 00:08:34,720 --> 00:08:38,160 Speaker 5: So we continue to see substantial growth in the in 162 00:08:38,200 --> 00:08:40,720 Speaker 5: the earnings and the cash flow of companies like Micron 163 00:08:40,760 --> 00:08:42,800 Speaker 5: and the other memory names. Now, don't forget though, the 164 00:08:42,840 --> 00:08:45,920 Speaker 5: base level though, of of of energy and electricity. That's 165 00:08:45,960 --> 00:08:49,040 Speaker 5: that's uh, that's probably the base level of the of 166 00:08:49,080 --> 00:08:51,520 Speaker 5: the five layer cake that that that Jensen Wong likes 167 00:08:51,559 --> 00:08:53,520 Speaker 5: to talk about. We have a lot of exposure to 168 00:08:53,920 --> 00:08:56,839 Speaker 5: that that as well. Energy is ours our expertise, I 169 00:08:56,840 --> 00:08:59,800 Speaker 5: guess in the past, and we still see energy electricity 170 00:08:59,840 --> 00:09:04,120 Speaker 5: being the real critical driver of this AI trade going. 171 00:09:04,000 --> 00:09:07,680 Speaker 2: Forward as well. Just very quickly, Rob, before we let 172 00:09:07,760 --> 00:09:10,840 Speaker 2: you go, sk Heinex is coming to the US capital 173 00:09:10,920 --> 00:09:15,160 Speaker 2: markets with a US listing for ADRs. How would you 174 00:09:15,200 --> 00:09:18,400 Speaker 2: respond to that, Bearing in mind that the sk has 175 00:09:18,440 --> 00:09:21,240 Speaker 2: for fifty seven percent market share something like that of 176 00:09:21,320 --> 00:09:23,080 Speaker 2: high bandwidth memory at present. 177 00:09:24,600 --> 00:09:29,160 Speaker 5: Well as investors in AI infrastructure, we welcome new entrants 178 00:09:29,600 --> 00:09:32,160 Speaker 5: or new companies that we can buy right and s 179 00:09:32,280 --> 00:09:33,960 Speaker 5: j Heinigs will be at the top of the list 180 00:09:34,080 --> 00:09:35,640 Speaker 5: to take a look at. We've been looking at that 181 00:09:35,720 --> 00:09:38,920 Speaker 5: company obviously and have the ability to buy it internationally, 182 00:09:38,920 --> 00:09:40,880 Speaker 5: but but to buy it on a buy it as 183 00:09:40,880 --> 00:09:43,760 Speaker 5: an ADR may make it more attractive for us. We'll 184 00:09:43,800 --> 00:09:45,680 Speaker 5: take a very close look at it though, because it 185 00:09:45,720 --> 00:09:48,600 Speaker 5: is part of that AI infrastructure that you were mentioning earlier, 186 00:09:48,640 --> 00:09:51,120 Speaker 5: and it's really critical and it's something that we're trying 187 00:09:51,120 --> 00:09:52,280 Speaker 5: to capture for investors. 188 00:09:53,559 --> 00:09:56,400 Speaker 2: Okay, Rob Farm or the Tortoise Capital somebody that is, 189 00:09:56,760 --> 00:09:59,120 Speaker 2: let's say, across the data center from top to tail. 190 00:09:59,559 --> 00:10:02,080 Speaker 2: Thank you for very much. Indeed, now coming up, super 191 00:10:02,120 --> 00:10:07,320 Speaker 2: Micro's Taiwan office is raided as authorities investigate alleged illegal 192 00:10:07,360 --> 00:10:10,199 Speaker 2: exports to China. Those are the US listed shares a 193 00:10:10,240 --> 00:10:13,199 Speaker 2: super micro and they are down almost five percent over 194 00:10:13,240 --> 00:10:16,600 Speaker 2: a two day basis. We'll track it. This is Bloomberg Tech. 195 00:10:26,520 --> 00:10:29,880 Speaker 2: Bloomberg's David Weston joins US now for an important clarification 196 00:10:30,080 --> 00:10:31,040 Speaker 2: on an earlier story. 197 00:10:31,160 --> 00:10:33,760 Speaker 7: David, Yes, thank you. Just a short time ago we 198 00:10:33,800 --> 00:10:38,360 Speaker 7: reported off of an NPR report the Justice Samuel Aldo 199 00:10:38,640 --> 00:10:41,720 Speaker 7: had resigned from the Supreme coo United States. That was 200 00:10:41,760 --> 00:10:45,440 Speaker 7: off of NPR. Since then has said they retract that statement. 201 00:10:45,520 --> 00:10:47,640 Speaker 7: I don't know what happened. The Supreme Court has said 202 00:10:47,679 --> 00:10:49,959 Speaker 7: they have made no announcement on the subject, so that 203 00:10:50,000 --> 00:10:53,520 Speaker 7: report was in error. The Samuel Leader, Supreme Court justice 204 00:10:53,760 --> 00:10:56,040 Speaker 7: has resigned from the Supreme Court the United States, and 205 00:10:56,080 --> 00:10:57,640 Speaker 7: we will keep you posted. 206 00:10:58,240 --> 00:11:00,920 Speaker 2: The most David Western, thank you very much. Turning back 207 00:11:00,960 --> 00:11:03,800 Speaker 2: to tech, the battle over AI chips is moving beyond 208 00:11:03,880 --> 00:11:09,400 Speaker 2: Washington and Beijing. Taiwanese authorities are investigating the alleged smuggling 209 00:11:09,559 --> 00:11:14,559 Speaker 2: of super microservers containing in video chips to China, underscoring 210 00:11:14,600 --> 00:11:20,280 Speaker 2: the growing focus on enforcing US export controls. Bloomberg's Minmnlo reports. 211 00:11:21,440 --> 00:11:24,720 Speaker 8: Sources have told us that Taiwanese authorities have raided the 212 00:11:24,760 --> 00:11:28,360 Speaker 8: residences of six individuals, as well as the sites of 213 00:11:28,480 --> 00:11:33,719 Speaker 8: three affiliated companies, including super Micro's office in Taiwan, as 214 00:11:33,720 --> 00:11:37,439 Speaker 8: well as one of its distributors, Albatron Technology, and a 215 00:11:37,520 --> 00:11:41,080 Speaker 8: Taiwanese data center operated Chief Telecom. Now, this is part 216 00:11:41,080 --> 00:11:44,160 Speaker 8: of an ongoing probe looking into the alleged smuggling of 217 00:11:44,240 --> 00:11:49,200 Speaker 8: super microservice containing in vidious chips to China. Now, that's 218 00:11:49,200 --> 00:11:52,560 Speaker 8: in violation of US export controls, but it's not in 219 00:11:52,640 --> 00:11:56,400 Speaker 8: violation of local laws in Taiwan. But the US has 220 00:11:56,440 --> 00:11:59,640 Speaker 8: been pressuring Taiwan to match the US wars, and we 221 00:11:59,679 --> 00:12:03,400 Speaker 8: know that how many these authorities are considering criminalizing the 222 00:12:03,480 --> 00:12:07,240 Speaker 8: act of exploring these high end chips to China. For now, though, 223 00:12:07,280 --> 00:12:11,480 Speaker 8: the only legal recourse is to find existing rules such 224 00:12:11,520 --> 00:12:16,120 Speaker 8: as falsifying documentation, for example, to charge any alleged chip 225 00:12:16,200 --> 00:12:20,560 Speaker 8: smuggling case. Investigation is ongoing, and super Micro has said 226 00:12:20,600 --> 00:12:25,160 Speaker 8: that they will work closely with law enforcement. Mimi Law, Bloomberg. 227 00:12:24,679 --> 00:12:31,520 Speaker 2: News, Hong Kong, Bloomberg's Midmilow reporting. Washington's lobbying industry is 228 00:12:31,559 --> 00:12:34,480 Speaker 2: feeling the impact of the US China tech rivalries. Some 229 00:12:34,800 --> 00:12:37,880 Speaker 2: of the capital's biggest lobbying firms are cutting ties with 230 00:12:37,960 --> 00:12:41,520 Speaker 2: Chinese set giants, including Ali, Barber and ten Cent, as 231 00:12:41,559 --> 00:12:45,520 Speaker 2: new US restrictions force them to choose between representing Chinese 232 00:12:45,520 --> 00:12:51,080 Speaker 2: companies or American defense contractors. Bloomberg's Ted Man joins us 233 00:12:51,080 --> 00:12:53,000 Speaker 2: for more. Ted it was really interesting to get into this. 234 00:12:53,080 --> 00:12:55,920 Speaker 2: I learned a lot about kind of what happens in 235 00:12:56,240 --> 00:13:00,040 Speaker 2: the hallways of the nation's capital. But let's start with 236 00:13:00,160 --> 00:13:02,960 Speaker 2: the sort of crux of what we're reporting. There is 237 00:13:03,000 --> 00:13:06,439 Speaker 2: a whole industry and a lobbying industry that basically now 238 00:13:06,640 --> 00:13:11,880 Speaker 2: feels it can't represent some of the biggest tech companies 239 00:13:11,920 --> 00:13:15,160 Speaker 2: in China. That's right. 240 00:13:15,440 --> 00:13:18,800 Speaker 9: The lobbying companies have reached a time for choosing here. 241 00:13:19,040 --> 00:13:21,959 Speaker 9: They have to decide whether to continue to represent those 242 00:13:21,960 --> 00:13:24,839 Speaker 9: companies or an In so doing, they would have to 243 00:13:24,880 --> 00:13:27,600 Speaker 9: forego working with companies that work with the Pentagon, and 244 00:13:27,600 --> 00:13:29,679 Speaker 9: that's just too big a slice of the book of 245 00:13:29,720 --> 00:13:33,080 Speaker 9: business for those major firms, and so they're having to 246 00:13:33,240 --> 00:13:36,560 Speaker 9: drop the Chinese companies instead. This is something that has 247 00:13:36,559 --> 00:13:40,000 Speaker 9: been building for more than a year, and then they're 248 00:13:40,040 --> 00:13:42,679 Speaker 9: finally at the deadline now where they had to drop. 249 00:13:42,520 --> 00:13:48,960 Speaker 2: Those companies historically sort of how effective are the tech 250 00:13:49,000 --> 00:13:52,760 Speaker 2: lobbyists on getting the government of the day to make 251 00:13:53,000 --> 00:13:56,240 Speaker 2: decisions or take action that's in their favor. I mean, 252 00:13:56,760 --> 00:14:01,160 Speaker 2: this administration in particular has sort of a range of 253 00:14:01,200 --> 00:14:06,280 Speaker 2: what they would call hawkishness on China. The exports or 254 00:14:06,320 --> 00:14:08,640 Speaker 2: not to export chips has kind of been at the 255 00:14:08,640 --> 00:14:10,280 Speaker 2: heart of that. But I think it would be really 256 00:14:10,360 --> 00:14:13,440 Speaker 2: valuable sub as a wanderings to kind of understand how 257 00:14:13,480 --> 00:14:16,079 Speaker 2: this has typically worked in practice. 258 00:14:17,880 --> 00:14:20,280 Speaker 9: Yeah, I mean, I think that the lobbying has been 259 00:14:20,520 --> 00:14:23,480 Speaker 9: very aggressive over the years, and there is a range 260 00:14:23,520 --> 00:14:24,400 Speaker 9: of views here. 261 00:14:24,960 --> 00:14:26,680 Speaker 2: It's also it's. 262 00:14:26,320 --> 00:14:30,160 Speaker 9: The current administration, but there's a range of a sort 263 00:14:30,160 --> 00:14:33,640 Speaker 9: of hawkishness in Congress too, where even during the Biden 264 00:14:33,640 --> 00:14:38,240 Speaker 9: administration you had some in Congress and in that administration 265 00:14:38,280 --> 00:14:40,600 Speaker 9: and this one who wanted to take a harder line 266 00:14:40,600 --> 00:14:43,600 Speaker 9: and basically, as someone was quoted saying in our story, 267 00:14:43,680 --> 00:14:47,080 Speaker 9: you know, basically stop this practice where you'd have a 268 00:14:47,080 --> 00:14:49,680 Speaker 9: lobbyist coming in on behalf of a Chinese company, talking 269 00:14:49,720 --> 00:14:52,280 Speaker 9: to a senator or talking to an executive branch official 270 00:14:52,320 --> 00:14:54,920 Speaker 9: and then turning around and advocating on behalf of an 271 00:14:54,920 --> 00:14:57,600 Speaker 9: American company or a vendor to the Pentagon, and so 272 00:14:57,720 --> 00:15:02,080 Speaker 9: for a range of people in the administration and also 273 00:15:02,120 --> 00:15:06,600 Speaker 9: in Congress, that just seemed like just too brazen to them. 274 00:15:06,640 --> 00:15:09,400 Speaker 9: And so this is where there's finally some teeth being 275 00:15:09,440 --> 00:15:13,000 Speaker 9: applied to the law that already created this list at 276 00:15:13,000 --> 00:15:15,600 Speaker 9: the Pentagon. It was a blacklist that didn't really have 277 00:15:15,640 --> 00:15:18,760 Speaker 9: a lot of ramifications before now, and now it really does. 278 00:15:18,840 --> 00:15:20,960 Speaker 9: It's going to cost real money to these firms that 279 00:15:20,960 --> 00:15:24,080 Speaker 9: they're going to have to choose. And as we're seeing 280 00:15:24,400 --> 00:15:27,680 Speaker 9: the biggest firms are deciding to drop those Chinese clients 281 00:15:27,680 --> 00:15:29,760 Speaker 9: because just the math doesn't add up. 282 00:15:30,680 --> 00:15:33,880 Speaker 2: They both said, man, thank you very much. Indeed, another story, 283 00:15:34,040 --> 00:15:38,840 Speaker 2: Apple is reportedly seeking the Trump administration's approval to buy 284 00:15:38,920 --> 00:15:44,240 Speaker 2: memory chips from blacklisted Chinese chip maker CXMT. The Financial 285 00:15:44,280 --> 00:15:47,240 Speaker 2: Time says the company has been lobbying US officials is 286 00:15:47,280 --> 00:15:50,320 Speaker 2: it looks to offset rising memory costs, a move that's 287 00:15:50,360 --> 00:15:54,640 Speaker 2: already drawing criticism from lawmakers. Senator Tom Cotton of Arkansas 288 00:15:54,680 --> 00:15:58,520 Speaker 2: warning on x quote using blacklisted Chinese chips would be 289 00:15:58,560 --> 00:16:01,640 Speaker 2: a grave mistake for Apple. Doing so would risk long 290 00:16:01,760 --> 00:16:05,480 Speaker 2: term shareholder value, the privacy of their customers, and the 291 00:16:05,520 --> 00:16:11,240 Speaker 2: security of this country. Coming up. How a costly mix 292 00:16:11,360 --> 00:16:14,640 Speaker 2: up shut more than a billion dollars in Korean demand 293 00:16:15,080 --> 00:16:19,760 Speaker 2: out of the SPACEXIPO the details Next, this is Blimberg. 294 00:16:19,360 --> 00:16:31,240 Speaker 4: Tech and Houston has checked your superameters. We are good 295 00:16:31,240 --> 00:16:34,520 Speaker 4: to go. So with that, Chris, you can open the 296 00:16:34,720 --> 00:16:35,880 Speaker 4: hatch thermal cover. 297 00:16:39,280 --> 00:16:42,480 Speaker 2: You're looking at pictures of NASA astronauts Chris Williams and 298 00:16:42,560 --> 00:16:46,240 Speaker 2: Jessica Meyer opening an air lock on the International Space Station. 299 00:16:46,320 --> 00:16:50,160 Speaker 2: The pair are replacing a risk joint on the canadarm II, 300 00:16:50,400 --> 00:16:53,960 Speaker 2: a seventeen meter long robotic arm that is essential to 301 00:16:54,000 --> 00:16:57,840 Speaker 2: the ISS for tasks like grabbing cargo carrying spacecraft before 302 00:16:57,840 --> 00:17:00,720 Speaker 2: they dock. The work's expected to keep the astronauts in 303 00:17:00,760 --> 00:17:04,880 Speaker 2: the vacuum of space for more than six hours. Definitely 304 00:17:05,000 --> 00:17:09,359 Speaker 2: not your average maintenance call. Now, turning to other things 305 00:17:09,359 --> 00:17:10,800 Speaker 2: that are out of this world, let's take a look 306 00:17:10,800 --> 00:17:14,400 Speaker 2: at today's big number. Eight hundred percent. That's how much 307 00:17:14,440 --> 00:17:18,320 Speaker 2: SpaceX's revenue could climb by twenty thirty based on a 308 00:17:18,359 --> 00:17:21,560 Speaker 2: new model from Bloomberg Intelligence. The main driver of that 309 00:17:22,440 --> 00:17:26,399 Speaker 2: data center agreements with its XAI unit. Still, even if 310 00:17:26,440 --> 00:17:30,240 Speaker 2: sales rise almost nine times, those same bi researchers say 311 00:17:30,440 --> 00:17:34,960 Speaker 2: SpaceX's valuation could be brought down closer to Earth as talks. 312 00:17:35,040 --> 00:17:38,080 Speaker 2: One of them, George Ferguson, leads aerospace defense coverage for 313 00:17:38,400 --> 00:17:43,040 Speaker 2: Bloomberg Intelligence. You've been keeping the deck and the research updated. 314 00:17:43,400 --> 00:17:46,639 Speaker 2: So that's really interesting. You see top line growth and 315 00:17:46,680 --> 00:17:51,879 Speaker 2: the reason why it's clear valuation being kept in Earth's 316 00:17:51,880 --> 00:17:53,680 Speaker 2: gravitational Paul, what's the thesis? 317 00:17:54,520 --> 00:17:56,560 Speaker 10: Yeah, I mean, I think the thesis here really is 318 00:17:56,680 --> 00:17:59,360 Speaker 10: there needs to be a lot of growth in order 319 00:17:59,440 --> 00:18:03,679 Speaker 10: for space x to sort of merge into valuations that 320 00:18:03,720 --> 00:18:09,520 Speaker 10: are similar to the other hyperscaler peers like Microsoft, Amazon, 321 00:18:10,240 --> 00:18:15,200 Speaker 10: you know, Google, those companies. So yeah, at the iprool price, 322 00:18:15,200 --> 00:18:19,240 Speaker 10: we measured everything. At the IPO price, SpaceX are very 323 00:18:19,400 --> 00:18:23,320 Speaker 10: richly valued. And so if you're involved in it, it's 324 00:18:23,359 --> 00:18:26,919 Speaker 10: all about this turning around the AI story. 325 00:18:27,000 --> 00:18:27,240 Speaker 11: Right. 326 00:18:27,600 --> 00:18:33,439 Speaker 10: They've built these big Colossus one and Colossus two, you 327 00:18:33,480 --> 00:18:37,080 Speaker 10: know computer sorry, it's the names escaped me here, the 328 00:18:37,680 --> 00:18:41,399 Speaker 10: data center sorry, in a long day already, and you 329 00:18:41,400 --> 00:18:43,879 Speaker 10: know they're going to sell out some of that capacity 330 00:18:44,400 --> 00:18:46,960 Speaker 10: and they need to build some of the advertising. They've 331 00:18:46,960 --> 00:18:48,800 Speaker 10: bought in some other businesses and they're going to try 332 00:18:48,800 --> 00:18:53,920 Speaker 10: to swing that business from five billion ish losses into 333 00:18:53,960 --> 00:18:57,040 Speaker 10: something that's a little bit you know, four billion profitable 334 00:18:57,560 --> 00:19:01,200 Speaker 10: by twenty thirty. And that swing is really what brings 335 00:19:01,680 --> 00:19:04,840 Speaker 10: SpaceX into profitability. But the revenue at that at that 336 00:19:04,920 --> 00:19:07,960 Speaker 10: group's going to grow and be largely responsible for getting 337 00:19:07,960 --> 00:19:11,080 Speaker 10: this company to one hundred and sixty billion dollars of 338 00:19:11,160 --> 00:19:14,560 Speaker 10: revenue potentially, we think at twenty thirty. So that that's 339 00:19:14,680 --> 00:19:18,480 Speaker 10: core to I think the thesis around SpaceX right now. 340 00:19:19,200 --> 00:19:23,280 Speaker 2: Bloomberg Intelligence analyst George Ferguson with the bpsis on SpaceX. 341 00:19:23,320 --> 00:19:26,199 Speaker 2: Thank you very much to stick with SpaceX. And if 342 00:19:26,240 --> 00:19:29,240 Speaker 2: you've ever felt left out, spare a thought for Maria 343 00:19:29,359 --> 00:19:33,200 Speaker 2: Assets Securities. The South Korean firm was the only one 344 00:19:33,240 --> 00:19:38,160 Speaker 2: of twenty three underwriters to receive no stock in SpaceX's 345 00:19:38,200 --> 00:19:41,840 Speaker 2: mega IPO, no allocation at all, and Bloomberg reporting it 346 00:19:41,920 --> 00:19:44,600 Speaker 2: was all one big and for the Korean asset manager, 347 00:19:44,680 --> 00:19:49,480 Speaker 2: embarrassing mix up. Bring those baby Lipscholtzers with us, try 348 00:19:49,520 --> 00:19:52,560 Speaker 2: and explain it. There's basically a process that happens in 349 00:19:52,640 --> 00:19:55,680 Speaker 2: the weeks before an IPO, and there has been a 350 00:19:55,720 --> 00:19:57,440 Speaker 2: miscommunication in that process. 351 00:19:58,440 --> 00:20:02,000 Speaker 12: Yeah, we go back to May all these banks were 352 00:20:02,119 --> 00:20:05,480 Speaker 12: compiling orders or at least indications of interest. So when 353 00:20:05,520 --> 00:20:09,560 Speaker 12: you're talking about tapping retail, promoting this across your retail channel, 354 00:20:09,840 --> 00:20:12,240 Speaker 12: saying basically, let us know if you're interested in buying 355 00:20:12,280 --> 00:20:14,320 Speaker 12: SpaceX and how much you would want to buy or 356 00:20:14,400 --> 00:20:16,239 Speaker 12: I did that and they said, well, we have more 357 00:20:16,320 --> 00:20:19,680 Speaker 12: than a billion dollars of demand from these indication of interests. 358 00:20:19,920 --> 00:20:22,440 Speaker 12: They submit that to Goldman Sachs Morgan Stanley, the big 359 00:20:22,520 --> 00:20:26,879 Speaker 12: banks running the initial public offering, and according to Bloomberg reporting, 360 00:20:27,280 --> 00:20:29,920 Speaker 12: they thought that's all they had to do, except when 361 00:20:29,960 --> 00:20:32,960 Speaker 12: you get closer to the launch and closer to the 362 00:20:33,080 --> 00:20:36,719 Speaker 12: pricing dates in June, you need to actually show up 363 00:20:36,960 --> 00:20:40,240 Speaker 12: with an order book. So according to our reporting, the 364 00:20:40,320 --> 00:20:44,320 Speaker 12: bank never submitted actual orders on behalf of its clients. 365 00:20:44,640 --> 00:20:46,240 Speaker 2: So more than a billion dollars. 366 00:20:46,000 --> 00:20:48,960 Speaker 12: That was indicated an interest, according to Maria, never actually 367 00:20:49,119 --> 00:20:52,560 Speaker 12: was passed along formally to Goldman Sachs and Morgan Stanley. 368 00:20:52,960 --> 00:20:54,080 Speaker 2: Therefore they get zero. 369 00:20:54,280 --> 00:20:57,600 Speaker 12: It certainly was an interesting update in something a lot 370 00:20:57,640 --> 00:21:00,480 Speaker 12: of people were questioning when we saw the initial discuss. 371 00:21:01,200 --> 00:21:04,760 Speaker 2: Bailey real quick that there are serious repercussions for Mariah. 372 00:21:04,800 --> 00:21:06,960 Speaker 2: What do we know about the fallout from this. 373 00:21:07,960 --> 00:21:10,200 Speaker 12: Yeah, the regulators in South Korea are looking through what 374 00:21:10,320 --> 00:21:12,320 Speaker 12: this ultimately could mean, not only for the bank, but 375 00:21:12,440 --> 00:21:14,920 Speaker 12: kind of for the entire distribution. It's something that I 376 00:21:15,080 --> 00:21:18,560 Speaker 12: think we will see a formal kind of ruling, a 377 00:21:18,720 --> 00:21:22,280 Speaker 12: formal investigation conducted. We remain to be it remains to 378 00:21:22,320 --> 00:21:24,679 Speaker 12: be seen when that will come out, but that's certainly 379 00:21:24,760 --> 00:21:27,200 Speaker 12: that's top of mind again for the entire organization and 380 00:21:27,240 --> 00:21:29,840 Speaker 12: really for everyone who's a retail trader in South Korea. 381 00:21:31,040 --> 00:21:34,080 Speaker 2: The most Bata Lipschualtz with what is the most read 382 00:21:34,160 --> 00:21:38,080 Speaker 2: story across Bloomberg today, in no surprise, linked to SpaceX. 383 00:21:38,119 --> 00:21:40,159 Speaker 2: We're looking at shares of AT and T, T Mobile, 384 00:21:40,240 --> 00:21:44,560 Speaker 2: and Verizon in all three cases down significantly and for 385 00:21:44,680 --> 00:21:48,840 Speaker 2: a second day. Bloomberg had reported June twenty sixth that 386 00:21:49,119 --> 00:21:54,320 Speaker 2: SpaceX and Charter Communications had held high level talks about 387 00:21:54,440 --> 00:22:00,760 Speaker 2: partnering on a consumer mobile phone offering and not drawing 388 00:22:00,920 --> 00:22:05,200 Speaker 2: too close a direct or causal link. Like everything I'm 389 00:22:05,240 --> 00:22:08,800 Speaker 2: reading on the Bloomberg terminal is that the downward pressure 390 00:22:08,880 --> 00:22:11,639 Speaker 2: on these carriers is the idea that SpaceX comes in. 391 00:22:11,840 --> 00:22:15,240 Speaker 2: If those reports from Bloomberg about SpaceX working with chatter 392 00:22:15,240 --> 00:22:18,680 Speaker 2: are accurate and disrupts that space SpaceX, particularly in the 393 00:22:18,760 --> 00:22:23,960 Speaker 2: context of Starlink, already has relationships with a couple of those, 394 00:22:24,080 --> 00:22:26,399 Speaker 2: so we'll continue to track that, but it's interesting to 395 00:22:26,440 --> 00:22:29,560 Speaker 2: see the level declines. It's halftime in the show. Coming up, 396 00:22:29,640 --> 00:22:33,040 Speaker 2: Gavin Uberti of x joins us his startup coming out 397 00:22:33,080 --> 00:22:37,240 Speaker 2: of stale, taking on chip Giant's low latency approach to 398 00:22:37,359 --> 00:22:41,600 Speaker 2: AI inference. Serious money being raised out of the gate 399 00:22:42,080 --> 00:22:45,679 Speaker 2: busy for two years until this point, haven't said very 400 00:22:45,760 --> 00:22:46,680 Speaker 2: much about what they're up to. 401 00:22:47,320 --> 00:23:02,560 Speaker 13: This is Bloomberg Tech. Welcome back to Bloomberg Tech. 402 00:23:02,680 --> 00:23:07,159 Speaker 2: Semiconductors and outperformance of semiconductors is the story, particularly in 403 00:23:07,200 --> 00:23:09,520 Speaker 2: public markets right now. Then, as that one hundred is 404 00:23:09,600 --> 00:23:12,399 Speaker 2: up more than a percentage point, the Philadelphia Semiconductor Index 405 00:23:12,720 --> 00:23:15,080 Speaker 2: is up three percent. That we've seen some big swings 406 00:23:15,359 --> 00:23:18,320 Speaker 2: in recent sessions, even with the overall rally of this week. 407 00:23:18,600 --> 00:23:20,879 Speaker 2: The bigger picture point is we're in the last trading 408 00:23:20,960 --> 00:23:23,840 Speaker 2: day of June and headed for the second quarter, and 409 00:23:23,920 --> 00:23:27,159 Speaker 2: the Philadelphia Semiconductor Index or SOCKS, is on track for 410 00:23:27,240 --> 00:23:30,680 Speaker 2: its best quarter ever. It gained eighty six percent in 411 00:23:30,760 --> 00:23:33,200 Speaker 2: the last three months. It is up almost one hundred 412 00:23:33,280 --> 00:23:36,240 Speaker 2: percent year to date, although we still have a half 413 00:23:36,320 --> 00:23:39,160 Speaker 2: of the calendar year to go. Who knows what happens next. 414 00:23:39,200 --> 00:23:41,360 Speaker 2: In a little while, we're going to have a conversation 415 00:23:41,400 --> 00:23:46,800 Speaker 2: about how private markets are viewing innovation in silicon and 416 00:23:46,920 --> 00:23:50,480 Speaker 2: also great inference demand and what that's doing for chips. 417 00:23:50,680 --> 00:23:53,120 Speaker 2: We have a lot of other tech headlines to get through. 418 00:23:53,280 --> 00:23:55,200 Speaker 2: Let's get out to New York where Bloomberg's your high 419 00:23:55,280 --> 00:23:57,240 Speaker 2: and is standing in by you Higher. 420 00:23:57,160 --> 00:24:00,560 Speaker 14: High hi adis time now for talking tech. First up, 421 00:24:00,960 --> 00:24:03,720 Speaker 14: ten cents is going on the offensive. After a three 422 00:24:03,840 --> 00:24:07,919 Speaker 14: hundred billion dollars wipeout the Chinese tech giants is ramping 423 00:24:08,000 --> 00:24:10,840 Speaker 14: up share buybacks to support US thought that's lost a 424 00:24:11,000 --> 00:24:14,120 Speaker 14: third of its value since October. The move comes as 425 00:24:14,200 --> 00:24:18,159 Speaker 14: investors demand proof that ten cents massive AI pivots. 426 00:24:18,200 --> 00:24:21,240 Speaker 2: Can generate real world revenue. Plus. 427 00:24:21,320 --> 00:24:24,720 Speaker 14: The global AI boom is fueling a massive fundraising wave 428 00:24:24,880 --> 00:24:27,640 Speaker 14: in Hong Kong, with companies looking to raise more than 429 00:24:27,760 --> 00:24:31,200 Speaker 14: six billion dollars through new listings. Leading to charge is 430 00:24:31,280 --> 00:24:35,359 Speaker 14: Apple supplier lux Share, the world's largest maker of AirPods. 431 00:24:35,400 --> 00:24:38,040 Speaker 14: The company is aiming to raise up to three point 432 00:24:38,119 --> 00:24:41,360 Speaker 14: one billion dollars in what would be Hong Kong's biggest 433 00:24:41,680 --> 00:24:45,080 Speaker 14: IPO of the year. And while Hong Kong's IPO market 434 00:24:45,160 --> 00:24:48,480 Speaker 14: is booming, it's been a much tougher year for Chinese tech. 435 00:24:49,000 --> 00:24:52,000 Speaker 14: The sector is on track for its worst underperformance against 436 00:24:52,040 --> 00:24:55,320 Speaker 14: global markets since two thousand and one, with the MSCI 437 00:24:55,520 --> 00:24:59,760 Speaker 14: China Index down fifteen percent this year, analysts saying week 438 00:25:00,800 --> 00:25:04,600 Speaker 14: consumer spending in a lack of homegrown AI hardware leaders 439 00:25:04,680 --> 00:25:08,200 Speaker 14: have left China trailing the global AI rally ed. 440 00:25:08,880 --> 00:25:11,639 Speaker 2: All right, let's get to private market. Startup Etched is 441 00:25:11,680 --> 00:25:15,359 Speaker 2: emerging from self with an itold taking on chip heavyweights 442 00:25:15,520 --> 00:25:19,040 Speaker 2: like Nvidia is the industry shifts from training AI models 443 00:25:19,240 --> 00:25:22,440 Speaker 2: to running them inference. The company's raised eight hundred million 444 00:25:22,480 --> 00:25:25,520 Speaker 2: dollars in funding out of the gate, including backing from 445 00:25:25,560 --> 00:25:29,000 Speaker 2: the likes of Jane Street and TSMC linked firm Venture 446 00:25:29,080 --> 00:25:33,720 Speaker 2: Tech Alliance. At CEO Gavin Aberti joins us. Now you 447 00:25:33,920 --> 00:25:42,879 Speaker 2: have introduced a RAX scale system, very interesting, split memory design, 448 00:25:43,880 --> 00:25:46,480 Speaker 2: very focused on inference. We'll get into all the money 449 00:25:46,520 --> 00:25:48,800 Speaker 2: you raised and what you've been doing quietly for two years, 450 00:25:49,000 --> 00:25:51,639 Speaker 2: but could we start on the system. What is Etched 451 00:25:51,680 --> 00:25:53,520 Speaker 2: what is it pitching. 452 00:25:53,440 --> 00:25:56,479 Speaker 15: Or eesch building rack scale inference systems or really if 453 00:25:56,480 --> 00:25:58,359 Speaker 15: I had a dayd ago and unveiled both are funding 454 00:25:58,840 --> 00:26:00,720 Speaker 15: and two of our qure tech nowlogies that make the 455 00:26:00,800 --> 00:26:03,960 Speaker 15: tech work what we call a le voltage inference and 456 00:26:04,040 --> 00:26:07,040 Speaker 15: custure scale memory. When do you think about inference, you 457 00:26:07,119 --> 00:26:09,960 Speaker 15: think about prefill and decode. Yes, that is reading in 458 00:26:10,280 --> 00:26:13,400 Speaker 15: all the input data and there producing up those decode tokens. 459 00:26:13,800 --> 00:26:17,040 Speaker 2: We're just showing an image of the system on the screen, 460 00:26:18,280 --> 00:26:19,720 Speaker 2: the rack scale inference system. 461 00:26:19,920 --> 00:26:20,359 Speaker 11: What is it? 462 00:26:20,560 --> 00:26:22,479 Speaker 2: We're looking at break it down for us. So right 463 00:26:22,560 --> 00:26:24,200 Speaker 2: here you see one of our racks. 464 00:26:24,560 --> 00:26:26,760 Speaker 15: This thing comes pre assembled with three or two of 465 00:26:26,800 --> 00:26:29,680 Speaker 15: our chips for our first gen product, and what you 466 00:26:29,800 --> 00:26:32,320 Speaker 15: have is our culture scale memory tech linking those chips 467 00:26:32,880 --> 00:26:36,639 Speaker 15: and this technology allows for a very communication between chips 468 00:26:37,119 --> 00:26:40,120 Speaker 15: and that then allows one chip to read and use 469 00:26:40,400 --> 00:26:44,280 Speaker 15: the HBM and STRAM memory of other chips in its system. 470 00:26:44,800 --> 00:26:47,399 Speaker 2: You have a very interesting story. You've clearly been busy 471 00:26:47,960 --> 00:26:50,680 Speaker 2: for a couple of years, but you've raised a significant 472 00:26:50,720 --> 00:26:53,440 Speaker 2: amount of money from a very interesting list of back 473 00:26:53,640 --> 00:26:56,000 Speaker 2: is just reflect on that for a minute. I mean, 474 00:26:56,600 --> 00:26:59,280 Speaker 2: how have you been able to convince a Jane Streets 475 00:27:00,119 --> 00:27:03,720 Speaker 2: the effectiveness of the technology and why raise capital at 476 00:27:03,760 --> 00:27:04,160 Speaker 2: that level. 477 00:27:04,440 --> 00:27:06,399 Speaker 15: Well, I think if you look at our backers, they 478 00:27:06,440 --> 00:27:09,640 Speaker 15: are extremely technical. To go ahead and take a bet 479 00:27:09,760 --> 00:27:12,840 Speaker 15: like this on somebody like myself and my co founders, 480 00:27:13,119 --> 00:27:15,320 Speaker 15: you need to go ahead and be very first principles driven, 481 00:27:15,800 --> 00:27:18,560 Speaker 15: understand the tech very very deeply, and see why the 482 00:27:18,640 --> 00:27:20,840 Speaker 15: market is going to be so big and why this 483 00:27:20,960 --> 00:27:25,080 Speaker 15: technology is fundamentally better. And our backers like James Street 484 00:27:25,320 --> 00:27:27,879 Speaker 15: and better Check Alliance have big chip teams and they 485 00:27:27,920 --> 00:27:28,200 Speaker 15: get it. 486 00:27:29,600 --> 00:27:32,399 Speaker 2: We had reported I think that that money was raised 487 00:27:32,400 --> 00:27:35,480 Speaker 2: at around a five billion dollar valuation. But what's astonishing is, 488 00:27:35,880 --> 00:27:39,239 Speaker 2: you know, with respect the brief history of the company, right, 489 00:27:39,400 --> 00:27:41,639 Speaker 2: how long when did you come up with the idea? 490 00:27:42,280 --> 00:27:44,960 Speaker 2: What was that first couple of years of trying to 491 00:27:45,000 --> 00:27:45,600 Speaker 2: get set. 492 00:27:45,480 --> 00:27:47,520 Speaker 15: Up like well, it took a lot of work to 493 00:27:47,560 --> 00:27:49,399 Speaker 15: get to this point or at a spot the day 494 00:27:49,400 --> 00:27:51,640 Speaker 15: where we have the rack scale inference system is running 495 00:27:52,160 --> 00:27:55,520 Speaker 15: and we have this tech proven out. And while this's 496 00:27:55,560 --> 00:27:58,639 Speaker 15: a not a low evaluation, the fact that we have 497 00:27:58,720 --> 00:28:00,600 Speaker 15: a tech to back it up, I think they makes sense. 498 00:28:00,960 --> 00:28:04,920 Speaker 2: I'm really interested by the specifics of the system on 499 00:28:05,040 --> 00:28:08,760 Speaker 2: the program of late. Just by way of example, you know, 500 00:28:08,840 --> 00:28:12,720 Speaker 2: the pitch of Cerebris is its insulation from what's happening 501 00:28:12,760 --> 00:28:15,440 Speaker 2: in high bandwidth memory because it doesn't rely on Hamburg 502 00:28:15,640 --> 00:28:20,439 Speaker 2: high bandwidth Memory SRAM is what they have configured. How 503 00:28:20,480 --> 00:28:22,399 Speaker 2: does that work for h in its system. 504 00:28:22,560 --> 00:28:24,280 Speaker 15: Well, for us, we're able to go ahead and get 505 00:28:24,359 --> 00:28:27,760 Speaker 15: more out of the same volume of HBM and one 506 00:28:27,760 --> 00:28:29,199 Speaker 15: of the those we did with the technically call low 507 00:28:29,280 --> 00:28:32,200 Speaker 15: voltage inferess, we want to be able to use the 508 00:28:32,280 --> 00:28:35,440 Speaker 15: same HBM in the same s RAM to run way 509 00:28:35,560 --> 00:28:38,600 Speaker 15: way more tokens for more users. But the bottomlike you're 510 00:28:38,640 --> 00:28:41,920 Speaker 15: typically is power. If you look at a GPU, it 511 00:28:42,080 --> 00:28:46,160 Speaker 15: thermally throttles. You cannot fit more compute under that same chip. 512 00:28:46,680 --> 00:28:49,200 Speaker 15: So what we do to lower the voltage a lot. 513 00:28:49,880 --> 00:28:53,040 Speaker 15: We run under half the voltage of typical Nvidio GPUs 514 00:28:53,640 --> 00:28:55,760 Speaker 15: and as a result we're able to get very large 515 00:28:55,840 --> 00:28:58,240 Speaker 15: power savings and that in term means you can have 516 00:28:58,560 --> 00:29:02,080 Speaker 15: more users on the same and get more mileage out 517 00:29:02,080 --> 00:29:02,960 Speaker 15: of your same HBM. 518 00:29:03,000 --> 00:29:05,640 Speaker 2: Baywan, if you think about the economic benefits for the 519 00:29:05,920 --> 00:29:08,600 Speaker 2: customer the user, then you would pitch this as being 520 00:29:08,800 --> 00:29:12,320 Speaker 2: superior on an essentially a dollar per token basis. 521 00:29:12,560 --> 00:29:15,600 Speaker 15: Absolutely, we think about this as economies of scale. You 522 00:29:15,680 --> 00:29:18,440 Speaker 15: can go have more infrastructure that allows you to go 523 00:29:18,480 --> 00:29:21,880 Speaker 15: ahead and get a much lower cost producer. And as 524 00:29:21,960 --> 00:29:24,800 Speaker 15: the models get bigger, that clutter scale memory dec it's 525 00:29:24,840 --> 00:29:26,400 Speaker 15: going to give more and more of an advantage. 526 00:29:27,280 --> 00:29:30,440 Speaker 2: The world we're in is that, you know, for some 527 00:29:30,640 --> 00:29:34,320 Speaker 2: time we've been talking about better performing systems coming from 528 00:29:34,440 --> 00:29:38,520 Speaker 2: video architecture. Right in video still has a technical monopoly 529 00:29:38,600 --> 00:29:42,720 Speaker 2: in the market. What does the pathway look like for 530 00:29:42,840 --> 00:29:45,479 Speaker 2: you to scale and is there anything you can say 531 00:29:45,520 --> 00:29:48,760 Speaker 2: about sort of real world workloads that are being run 532 00:29:48,840 --> 00:29:52,440 Speaker 2: on your systems, real world revenues that you're already able 533 00:29:52,520 --> 00:29:54,720 Speaker 2: to point to as evidence that this will be out 534 00:29:54,760 --> 00:29:56,040 Speaker 2: there in a meaningful way. 535 00:29:56,480 --> 00:29:58,800 Speaker 15: Or we are praying some benuremarks and we are seeing 536 00:29:58,840 --> 00:29:59,880 Speaker 15: best in the world performance. 537 00:30:00,320 --> 00:30:03,120 Speaker 2: What gets me really excited inside of the lab environment, though. 538 00:30:03,680 --> 00:30:05,960 Speaker 15: We've had some customers to test the system. What gets 539 00:30:06,000 --> 00:30:09,840 Speaker 15: me really excited on this production that to make a 540 00:30:09,880 --> 00:30:12,040 Speaker 15: big difference you have to build a lot of these products, 541 00:30:12,480 --> 00:30:14,960 Speaker 15: and we have a world class platform in production. And 542 00:30:15,000 --> 00:30:17,720 Speaker 15: that explains the capital that you raised. It's a key 543 00:30:17,760 --> 00:30:19,560 Speaker 15: part of it. The team is such a massive part 544 00:30:19,600 --> 00:30:22,800 Speaker 15: of our story. Around half of our platform team it's 545 00:30:22,840 --> 00:30:25,280 Speaker 15: from the video. We're likely to have some of the best, 546 00:30:25,680 --> 00:30:29,560 Speaker 15: including our VP who ran in videos, DJs and HGX team. 547 00:30:30,600 --> 00:30:35,600 Speaker 2: Gaviniberti introducing h to US. CEO of ETCHED raising a 548 00:30:35,640 --> 00:30:39,040 Speaker 2: lot of money after two very quiet years building a 549 00:30:39,120 --> 00:30:42,080 Speaker 2: low voltage inference system, come back, thank you very much. Indeed, 550 00:30:42,120 --> 00:30:45,040 Speaker 2: now coming up on the show. The US has long 551 00:30:45,120 --> 00:30:49,600 Speaker 2: been a top destination for talented immigrants, but increasingly restrictive 552 00:30:49,600 --> 00:30:53,480 Speaker 2: immigration policies are changing the corporate landscape. We have a 553 00:30:53,520 --> 00:31:03,480 Speaker 2: few stories next. This has been a big tech Today, 554 00:31:03,600 --> 00:31:08,920 Speaker 2: the Supreme Court rejected President Trump's planned restrictions on birthright citizenship, 555 00:31:09,280 --> 00:31:13,200 Speaker 2: invalidating a central plank of his immigration agenda. The Court 556 00:31:13,280 --> 00:31:17,760 Speaker 2: saying the executive order couldn't be squared with the Constitution's 557 00:31:17,800 --> 00:31:22,640 Speaker 2: fourteenth Amendment, which guarantees citizenship to virtually everyone born on 558 00:31:22,840 --> 00:31:26,320 Speaker 2: US soil. Bloomberg's Washington correspondent, Tyler Kendall, was standing by 559 00:31:26,680 --> 00:31:28,920 Speaker 2: outside of the Supreme Court. What do we need to know? 560 00:31:31,520 --> 00:31:33,840 Speaker 16: Well, just to underscore the point that you made about 561 00:31:33,840 --> 00:31:37,200 Speaker 16: how this really does reject a central pillar of President 562 00:31:37,200 --> 00:31:40,360 Speaker 16: Trump's immigration agenda. I was in this very spot back 563 00:31:40,400 --> 00:31:43,240 Speaker 16: in April when President Trump became the first sitting US 564 00:31:43,360 --> 00:31:47,720 Speaker 16: president in modern history to ever attend oral arguments in person. 565 00:31:47,840 --> 00:31:51,760 Speaker 16: That is how critical the administration viewed this policy to 566 00:31:52,120 --> 00:31:55,640 Speaker 16: its broader agenda. But ultimately the decision isn't surprising, right. 567 00:31:55,720 --> 00:31:58,840 Speaker 16: We got these signals from the justices during oral arguments 568 00:31:58,880 --> 00:32:00,840 Speaker 16: that this was how this case was going to go. 569 00:32:00,920 --> 00:32:03,240 Speaker 16: In fact, President Trump himself had indicated that he didn't 570 00:32:03,280 --> 00:32:05,960 Speaker 16: think that the Justices were going to go his way. 571 00:32:06,120 --> 00:32:08,000 Speaker 16: But as you mentioned, it really does come down to 572 00:32:08,080 --> 00:32:11,040 Speaker 16: that executive order that President Trump signed on his very 573 00:32:11,120 --> 00:32:14,840 Speaker 16: first day in office, which sought to restrict automatic birthright 574 00:32:14,880 --> 00:32:17,440 Speaker 16: citizenship to those with at least one parent who is 575 00:32:17,480 --> 00:32:19,720 Speaker 16: a US citizen or Green card holder, which meant that 576 00:32:19,840 --> 00:32:22,800 Speaker 16: newborn children of parents that are on temporary visas would 577 00:32:22,880 --> 00:32:26,880 Speaker 16: not be able to be eligible. But the justices ultimately 578 00:32:26,920 --> 00:32:30,000 Speaker 16: looked at those five keywords of the citizenship clause under 579 00:32:30,120 --> 00:32:32,440 Speaker 16: the fourteenth Amendment, which says that if you were born 580 00:32:32,560 --> 00:32:36,280 Speaker 16: in the United States and quote subject to the jurisdiction thereof, 581 00:32:36,400 --> 00:32:39,720 Speaker 16: then you are a US citizen. These justices sided with 582 00:32:39,800 --> 00:32:42,600 Speaker 16: that long standing legal precedent and ed. As you well know, 583 00:32:42,680 --> 00:32:45,240 Speaker 16: it could have very broad ranging impacts for the nearly 584 00:32:45,320 --> 00:32:49,720 Speaker 16: two hundred and fifty five thousand children of undocumented immigrants 585 00:32:49,760 --> 00:32:51,520 Speaker 16: that are born in this country each year. 586 00:32:52,440 --> 00:32:54,560 Speaker 2: Being both side the Kendall, thank you very much. The 587 00:32:54,720 --> 00:32:58,240 Speaker 2: US has long been a top destination for talented immigrants. 588 00:32:58,320 --> 00:33:01,080 Speaker 2: In fact, two thirds of its tech workers in Silicon 589 00:33:01,200 --> 00:33:04,480 Speaker 2: Valley are foreign born. While the US has spent decades 590 00:33:04,560 --> 00:33:07,880 Speaker 2: relying on top ten talent from nations like India and China, 591 00:33:08,400 --> 00:33:12,640 Speaker 2: increasingly restrictive immigration policies are changing the calculus. Bloomberger Regionals 592 00:33:12,880 --> 00:33:13,600 Speaker 2: did a deep dive. 593 00:33:15,000 --> 00:33:15,920 Speaker 16: This story is. 594 00:33:16,040 --> 00:33:19,560 Speaker 11: All about the American dream. It used to be that 595 00:33:19,880 --> 00:33:23,560 Speaker 11: coming to the US was this beacon of opportunity for 596 00:33:23,720 --> 00:33:27,400 Speaker 11: skilled immigrants who wanted to pursue their fortune, and in turn, 597 00:33:27,560 --> 00:33:31,120 Speaker 11: the US received all the innovation that comes with having 598 00:33:31,200 --> 00:33:35,920 Speaker 11: top talent. But now with all the administration chaos and 599 00:33:36,080 --> 00:33:39,080 Speaker 11: the visa back and forth, that calculus is changing. 600 00:33:39,400 --> 00:33:41,680 Speaker 17: The trend we're seeing is that immigrant workers who feel 601 00:33:41,720 --> 00:33:43,880 Speaker 17: like they've done everything right to stay in this country 602 00:33:43,960 --> 00:33:46,000 Speaker 17: are now wondering if the country wants them at all. 603 00:33:46,200 --> 00:33:49,520 Speaker 17: This could affect the competitive advantages that America's held for decades. 604 00:33:51,680 --> 00:33:54,440 Speaker 17: Each year, the government issues four hundred thousand, f one 605 00:33:54,520 --> 00:33:56,640 Speaker 17: visas for foreign students who want to come to an 606 00:33:56,640 --> 00:33:59,920 Speaker 17: American university. At the same time, they offer eighty five 607 00:34:00,000 --> 00:34:02,960 Speaker 17: one thousand h ONEV visas for international talent they want 608 00:34:02,960 --> 00:34:05,120 Speaker 17: to come and work at American companies. The H one 609 00:34:05,160 --> 00:34:07,320 Speaker 17: B visa has been seen as the most viable way 610 00:34:07,440 --> 00:34:09,680 Speaker 17: to secure a green card and eventually have a life 611 00:34:09,719 --> 00:34:10,239 Speaker 17: in the US. 612 00:34:11,440 --> 00:34:16,040 Speaker 11: Two thirds of tech workers in Silicon Valley are foreign born. 613 00:34:16,360 --> 00:34:19,360 Speaker 11: Even though the US is now a powerhouse of the 614 00:34:19,520 --> 00:34:23,400 Speaker 11: AI fueled economy, it's relied for years and years and 615 00:34:23,520 --> 00:34:26,240 Speaker 11: years on a lot of skilled immigrants coming from countries 616 00:34:26,360 --> 00:34:27,640 Speaker 11: like India and China. 617 00:34:27,600 --> 00:34:30,960 Speaker 17: Making matters more complicated for skilled immigrants. The Trump administration 618 00:34:31,120 --> 00:34:32,880 Speaker 17: is flip flopping on immigration policy. 619 00:34:33,120 --> 00:34:37,080 Speaker 11: All the fifty skilled immigrants that we spoke to, that 620 00:34:37,239 --> 00:34:39,320 Speaker 11: was a sticking point for them. That was a moment 621 00:34:39,520 --> 00:34:42,320 Speaker 11: where the path that they had charted would not be 622 00:34:42,400 --> 00:34:43,240 Speaker 11: possible anymore. 623 00:34:47,480 --> 00:34:50,480 Speaker 2: You can read and watch the full report on Bloomberg 624 00:34:50,520 --> 00:34:53,640 Speaker 2: dot com more on immigration. So after the Supreme Court 625 00:34:53,719 --> 00:34:58,839 Speaker 2: officially struck down President Trump's signature restrictions on birthright citizenship 626 00:34:58,880 --> 00:35:02,560 Speaker 2: this morning with implications for the future US tech workforce, 627 00:35:02,640 --> 00:35:06,720 Speaker 2: which relies heavily on immigrant talent. As was just outline, 628 00:35:06,800 --> 00:35:10,560 Speaker 2: new data by Handshake shows the share of entry level 629 00:35:10,680 --> 00:35:15,200 Speaker 2: jobs offering visa sponsorship has plummeted to just two point 630 00:35:15,320 --> 00:35:18,359 Speaker 2: three percent this year. Joining US now to break down 631 00:35:18,640 --> 00:35:21,080 Speaker 2: what is its stake and what is happening? Hebra Am, 632 00:35:21,160 --> 00:35:25,560 Speaker 2: a partner with Ericsson Immigration Group. Welcome back to Bloomberg Tech. 633 00:35:26,160 --> 00:35:28,640 Speaker 2: I want to start with the Supreme Court decision if 634 00:35:28,719 --> 00:35:34,239 Speaker 2: I may. You know, as many would point out, those 635 00:35:34,320 --> 00:35:37,799 Speaker 2: in the technology industry from all around the world come 636 00:35:37,840 --> 00:35:40,359 Speaker 2: to the United States in part for their careers right 637 00:35:40,520 --> 00:35:43,560 Speaker 2: to seek a job in the technology sector, but also 638 00:35:43,680 --> 00:35:45,239 Speaker 2: with the hope that you know, if they have a 639 00:35:45,360 --> 00:35:48,040 Speaker 2: family here, it offers a future to their family. I 640 00:35:48,080 --> 00:35:50,439 Speaker 2: would just ask you know, based on your client base, 641 00:35:50,560 --> 00:35:54,080 Speaker 2: your experience working as an attorney in the field, what 642 00:35:54,560 --> 00:35:57,480 Speaker 2: you make of the decision by the Supreme Court this morning. 643 00:35:59,080 --> 00:36:00,640 Speaker 18: So I think that this is going to be good 644 00:36:00,680 --> 00:36:04,160 Speaker 18: news for the tech industry, the healthcare industry, financial services. 645 00:36:04,320 --> 00:36:07,480 Speaker 18: These are all industries that rely very heavily on workers 646 00:36:07,600 --> 00:36:10,839 Speaker 18: from countries that are heavily represented in the H one 647 00:36:10,960 --> 00:36:13,960 Speaker 18: B and the L one pipelines in particular. So I 648 00:36:14,040 --> 00:36:17,359 Speaker 18: think that first and foremost, this is a win for stability. 649 00:36:17,880 --> 00:36:21,720 Speaker 18: The last eighteen months have involved a tremendous amount of change, 650 00:36:22,280 --> 00:36:24,560 Speaker 18: and I think that, you know, had the decision gone 651 00:36:24,600 --> 00:36:27,680 Speaker 18: a different way than the immediate question that would have 652 00:36:27,880 --> 00:36:30,440 Speaker 18: occurred to a lot of business leaders is if the 653 00:36:30,520 --> 00:36:33,320 Speaker 18: court was willing to go that far, what other reinterpretations 654 00:36:33,360 --> 00:36:36,120 Speaker 18: would have been possible? And so, you know, with the 655 00:36:36,320 --> 00:36:40,600 Speaker 18: uncertainty itself being arguably economically damaging, I think that again, 656 00:36:40,680 --> 00:36:43,319 Speaker 18: this is a win for stability and continuity pe. 657 00:36:43,440 --> 00:36:44,840 Speaker 2: But I want to go back to that data, the 658 00:36:44,920 --> 00:36:48,880 Speaker 2: Handshake data that the share of entry level jobs offering 659 00:36:49,080 --> 00:36:52,960 Speaker 2: visa sponsorship has plummeted to just two point three percent. 660 00:36:53,040 --> 00:36:55,840 Speaker 2: You can see based on that chart the visual alone, 661 00:36:56,400 --> 00:36:59,920 Speaker 2: that you know, into twenty twenty three we're above ten 662 00:37:00,200 --> 00:37:04,080 Speaker 2: per What does that data tell you about the impact 663 00:37:04,160 --> 00:37:04,720 Speaker 2: of policy. 664 00:37:05,920 --> 00:37:09,040 Speaker 18: I mean, I think that we're seeing basically a acceleration 665 00:37:09,200 --> 00:37:12,959 Speaker 18: of the talent drain. Right, we're already seeing brain drain. 666 00:37:13,760 --> 00:37:16,880 Speaker 18: Foreign job seeker interest in the United States is at 667 00:37:16,920 --> 00:37:20,000 Speaker 18: a six year low. There are estimates that there has 668 00:37:20,080 --> 00:37:22,960 Speaker 18: been a seventeen percent drop in enrollment on the part 669 00:37:23,000 --> 00:37:27,360 Speaker 18: of international students. The economic consequences of that foreign student 670 00:37:27,480 --> 00:37:30,759 Speaker 18: enrollment drop is by some measures quantified to be over 671 00:37:30,880 --> 00:37:33,680 Speaker 18: one billion dollars. And so I think what this is 672 00:37:34,160 --> 00:37:37,560 Speaker 18: demonstrating is again the fact that interest is starting to 673 00:37:37,600 --> 00:37:42,040 Speaker 18: be directed elsewhere, but that you cannot divorce the overlap 674 00:37:42,160 --> 00:37:46,560 Speaker 18: between the immigration policies as well as their downstream economic effects. 675 00:37:46,920 --> 00:37:48,560 Speaker 18: And a lot of these trends are starting to have 676 00:37:48,719 --> 00:37:52,640 Speaker 18: economic impact on US workers and the US labor force 677 00:37:52,719 --> 00:37:53,000 Speaker 18: as well. 678 00:37:54,880 --> 00:37:58,520 Speaker 2: If you could give any case study or just sort 679 00:37:58,560 --> 00:38:02,520 Speaker 2: of illertiative exams, if you're working for a technology company 680 00:38:03,000 --> 00:38:06,360 Speaker 2: or you are representing somebody that's trying to secure a 681 00:38:06,520 --> 00:38:09,759 Speaker 2: job in the tech sector in this country, like, what 682 00:38:09,960 --> 00:38:13,360 Speaker 2: is the real world experience of those two entities in 683 00:38:13,480 --> 00:38:16,160 Speaker 2: securing a visa in this current environment. 684 00:38:17,120 --> 00:38:19,480 Speaker 18: I mean, there is a tremendous amount of scrutiny, there's 685 00:38:19,520 --> 00:38:23,759 Speaker 18: a tremendous amount of vetting. The evidentiary burden is becomingly 686 00:38:23,920 --> 00:38:27,280 Speaker 18: increasingly more strict. And then when it comes to individuals 687 00:38:27,360 --> 00:38:31,040 Speaker 18: who are from certain countries, the waytime before an individual 688 00:38:31,160 --> 00:38:34,560 Speaker 18: can convert to a legal permanent resident of the United 689 00:38:34,600 --> 00:38:39,040 Speaker 18: States is years and years, if not decades, and so 690 00:38:39,640 --> 00:38:41,680 Speaker 18: you know, this is not as simple as saying that 691 00:38:41,800 --> 00:38:45,240 Speaker 18: foreign workers are taking jobs from US workers. Foreign workers 692 00:38:45,320 --> 00:38:48,080 Speaker 18: actually have to jump through a lot of hoops to 693 00:38:48,200 --> 00:38:51,360 Speaker 18: be able to achieve their immigration goals. And had this 694 00:38:51,560 --> 00:38:53,960 Speaker 18: decision gone the other way, I think that there would 695 00:38:54,000 --> 00:38:57,160 Speaker 18: have been a segment of that high skilled foreign talent 696 00:38:57,280 --> 00:39:03,120 Speaker 18: demographic that would have seriously, you know, rethought the decision 697 00:39:03,160 --> 00:39:06,040 Speaker 18: to choose, you know, the United States as their permanent home. 698 00:39:07,719 --> 00:39:11,080 Speaker 2: What are the technology companies doing. Are they spending more 699 00:39:11,200 --> 00:39:15,680 Speaker 2: to secure that talent? Are they lobbying hard to have 700 00:39:15,880 --> 00:39:16,720 Speaker 2: policy change. 701 00:39:17,800 --> 00:39:19,719 Speaker 18: I think companies just want to be able to hire 702 00:39:19,760 --> 00:39:22,239 Speaker 18: who they want to hire, and when it comes down 703 00:39:22,280 --> 00:39:25,400 Speaker 18: to finding the right talent, they're interested in onboarding that 704 00:39:25,520 --> 00:39:28,920 Speaker 18: individual as quickly as possible, regardless of where that individual 705 00:39:29,040 --> 00:39:31,799 Speaker 18: is from. The Industries in which a lot of tech 706 00:39:31,840 --> 00:39:35,880 Speaker 18: companies operate are becoming increasingly competitive. It always feels like 707 00:39:35,960 --> 00:39:38,560 Speaker 18: it's a race to innovate all the time, and so 708 00:39:38,680 --> 00:39:41,439 Speaker 18: companies just want to be able to hire without any 709 00:39:41,520 --> 00:39:44,640 Speaker 18: sort of you know, burden or hindrance so that they 710 00:39:44,680 --> 00:39:47,440 Speaker 18: can remain competitive. And so when they hone in on 711 00:39:47,680 --> 00:39:51,040 Speaker 18: a qualified candidate. They want to be able to do 712 00:39:51,200 --> 00:39:52,920 Speaker 18: whatever it takes to be able to bring them on. 713 00:39:53,480 --> 00:39:56,680 Speaker 18: The more barriers that are presented in a company, a 714 00:39:56,840 --> 00:40:00,480 Speaker 18: US company's ability to do that, the further advantage creates 715 00:40:00,560 --> 00:40:03,719 Speaker 18: for companies outside of the United States. And so you know, 716 00:40:04,000 --> 00:40:08,440 Speaker 18: what I've observed is that companies, while the overall demand 717 00:40:08,760 --> 00:40:13,400 Speaker 18: for you know, foreign talent has decreased slightly, because companies 718 00:40:13,520 --> 00:40:16,480 Speaker 18: seem to be hiring less in general. Once they decide 719 00:40:16,520 --> 00:40:19,399 Speaker 18: that there's a particular candidate that fits their need, they're 720 00:40:19,440 --> 00:40:21,319 Speaker 18: really willing to do whatever it takes to be able 721 00:40:21,360 --> 00:40:23,600 Speaker 18: to bring them on because the stakes are so high. 722 00:40:24,880 --> 00:40:28,320 Speaker 2: Piva am ver ericson Immigration Group on what's happening with 723 00:40:28,480 --> 00:40:32,080 Speaker 2: the technology workforce with immigration policy and this morning Supreme 724 00:40:32,120 --> 00:40:34,840 Speaker 2: Court decision. Thank you very much. Indeed, now coming up, 725 00:40:35,000 --> 00:40:40,200 Speaker 2: media conglomerate Comcast has plans to split up NBC Universal 726 00:40:40,239 --> 00:40:42,319 Speaker 2: and Sky from the rest of its business. It's been 727 00:40:42,400 --> 00:40:46,000 Speaker 2: off is meant to allow for more strategic flexibility. We 728 00:40:46,239 --> 00:40:46,920 Speaker 2: have more details. 729 00:40:47,000 --> 00:40:47,160 Speaker 8: Next. 730 00:40:47,280 --> 00:41:02,520 Speaker 2: This is Blomberg Tech media giant Comcast as plans to 731 00:41:02,640 --> 00:41:06,640 Speaker 2: split up NBC Universe and Sky it's European media unit 732 00:41:07,040 --> 00:41:11,120 Speaker 2: into standalone companies. The strategic move is intended to set 733 00:41:11,200 --> 00:41:13,360 Speaker 2: the company up with more deals down the line, and 734 00:41:13,440 --> 00:41:15,840 Speaker 2: it comes off to Comcast lost out on the bidding 735 00:41:15,920 --> 00:41:19,359 Speaker 2: war for Warner Brothers months prior. Blimbus Kelsey Griffiths has 736 00:41:19,400 --> 00:41:22,440 Speaker 2: been all over it. I kind of have two big questions, 737 00:41:22,480 --> 00:41:26,120 Speaker 2: which is why, you know, why split up the company 738 00:41:26,200 --> 00:41:28,360 Speaker 2: in that way? But also it kind of feels like 739 00:41:29,080 --> 00:41:32,839 Speaker 2: the era of the big conglomerate is kind of over. 740 00:41:34,640 --> 00:41:35,080 Speaker 2: That's right. 741 00:41:35,160 --> 00:41:38,280 Speaker 6: I think we've seen Comcast sort of following this playbook 742 00:41:38,400 --> 00:41:40,680 Speaker 6: that has come before them. We saw AT and T 743 00:41:41,480 --> 00:41:45,719 Speaker 6: try the big media conglomerate strategy and they ended up 744 00:41:45,719 --> 00:41:47,759 Speaker 6: selling off Warner Media a couple of years ago. So 745 00:41:48,480 --> 00:41:51,640 Speaker 6: Comcast it's their time now. They have decided that the 746 00:41:51,719 --> 00:41:56,080 Speaker 6: market would prefer pure play media, pure play broadband companies, 747 00:41:56,200 --> 00:41:58,719 Speaker 6: and that's a strategy that they're going to go forward with. 748 00:41:59,200 --> 00:42:01,239 Speaker 6: I think one of the reasons that this might be 749 00:42:01,480 --> 00:42:03,400 Speaker 6: strategic for them in the short term is that they 750 00:42:03,400 --> 00:42:06,759 Speaker 6: don't want to see a bad broadband quarter wayne on 751 00:42:06,880 --> 00:42:09,800 Speaker 6: an otherwise good quarter for say the box office or 752 00:42:10,239 --> 00:42:13,680 Speaker 6: their theme parks. So that's potentially part of the thinking there. 753 00:42:15,840 --> 00:42:19,600 Speaker 2: I'm very interested in what happens to Brian Roberts, the 754 00:42:19,800 --> 00:42:23,080 Speaker 2: chairman and co CEO of Comcast, somebody that I've tried 755 00:42:23,160 --> 00:42:26,920 Speaker 2: to interview several times. I would love to speak to him. So, 756 00:42:27,000 --> 00:42:32,040 Speaker 2: if you're watching, mister Roberts, if this plan proceeds. 757 00:42:34,760 --> 00:42:38,160 Speaker 6: Yeah, So I got to speak with the executive team 758 00:42:38,280 --> 00:42:41,680 Speaker 6: yesterday and they were telling me that Brian Roberts has 759 00:42:41,760 --> 00:42:44,800 Speaker 6: no plans to retire anytime soon. He is going to 760 00:42:44,880 --> 00:42:47,879 Speaker 6: take a step back from being co chairman or sorry, 761 00:42:47,960 --> 00:42:51,120 Speaker 6: co CEO of Comcast. We don't know yet what he 762 00:42:51,280 --> 00:42:53,799 Speaker 6: is going to do in terms of his chairmanship role. 763 00:42:53,840 --> 00:42:57,120 Speaker 6: Will he stick around as chairman of both of the 764 00:42:57,160 --> 00:42:59,759 Speaker 6: split companies of just one of them, We're not sure, 765 00:43:00,000 --> 00:43:03,160 Speaker 6: but he does say he plans to be involved on 766 00:43:03,360 --> 00:43:05,680 Speaker 6: some level, much in the same way that he has 767 00:43:05,760 --> 00:43:06,440 Speaker 6: been in the past. 768 00:43:07,320 --> 00:43:09,920 Speaker 2: Kelsey is shifting to another story really quickly. A lot 769 00:43:09,960 --> 00:43:12,600 Speaker 2: of the carriers are under pressure. You reported on the 770 00:43:12,640 --> 00:43:16,200 Speaker 2: twenty six SpaceX is talking with Charter that seems to 771 00:43:16,280 --> 00:43:17,640 Speaker 2: be account for some of that pressure. 772 00:43:19,520 --> 00:43:21,839 Speaker 6: Yes, this has been a really interesting thing to watch. 773 00:43:21,960 --> 00:43:24,839 Speaker 6: About a year ago we saw SpaceX by a lot 774 00:43:24,920 --> 00:43:28,160 Speaker 6: of spectrum from EchoStar and at the time, they were 775 00:43:28,200 --> 00:43:31,200 Speaker 6: saying they really just wanted to do this to boost 776 00:43:31,239 --> 00:43:35,160 Speaker 6: their direct to sell offerings. Since then, SpaceX has obviously 777 00:43:35,160 --> 00:43:38,640 Speaker 6: gone through a massive IPO and it has started edging 778 00:43:38,760 --> 00:43:43,680 Speaker 6: closer to potentially setting up plans for a standalone mobile carrier. 779 00:43:44,000 --> 00:43:46,160 Speaker 6: And I think that's what's interesting about this report. This 780 00:43:46,320 --> 00:43:48,800 Speaker 6: is some of the first concrete evidence we've seen that 781 00:43:49,000 --> 00:43:50,640 Speaker 6: this is potentially on the table. 782 00:43:51,719 --> 00:43:55,360 Speaker 2: The most Kelsey Griffith, thank you very much. Indeed, that 783 00:43:55,440 --> 00:43:58,759 Speaker 2: does it for this edition a Bloomberg Tech really packed show. 784 00:43:58,920 --> 00:44:01,239 Speaker 2: Lots to recap on the podcast. You know where to 785 00:44:01,320 --> 00:44:03,840 Speaker 2: find it on the Bloomberg terminal as well as online 786 00:44:03,880 --> 00:44:08,480 Speaker 2: on Apple, Spotify and iHeart from San Francisco. This is 787 00:44:08,520 --> 00:44:09,279 Speaker 2: Bloomberg Tech.