WEBVTT - Trump Wants Big Tech to Pay for Power 0:00:00.280 --> 0:00:13.760 Bloomberg Audio Studios, podcasts, radio news. Bloomberg Tech is alive 0:00:13.800 --> 0:00:17.599 from coast to coast with Caroline Hyde in New York 0:00:17.920 --> 0:00:21.040 and Ed Lovelow in San Francisco. 0:00:23.200 --> 0:00:26.720 This is Bloomberg Tech coming up, Bloomberg Reporting. President Trump 0:00:26.760 --> 0:00:30.479 moves to have tech companies pay for surging energy prices 0:00:30.520 --> 0:00:33.920 following the rise of AI data centers plus AI coding 0:00:33.960 --> 0:00:37.720 startup RECLT could hit evaluation of nine billion dollars in 0:00:37.760 --> 0:00:40.080 a new round of funding. We have the details, and 0:00:40.120 --> 0:00:42.480 we'll be joined by the Chairman of the Federal Trade 0:00:42.479 --> 0:00:46.720 Commission for a conversation on antitrust, consumer protection, and AI 0:00:46.840 --> 0:00:51.320 regulation that is later this hour. Our top story to 0:00:51.400 --> 0:00:55.000 Trump administration and governors from Northeastern states are taking an 0:00:55.120 --> 0:00:59.320 unprecedented step to get tech companies to pay for surging 0:00:59.400 --> 0:01:02.720 energy price is as data centers gobble up more electricity, 0:01:02.760 --> 0:01:06.760 the administration will direct the country's largest grid operator, PJAM 0:01:06.840 --> 0:01:11.080 Interconnection to hold an emergency power auction. That's according to 0:01:11.120 --> 0:01:13.959 a White House official. Those companies will be able to 0:01:13.959 --> 0:01:17.960 bid on fifteen year contracts for new electricity generation, which 0:01:17.959 --> 0:01:22.080 could support potentially fifteen billion dollars worth of new power plants. 0:01:22.200 --> 0:01:25.960 To fuel the AI boom. Also, according to the official Bloomberg, 0:01:25.959 --> 0:01:28.280 Senior Tech editor Mike Shepherd joins us, Now, what do 0:01:28.319 --> 0:01:30.080 we need to know here, shep, I mean, this is 0:01:30.240 --> 0:01:33.440 very much in the consciousness of this administration, but also 0:01:33.560 --> 0:01:36.199 at the forefront of what the hyperscalers are thinking about too. 0:01:37.520 --> 0:01:39.440 Bill really is in a lot of ways, while the 0:01:39.640 --> 0:01:42.679 administration is trying to find a way to make the 0:01:42.680 --> 0:01:45.240 tech companies pay for some of this new power that 0:01:45.319 --> 0:01:48.960 they would need to sustain and keep the AI boom going, 0:01:49.440 --> 0:01:52.240 they are also giving the tech industry, in large part, 0:01:52.320 --> 0:01:56.440 especially the biggest companies the hyperscalers, what they've wanted. They 0:01:56.440 --> 0:01:59.560 are willing to pay for some of their extra power generation, 0:02:00.080 --> 0:02:03.560 but they need somehow the wherewithal from the grid operators 0:02:03.560 --> 0:02:07.040 themselves to add that infrastructure. So this would provide some 0:02:07.160 --> 0:02:10.360 stability in the funding, and it would also provide a 0:02:10.520 --> 0:02:13.800 slug of money for the grid operator in this case 0:02:13.880 --> 0:02:17.520 PGM Interconnection, to start laying the groundwork for some of 0:02:17.560 --> 0:02:21.440 that infrastructure. Now, what's interesting about this is PGM actually 0:02:21.600 --> 0:02:25.440 runs these kinds of so called emergency auctions on a 0:02:25.480 --> 0:02:28.520 somewhat regular basis, and they almost have one in the 0:02:28.560 --> 0:02:32.400 pipeline for this very need for the question of data centers, 0:02:32.400 --> 0:02:36.240 but they usually go on twelve month intervals, not fifteen 0:02:36.320 --> 0:02:39.960 year intervals, so this really would provide pricing stability for 0:02:40.120 --> 0:02:43.880 those buyers. Now, one other question, though ed, is how 0:02:44.000 --> 0:02:47.560 the grid operators will proceed with this, and PGM was 0:02:47.600 --> 0:02:50.160 not invited to the White House event. We are expecting 0:02:50.200 --> 0:02:54.519 an announcement shortly from Interior Secretary Doug Bergham and Energy 0:02:54.560 --> 0:02:58.639 Secretary Chris Right, and it's unclear exactly how enthusiastically they 0:02:58.680 --> 0:02:59.600 will be embracing this. 0:03:01.280 --> 0:03:04.320 The other big news overnight, the United States and Taiwan 0:03:04.600 --> 0:03:08.320 have reached a deal to lower tariffs on some goods 0:03:08.639 --> 0:03:12.960 but critically cooperate and co invest in semiconductor capacity. 0:03:13.000 --> 0:03:14.160 What do we need to know, they might. 0:03:15.360 --> 0:03:18.280 Well, what's really key in this deal is that not 0:03:18.400 --> 0:03:21.000 only are we seeing the tariff rate on goods from 0:03:21.000 --> 0:03:24.280 Taiwan dropped to fifteen percent from the current twenty percent, 0:03:25.040 --> 0:03:29.400 just as importantly, we are seeing the outlines of a more, 0:03:30.760 --> 0:03:35.840 much more significant investment in American semiconductor of manufacturing. And 0:03:35.960 --> 0:03:38.600 the announcement from the White House didn't name any names 0:03:38.640 --> 0:03:41.080 in terms of companies, but you and I our heads 0:03:41.120 --> 0:03:44.360 are going to go right to Taiwan Semiconductor Manufacturing Company. 0:03:44.600 --> 0:03:47.760 They are the top maker of AI chips and they 0:03:47.800 --> 0:03:51.400 are really the leading edge of vanguard of tai chip 0:03:51.400 --> 0:03:54.440 production in Taiwan. And under this deal, we have been 0:03:54.520 --> 0:04:00.800 told by sources, TSMC would commit to building for additional 0:04:01.000 --> 0:04:06.560 semiconductor manufacturing plants here in Arizona are through the twenty thirties. 0:04:06.560 --> 0:04:08.520 And that's on top of six plants that they have 0:04:08.600 --> 0:04:11.920 already planned to build in two advanced packaging pros. So 0:04:12.040 --> 0:04:16.040 we're looking for additional investment and the total added investment 0:04:16.160 --> 0:04:19.520 would be as much as one hundred billion dollars, and 0:04:19.560 --> 0:04:22.560 that is what we're hearing from Comerace Secretary Howard Lutnek. 0:04:23.200 --> 0:04:26.159 TSMC's US sisters shares hitting record highs on track for 0:04:26.200 --> 0:04:27.840 their best week since June, and it has a lot 0:04:27.880 --> 0:04:30.880 of momentum to do with their earnings print as well. 0:04:30.920 --> 0:04:33.560 But of course cooperation with the US members. Mike Sheppard, 0:04:33.600 --> 0:04:35.720 thank you very much. Both of those stories that that 0:04:35.839 --> 0:04:38.520 might just outlined are playing out in markets. Let's start 0:04:38.520 --> 0:04:42.760 with the energy auction piece. First, Utilities probably the second 0:04:42.839 --> 0:04:45.440 biggest decliner as a group within the s and P 0:04:45.560 --> 0:04:49.000 five hundred SMP Utilities Index down seven ten percent, and 0:04:49.000 --> 0:04:52.280 within that you see single names like Constellation, like Vistra 0:04:52.680 --> 0:04:56.560 with very market almost double digit declines in the session. 0:04:56.960 --> 0:05:00.960 The big story on semiconductors still is the bottleneck that 0:05:01.080 --> 0:05:03.880 is memory. But the bottleneck in memory drives up the 0:05:03.880 --> 0:05:07.760 main players. Micron again another name in equity markets. So 0:05:07.839 --> 0:05:10.200 this pushing record highs. It is on track for its 0:05:10.320 --> 0:05:14.720 eighth straight week of gains, eight straight weekly gain. That's 0:05:14.720 --> 0:05:17.359 his best run since twenty sixteen. 0:05:17.839 --> 0:05:18.640 Price is a high. 0:05:18.720 --> 0:05:21.120 Supply is tight, and that's good when you're one of 0:05:21.120 --> 0:05:23.640 the main players. Let's talk more about markets. Martin Norton, 0:05:24.000 --> 0:05:26.760 mpower chief investment strategist joins us now, and I want 0:05:26.760 --> 0:05:28.120 to start with the memory bottleneck. 0:05:28.920 --> 0:05:31.599 It seems to me to be severe justin in the 0:05:31.600 --> 0:05:32.120 control room. 0:05:32.200 --> 0:05:35.719 Let's get that chart ready on DDR four, because as 0:05:35.720 --> 0:05:38.159 we enter the new year, this has become a great 0:05:38.200 --> 0:05:41.240 story for all parts of the data center supply chain. 0:05:41.520 --> 0:05:43.160 How are you looking into that, Marta. 0:05:44.080 --> 0:05:46.440 Well, I think it seems like as we're entering twenty 0:05:46.520 --> 0:05:49.840 twenty six, we're reaching that phase that we had anticipated 0:05:50.440 --> 0:05:54.839 within the AI life cycle, and that's the bottleneck phase 0:05:54.960 --> 0:05:58.640 where the building and the demand and the supply are 0:05:59.480 --> 0:06:03.039 exponential moving higher, and that's putting us at a point 0:06:03.360 --> 0:06:07.440 where we're facing constraints and running headlong into them every 0:06:07.440 --> 0:06:10.480 which way we turn. And I think the memory element 0:06:10.720 --> 0:06:13.279 is a big one, not just because it relates to 0:06:13.320 --> 0:06:17.640 the AI trade, but because it has ramifications outside of AI. 0:06:18.200 --> 0:06:20.520 So I think that's something that we're going to need 0:06:20.720 --> 0:06:23.440 to watch in earning season to get a sense for 0:06:23.480 --> 0:06:26.400 what that looks like, not just for the AI trade, 0:06:26.400 --> 0:06:28.880 but for some of these other areas like PC's and 0:06:28.920 --> 0:06:31.240 get a sense for how folks are navigating that. 0:06:32.800 --> 0:06:36.360 There are many still grappling with the demand supply supply 0:06:36.440 --> 0:06:37.239 demand equation. 0:06:37.640 --> 0:06:39.560 Three things have happened. We went through memory. 0:06:40.279 --> 0:06:45.840 PGM actually downgraded and cut its peak time demand forecasts 0:06:46.040 --> 0:06:48.040 because a lot of the data centers aren't actually being 0:06:48.080 --> 0:06:51.080 built right. There are issues with construction and actually securing 0:06:51.120 --> 0:06:54.480 electricity supply, and then there's the actions of this administration. 0:06:55.080 --> 0:06:56.960 Is that any of that giving you kind of paused 0:06:57.120 --> 0:07:00.800 for thought on how intact cycle. 0:07:00.600 --> 0:07:06.000 Is well, I guess, just taking say the downgrade of 0:07:06.240 --> 0:07:09.520 energy there, I think one consideration is that timelines have 0:07:09.640 --> 0:07:11.400 always been in flux. When you look at what it 0:07:11.440 --> 0:07:14.440 takes to build a data center, the permitting that goes 0:07:14.480 --> 0:07:17.200 along with it, the construction, all the different elements. It's 0:07:17.200 --> 0:07:21.000 a multi year proposition, and so it's no surprise that 0:07:21.240 --> 0:07:24.800 it's hard to really understand when exactly the data center 0:07:24.920 --> 0:07:27.440 is coming online. So I think, you know, the progression 0:07:27.520 --> 0:07:29.680 creates clarity. I don't know if that's a change to 0:07:29.720 --> 0:07:32.000 the AI narrative. And then of course you have President 0:07:32.080 --> 0:07:36.480 Trump and the affordability issues running headlong into the AI trade, 0:07:36.680 --> 0:07:39.520 so another factor to consider. But I guess what we'll 0:07:39.520 --> 0:07:43.400 be looking for in earning season is does the boondazzel continue? Yes, 0:07:43.480 --> 0:07:46.600 we have all these factors that I think create moments 0:07:46.600 --> 0:07:49.760 of doubt for investors over the course of twenty twenty six, 0:07:50.120 --> 0:07:52.840 But are we seeing the boondazzel continue? Can we take 0:07:52.880 --> 0:07:56.000 comfort from the idea like we did with with Taiwan Semi, 0:07:56.640 --> 0:08:00.400 that Capex continues, that revenues continue, And that's what I 0:08:00.440 --> 0:08:02.400 think is really the focus as we as we move 0:08:02.440 --> 0:08:03.680 into Q four earning season. 0:08:04.400 --> 0:08:07.560 Let's go back to our top story, Bloomberg reporting that 0:08:07.760 --> 0:08:10.920 the White House, in association with governors from the northeast 0:08:11.000 --> 0:08:14.680 wants to force an auction for the data center operators. 0:08:15.760 --> 0:08:18.160 We talk about the heavy hand of the White House 0:08:18.240 --> 0:08:20.760 or Washington, but how heavy is that hand? 0:08:20.800 --> 0:08:23.240 To your mind, Well, it's. 0:08:23.400 --> 0:08:26.480 We're still I think somewhat I mean, broadly speaking around 0:08:26.480 --> 0:08:29.440 the affordability issues and a trial balloon phase right where 0:08:29.440 --> 0:08:32.480 different ideas are suggested and we see how the market 0:08:32.520 --> 0:08:36.240 anticipates and how we actually implement this those ideas. This 0:08:36.280 --> 0:08:39.760 one seems a little bit more fully baked, simply because 0:08:39.760 --> 0:08:42.560 it's coming not just from truth Social but from the 0:08:42.600 --> 0:08:48.600 administration officials more broadly, and it's you know, to Bloomberg's reporting, 0:08:49.080 --> 0:08:53.199 it's not just focused on the affordability issue, but really 0:08:53.200 --> 0:08:55.760 directed at the heart of the need of AI, which 0:08:55.800 --> 0:08:58.800 is bringing energy online. So it seems as though it's 0:08:58.840 --> 0:09:01.800 serving multiple per P says. What will be interesting to 0:09:01.840 --> 0:09:07.120 see is how the AI providers respond to this. Is 0:09:07.200 --> 0:09:10.480 this considered a good news, does it bring on and 0:09:10.640 --> 0:09:13.840 leieve some constraint for them, or is there frustration on 0:09:13.880 --> 0:09:15.680 what it could mean for their margins? And I think 0:09:15.679 --> 0:09:16.800 that's something we're going to need. 0:09:16.720 --> 0:09:18.440 To watch your margins. 0:09:18.440 --> 0:09:21.319 We'll find out about, as you said, during earnings, which 0:09:21.400 --> 0:09:24.720 start in earnest next week, can we discuss that a 0:09:24.720 --> 0:09:27.760 bit more marked to finish. You know, where are you 0:09:27.840 --> 0:09:31.559 looking within the earnings window? What kind of data points 0:09:31.600 --> 0:09:33.719 are going to drive your thesis for this coming year 0:09:33.720 --> 0:09:34.880 in the tech sector. 0:09:35.559 --> 0:09:38.600 Well, we're in an AI moment of doubt as we speak. 0:09:38.640 --> 0:09:41.480 I mean, really since Oxtober twenty ninth, we've seen this 0:09:41.679 --> 0:09:44.960 kind of malaise hit AI names as folks start to 0:09:44.960 --> 0:09:47.319 worry about some of the very issues we're discussing, and 0:09:47.840 --> 0:09:51.160 also including things like circular financing and all the big 0:09:51.160 --> 0:09:53.840 spending that we're seeing. And so I think at a 0:09:54.000 --> 0:09:56.320 very high level, we're still going to look for those 0:09:56.320 --> 0:09:59.240 things around revenue, for those things around cafex, to just 0:09:59.240 --> 0:10:02.760 get a sense for whether the company's confidence remains and 0:10:02.760 --> 0:10:06.480 whether they're continuing to improve their collection of rents from 0:10:06.559 --> 0:10:08.600 this space. I also think we're going to want to 0:10:08.600 --> 0:10:11.560 look more broadly at the overall market and get a 0:10:11.640 --> 0:10:14.960 sense where are we seeing AI implementation? And I think 0:10:15.000 --> 0:10:19.319 that can mean looking at what folks are spending on clouds, costs, 0:10:19.360 --> 0:10:21.680 looking at what they're spending on software. Are we beginning 0:10:21.720 --> 0:10:25.160 to see more than just general purpose implementation? I think 0:10:25.160 --> 0:10:27.280 that's something that we're going to need to watch as well. 0:10:27.360 --> 0:10:29.960 So I think in this earning season it's really casting 0:10:29.960 --> 0:10:32.839 a wide net and as always listening to guidance and 0:10:33.160 --> 0:10:36.080 how companies are planning to navigate the challenges that they're facing. 0:10:37.040 --> 0:10:39.200 Martin Awson, Empire, thank you very much. 0:10:39.760 --> 0:10:42.800 Now coming up, AI coding startup Replic could be reaching 0:10:42.840 --> 0:10:46.600 evaluation of nine billion dollars. We have more in the reporting. Next, 0:10:47.000 --> 0:10:59.240 this is Bloomberg Tech. Open Ai and Microsoft fail to 0:10:59.320 --> 0:11:02.400 avoid a try over. Elon Musk claims that open ai 0:11:02.920 --> 0:11:06.200 betrayed its founding mission as a public charity when it 0:11:06.200 --> 0:11:09.199 took billions in funding from the software giants and made 0:11:09.240 --> 0:11:12.000 plans to operate as a for profit business. The case 0:11:12.080 --> 0:11:16.160 was ordered to proceed to a jury trial in late April. 0:11:17.600 --> 0:11:20.600 Another top story AI coding startup replet It's nearing a 0:11:20.640 --> 0:11:22.880 deal for a new round of funding that would roughly 0:11:22.960 --> 0:11:27.120 triple its valuation to nine billion dollars. That's according to sources. 0:11:27.280 --> 0:11:30.880 Let's get the details with bloomberg'sventure capital reporter Natasha Mascaraine 0:11:30.960 --> 0:11:32.400 is part of the team that broke the story. 0:11:32.600 --> 0:11:33.480 Let's start with the round. 0:11:33.520 --> 0:11:36.920 What do we know about the size, Who's participating and 0:11:36.960 --> 0:11:39.800 the emphasis on that big jump in valuation totally. 0:11:39.880 --> 0:11:42.400 So the last time we reported that replat was raising 0:11:42.440 --> 0:11:44.880 it was only four months ago in September, at a 0:11:44.880 --> 0:11:48.240 three billion valuation. Now we're hearing that it's at a 0:11:48.360 --> 0:11:52.040 nine billion valuation, gone to raise four hundred million from 0:11:52.040 --> 0:11:56.640 an existing investor, led round Georgian, and we're not surprised. 0:11:56.720 --> 0:11:59.040 Back to back fundraising is more common than not an 0:11:59.080 --> 0:12:01.640 AI startup to these, and Replet as a coding company 0:12:01.960 --> 0:12:02.640 is no exception. 0:12:02.960 --> 0:12:04.880 It is more common than not, and so is a 0:12:04.920 --> 0:12:08.760 discussion around vibe coding and the different coding platforms that 0:12:08.800 --> 0:12:11.080 are out there. I guess the best place to go 0:12:11.160 --> 0:12:14.400 next is is why there's an attraction to replt and 0:12:14.400 --> 0:12:16.120 why it's different from the others. 0:12:16.400 --> 0:12:16.559 Yeah. 0:12:16.640 --> 0:12:19.240 When I think about coding startups, I think about two buckets. 0:12:19.240 --> 0:12:21.640 I think about the ones that are focusing on selling 0:12:21.960 --> 0:12:26.400 to enterprises, the big corporations. I think that's cursor and anthropic. 0:12:26.760 --> 0:12:28.640 Then there's a second bucket, and that's where I really 0:12:28.679 --> 0:12:32.720 place Replet. It's focusing on getting the non developer to 0:12:32.920 --> 0:12:35.679 use vibe coding at their organization, and it's a big 0:12:35.720 --> 0:12:38.600 theme I'm hearing as I asked CEOs what they're thinking 0:12:38.600 --> 0:12:41.240 about for twenty twenty six. They want you to prototype 0:12:41.720 --> 0:12:44.480 and not be living in your docs and decks too much. 0:12:44.880 --> 0:12:47.240 So when I think about investor interest in Replet, it's 0:12:47.280 --> 0:12:50.560 really a bet on getting the non technical coders start 0:12:50.640 --> 0:12:53.160 shipping apps and trying it themselves. 0:12:53.360 --> 0:12:57.640 The Attasha Mascarinus making the Bloomberg Tech debut starting twenty 0:12:57.679 --> 0:12:59.960 twenty six strong, what do we expect? What is the 0:13:00.240 --> 0:13:05.240 Replet round signal for the fundraising environment for AI? My 0:13:05.360 --> 0:13:06.839 hunch is that you're going to be pretty busy. 0:13:07.000 --> 0:13:08.400 It's going to be a very busy year. 0:13:08.679 --> 0:13:08.880 You know. 0:13:08.920 --> 0:13:10.840 When I think about what used to be the standard 0:13:10.840 --> 0:13:12.720 in Silicon Valley, I used to think you need an 0:13:12.760 --> 0:13:15.960 outside investor to value each new round, to give you 0:13:16.040 --> 0:13:18.079 sort of a new mark in Silicon Valley of being 0:13:18.120 --> 0:13:21.920 deserving of that valuation. For me, now that's a retired concept. 0:13:21.960 --> 0:13:26.040 Seeing Georgian come in and back an existing investment at 0:13:26.040 --> 0:13:29.120 that higher evaluation is considered them making a smart and 0:13:29.160 --> 0:13:32.680 savvy move with asymmetric information. So I expect to see 0:13:32.720 --> 0:13:36.320 a lot more existing investors marking up their stakes, looking 0:13:36.320 --> 0:13:39.440 to double down on their big bets and get bigger 0:13:39.520 --> 0:13:41.840 chunks of those startups, because we definitely know they're going 0:13:41.880 --> 0:13:44.959 to be raising more and maybe diluting that ownership as 0:13:45.000 --> 0:13:45.600 time goes on. 0:13:45.920 --> 0:13:49.360 Bloomberg's and Tash Mascarinus thank you. Another story in the 0:13:49.360 --> 0:13:53.560 world of private markets, German investor Digital Transformation Capital Partners 0:13:53.880 --> 0:13:56.480 is gearing up to raise what would be Europe's largest 0:13:56.480 --> 0:13:59.800 ever venture capital fund focused on defense startups, with the 0:13:59.800 --> 0:14:03.319 target of about five hundred and eighty million US dollars. 0:14:03.760 --> 0:14:07.560 Tech reporter Christina Kuriosoglu has been following the story. Joins 0:14:07.640 --> 0:14:10.680 us now from Berlin. Tell me about the fund, you know, 0:14:10.760 --> 0:14:14.280 the size is the headline, but the players and what 0:14:14.320 --> 0:14:15.839 they hope to do with the money raised. 0:14:18.120 --> 0:14:18.480 Yeah. 0:14:18.559 --> 0:14:24.000 So TDCP is a firm that's Hamburg based and they 0:14:24.480 --> 0:14:29.600 will be looking at supply chains, materials, robotics and all 0:14:29.680 --> 0:14:33.800 kinds of things that are concerned with the defense infrastructure 0:14:33.880 --> 0:14:38.240 in Europe. And you know, the process here, it's been 0:14:38.400 --> 0:14:44.200 pretty slow. Russia's full scale invasion into Ukraine happened nearly 0:14:44.280 --> 0:14:48.160 four years ago, and just now we slowly see more 0:14:48.200 --> 0:14:51.320 and more funds being raised across Europe that are looking 0:14:51.360 --> 0:14:55.120 at at the tech Even though that conflict has shown 0:14:55.320 --> 0:15:00.640 how important drones become and what kind of like modern warfare. 0:15:00.720 --> 0:15:02.000 Questions have come. 0:15:01.880 --> 0:15:05.240 Up that there are companies and telling out there, but 0:15:05.320 --> 0:15:07.080 in your reporting, do you get a sense for why 0:15:07.120 --> 0:15:09.520 it's been so slow if the techs there in the 0:15:09.520 --> 0:15:13.360 capital is finally there as well, Yes. 0:15:13.440 --> 0:15:16.600 So a couple of reasons, of course. So what we've 0:15:16.640 --> 0:15:21.320 been hearing from our sources is that, for example, DTCPS 0:15:21.480 --> 0:15:26.040 funds now is being anchored by Porsche, the family holding, 0:15:26.040 --> 0:15:29.920 and by Deutsche Telecom and this marks a massive shift 0:15:30.080 --> 0:15:33.360 in mentality that these companies are now investing in a 0:15:33.440 --> 0:15:38.520 fund that potentially will back weapons systems. So private companies 0:15:38.600 --> 0:15:41.040 put a lot of restrictions on themselves when it came 0:15:41.080 --> 0:15:45.600 to weapon investments, and also public funds were very slow 0:15:45.680 --> 0:15:50.040 to adjust. And at the same time, you know, it's 0:15:50.080 --> 0:15:55.160 not just the companies, it's also a market question in 0:15:55.320 --> 0:15:59.800 terms of if you're looking to develop weapons and military 0:16:00.560 --> 0:16:04.680 then you need a founder that has the respective experience, 0:16:04.880 --> 0:16:08.760 and you also probably only have one customer, and that 0:16:08.880 --> 0:16:12.240 is the government or a government for that matter, and 0:16:12.320 --> 0:16:15.600 that's a tough customer to have because the procurement processes 0:16:15.640 --> 0:16:20.280 are slow and the you know, it's very complex, so 0:16:20.560 --> 0:16:25.400 I think to come back to DTCP, they kind of 0:16:25.440 --> 0:16:28.040 like broadened the space that they want to invest in 0:16:28.040 --> 0:16:31.880 in a bit to look at products that are being 0:16:31.920 --> 0:16:36.440 made for civilian and military technologies and for weapons, to 0:16:36.520 --> 0:16:38.720 kind of like circle around that question. 0:16:39.880 --> 0:16:42.960 Bloomberg Tech reporter Christina curious Oglu, Great to have you 0:16:43.000 --> 0:16:45.440 on the show, Thank you so much. Coming up the 0:16:45.560 --> 0:16:49.200 rise of humanoid robots and why Barclays says it could 0:16:49.200 --> 0:16:51.520 be the next two hundred billion dollar market. 0:16:52.080 --> 0:16:53.000 This is Blouemberg Tech. 0:17:06.760 --> 0:17:09.040 Just a few years ago. The robots here were kind 0:17:09.040 --> 0:17:09.480 of yankee. 0:17:09.640 --> 0:17:12.159 They were wobbly, and they had wires coming out of 0:17:12.160 --> 0:17:16.200 their head, and they looked like science experiments. And it's 0:17:16.520 --> 0:17:20.080 really fascinating to see just how quickly humanoids are turning 0:17:20.080 --> 0:17:21.720 it into a real product. 0:17:23.359 --> 0:17:26.040 That was Open Mind CEO Jan Liphart speaking of US 0:17:26.080 --> 0:17:30.600 at CEES, where humanoid robots dominated the conversation. This week, 0:17:30.960 --> 0:17:35.359 Barkley's published its forecast for the humanoid market, with expectations 0:17:35.600 --> 0:17:38.160 that it could grow to two hundred billion dollars by 0:17:38.200 --> 0:17:40.960 twenty thirty five. Here to break down that research is 0:17:41.000 --> 0:17:46.280 zonitza Tudor rover Barkley's Somatic FICC research team director. Really 0:17:46.320 --> 0:17:49.639 timely piece of research as well. And what is interesting 0:17:49.720 --> 0:17:52.919 is there's the two hundred billion dollar market forecast, but 0:17:53.040 --> 0:17:56.439 you identify two or three key factors about why you 0:17:56.520 --> 0:17:58.800 now see the commercial viability. 0:17:59.080 --> 0:18:00.560 What are those factors? 0:18:01.600 --> 0:18:04.000 Hello and Capornik, and thanks for having me on the show. 0:18:04.160 --> 0:18:07.480 AI is indeed getting very physical, and I think humanoid 0:18:07.640 --> 0:18:10.560 robots are the forefront of his strength. But I think 0:18:10.640 --> 0:18:14.400 the reason why the proposition is changing and why there's 0:18:14.440 --> 0:18:17.000 going to be demand for these robots is because the 0:18:17.080 --> 0:18:21.800 demographics of the war population is changing. The reality is 0:18:21.920 --> 0:18:26.359 that humans are getting older. I think by twenty fifty 0:18:27.119 --> 0:18:30.280 the share of people aged about sixty five is going 0:18:30.359 --> 0:18:33.560 to double. Also, humans don't really want to live in 0:18:33.640 --> 0:18:36.160 rural areas anymore. They want to be based in cities. 0:18:36.520 --> 0:18:39.880 But the factories and the manufacturing facilities are actually located 0:18:39.920 --> 0:18:42.280 outside of these cities, which creates a bit of a 0:18:42.359 --> 0:18:45.800 structural issue. And then let's not forget that. 0:18:48.280 --> 0:18:50.440 Workers' preferences are changing. 0:18:50.119 --> 0:18:53.680 So which means that they will be certain essential jobs 0:18:53.760 --> 0:18:57.760 yet undesirable which might not be sealed. And I think 0:18:57.800 --> 0:19:01.240 this is precisely where humanoid enter the picture. I think 0:19:01.280 --> 0:19:05.639 they could be taking on these repetitive jobs, this now dirty, 0:19:05.680 --> 0:19:11.359 potentially dangerous jobs of maintain productivity and producing efficiency gains 0:19:11.400 --> 0:19:12.920 for the industry as a whole. 0:19:14.480 --> 0:19:16.960 The main players in this field talk a lot about 0:19:17.000 --> 0:19:21.480 the labor shortage in particular manufacturing and heavy industry. You're 0:19:21.480 --> 0:19:23.120 saying it's going to be a two hundred billion dollar 0:19:23.720 --> 0:19:26.359 market by twenty thirty five. In your research, do you 0:19:26.440 --> 0:19:28.760 have any sense of how real the market is today, 0:19:29.119 --> 0:19:34.120 how many humanoids are actually out there in manufacturing contexts 0:19:34.160 --> 0:19:37.440 around the world in present day. 0:19:38.000 --> 0:19:40.760 The reality is that I think that humanoid robots are 0:19:41.400 --> 0:19:44.280 no longer confined to research labs. They're stepping out of 0:19:44.280 --> 0:19:46.840 the lab and they're getting into the real economy, into 0:19:46.840 --> 0:19:52.040 the real world. They are a couple of thousand humanoid 0:19:52.080 --> 0:19:57.639 robots already deployed on factory floors doing actual work in manufacturing, 0:19:57.720 --> 0:20:00.399 which I think is a natural starting place. Goes the 0:20:00.840 --> 0:20:04.960 jobs that are more structured, more easily defined, so it 0:20:05.000 --> 0:20:08.440 is the place to start for humanoids. But I think 0:20:08.440 --> 0:20:10.280 this is going to change in the next five to 0:20:10.320 --> 0:20:13.359 ten years as the technology matures and as this market 0:20:13.400 --> 0:20:15.600 picks up space. So I think I wouldn't be surprised 0:20:15.640 --> 0:20:19.080 that by twenty thirty five, when we actually project that 0:20:19.320 --> 0:20:21.720 the humanoids market is going to be as big as 0:20:21.720 --> 0:20:25.400 two hundred billion dollars, that we'll see tens of thousands, 0:20:25.520 --> 0:20:28.240 potentially even millions of humanoid robots. 0:20:29.560 --> 0:20:33.800 The supply chain for the humanoid robot is fascinating. I 0:20:33.800 --> 0:20:35.919 think a lot of emphasis at CES was on the 0:20:35.920 --> 0:20:39.639 breakthroughs in the models themselves, the ability to solve, in 0:20:39.720 --> 0:20:44.159 particular the real world data limitations. But you've got some 0:20:44.240 --> 0:20:47.000 interesting pricing you're talking about, like you know, per unit 0:20:47.080 --> 0:20:50.360 three million down to one hundred thousand, the economics changing 0:20:50.480 --> 0:20:54.280 fifteen thousand dollars for a humanoid robot. What's happening that 0:20:54.320 --> 0:20:57.119 gives you conviction that you'll see that change in the 0:20:57.160 --> 0:20:58.840 economics of a humanoid robot? 0:21:00.200 --> 0:21:00.360 Right? 0:21:00.400 --> 0:21:02.200 So, I think that there are a couple of things 0:21:02.200 --> 0:21:04.280 happening at the same time. First of all, over the 0:21:04.320 --> 0:21:08.280 past three to five years, we've seen really significant breakthroughs 0:21:08.280 --> 0:21:11.600 in cognitive AI models. But it's not only the AI 0:21:11.720 --> 0:21:14.760 that is getting better. The batteries are also getting better, 0:21:14.880 --> 0:21:18.600 more efficient, more powerful, and they also have been significant 0:21:18.640 --> 0:21:21.679 breakthroughs in high precision manufacturing and When I put all 0:21:21.720 --> 0:21:25.440 these three components together, the result is that the unit 0:21:25.560 --> 0:21:30.560 costs for producing humanoid robots is coming down. We estimate 0:21:30.600 --> 0:21:34.000 in our research that costs have been thirty have declined 0:21:34.000 --> 0:21:36.800 by thirty times over the past five years, from three 0:21:36.840 --> 0:21:40.200 million dollars per unit to one hundred thousand dollars per unit, 0:21:40.400 --> 0:21:43.480 and this means that the economics are getting optically more attractive. 0:21:43.640 --> 0:21:46.280 The use cases are expanding, and hence I think that 0:21:46.280 --> 0:21:50.200 the investment opportunity is also becoming more real and stronger. 0:21:50.480 --> 0:21:54.840 Right, very quickly, Zanita, who's ahead? We have fifteen seconds 0:21:54.960 --> 0:21:56.880 China or the United States here. 0:21:58.760 --> 0:22:00.920 I think it's going to be a tough competition. I think, 0:22:00.920 --> 0:22:03.000 for the moment if China, this is where we've seen 0:22:03.040 --> 0:22:04.879 most of the Roberts being deployed. But I think the 0:22:05.000 --> 0:22:06.119 US is quickly catching up. 0:22:07.960 --> 0:22:10.639 So I need to throw over Berkley to again out 0:22:10.680 --> 0:22:12.919 this week with the forecast that humanoid will be a 0:22:12.920 --> 0:22:15.920 two hundred billion dollar market by twenty thirty five. Thank 0:22:15.960 --> 0:22:18.240 you so much for joining us on the program. Now 0:22:18.240 --> 0:22:21.280 coming up an important conversation, We're going to be joined 0:22:21.280 --> 0:22:25.440 by the Chairman of the Federal Trade Commission to discuss antitrust, 0:22:25.760 --> 0:22:29.600 consumer protection, and AI regulation. A lot has happened this 0:22:29.720 --> 0:22:33.639 week with regards to the FTC. A conversation also about 0:22:34.320 --> 0:22:38.000