WEBVTT - Anthropic’s AI Spending Surges as Microsoft Rethinks Copilot 0:00:02.520 --> 0:00:04.210 Bloomberg Audio Studios. 0:00:04.690 --> 0:00:05.370 Podcasts. 0:00:05.690 --> 0:00:06.210 Radio. 0:00:06.650 --> 0:00:07.050 News. 0:00:14.040 --> 0:00:17.580 Bloomberg Tech is live from the heart of Silicon Valley 0:00:18.100 --> 0:00:20.380 with Ed Ludlow in San Francisco. 0:00:23.430 --> 0:00:27.950 Coming up, Anthropic signs an $ 11. 6 billion seven-year contract with 0:00:28.090 --> 0:00:32.720 Akamai for computing power. It's the largest deal in Akamai's history. Plus, 0:00:32.740 --> 0:00:35.440 President Trump calls on his meeting with Xi Jinping, quote, 0:00:35.580 --> 0:00:38.550 very productive. We'll discuss if there's anything to show for it. 0:00:39.200 --> 0:00:41.790 And Microsoft steps back from the race to build a 0:00:41.950 --> 0:00:45.810 personal chatbot and merging the consumer and workplace versions of 0:00:45.830 --> 0:00:50.379 its co-pilot AI assistant into one. I'm Tim Stenevek. In 0:00:50.500 --> 0:00:54.330 for Ed Ludlow, it is Bloomberg Technology. First, though, a 0:00:54.390 --> 0:00:57.890 check on markets. Bloomberg's Nora Melinda joins us now for more. 0:00:57.910 --> 0:00:58.190 Hey, Nora. 0:00:59.220 --> 0:01:01.500 Hey, Tim, we are seeing a bounce back of major 0:01:01.580 --> 0:01:04.320 indexes from early session lows. Those session lows, of course, 0:01:04.360 --> 0:01:06.400 coming off of that consumer sentiment data that we got 0:01:06.440 --> 0:01:09.080 earlier this morning, showing that consumer sentiment fell to a 0:01:09.160 --> 0:01:12.470 four month low. We've also got cooling oil really aiding 0:01:12.530 --> 0:01:16.070 sentiment here and a Nasdaq 100 that's actually heading for 0:01:16.110 --> 0:01:18.870 its biggest weekly gain in seven weeks. But one of 0:01:18.890 --> 0:01:20.530 the laggards that I want to look at right now, 0:01:20.590 --> 0:01:24.390 let's look at shares of Tesla. Currently pointing lower by 0:01:24.450 --> 0:01:27.820 about 1.7%. Now this coming after the fact that the 0:01:27.880 --> 0:01:31.959 company's ramping up production of its Optimus humanoid robot. But 0:01:31.980 --> 0:01:34.730 the concern here is about the production specifically as it 0:01:34.770 --> 0:01:35.730 pertains to. 0:01:35.810 --> 0:01:36.850 The robot's arms. 0:01:36.890 --> 0:01:39.369 Now we know that this robot is something specifically that 0:01:39.390 --> 0:01:43.060 the company plans to lease to other commercial companies. clients, 0:01:43.120 --> 0:01:45.080 but not necessarily sell because they would like to have 0:01:45.100 --> 0:01:47.700 the opportunity to be able to actually revamp and refurbish 0:01:47.720 --> 0:01:50.560 some of these robots. And lastly, looking at shares of Meta, 0:01:50.860 --> 0:01:53.890 they are softer today, but the stock actually heading for 0:01:53.990 --> 0:01:58.770 its biggest monthly gain since 2013 here. And this all really, 0:01:58.810 --> 0:02:00.850 of course, coming from the optimism that we've been seeing 0:02:00.990 --> 0:02:03.820 on Wall Street about that Muse AI Assistant. 0:02:04.300 --> 0:02:06.260 All right, Norm Malinda, thank you. We'll be checking in 0:02:06.280 --> 0:02:09.359 with Nora throughout the hour here on Bloomberg Tech. Meantime, 0:02:09.380 --> 0:02:12.459 Anthropic locking up even more computing power to keep pace 0:02:12.500 --> 0:02:16.120 with demand for Claude. The AI startup has struck an 0:02:16.120 --> 0:02:19.510 $ 11. 6 billion seven-year deal with Akamai. It comes, though, with 0:02:19.550 --> 0:02:22.730 an unusual twist, a warrant that could give Anthropic a 0:02:22.810 --> 0:02:25.790 stake in the company Bloomberg set up. joins us now 0:02:25.810 --> 0:02:28.410 with more. Hey, Seth, I was saying, I mean, when 0:02:28.450 --> 0:02:32.170 this crossed yesterday, I'm thinking, and Akamai, like, you know, CDN, 0:02:32.190 --> 0:02:35.090 like Content Delivery Network, Netflix, like a dozen years ago, 0:02:35.630 --> 0:02:36.829 are they going all in on AI? 0:02:37.090 --> 0:02:39.580 Yeah, well, it feels like at the moment, Anthropic and 0:02:39.660 --> 0:02:42.019 OpenAI and others are turning over every stone they can 0:02:42.080 --> 0:02:43.920 right now to get the chips and data centers and 0:02:43.940 --> 0:02:47.060 computer power they need to meet demands. So certainly for Akamai, 0:02:47.080 --> 0:02:48.899 you're right. I think they're taking a firm that focused 0:02:48.919 --> 0:02:51.760 on cybersecurity and making their biggest bet yet to really 0:02:51.800 --> 0:02:54.500 claim a piece of the AI pie. But for Anthropic, 0:02:54.560 --> 0:02:57.920 these deals, which can be sort of unprecedented for Akamai, 0:02:58.160 --> 0:03:02.840 are becoming routine. They're tapping Bitcoin miners and neo-clouds as 0:03:02.900 --> 0:03:05.780 well as rivals like Google and SpaceX to meet their demands. 0:03:05.840 --> 0:03:09.060 So it's part of the demand story here, like going 0:03:09.120 --> 0:03:11.280 anywhere to look for this compute at this point. 0:03:11.310 --> 0:03:13.089 Yeah, I mean, at one point, Anthropic said that their 0:03:13.590 --> 0:03:17.750 demand was growing 80x from the prior year. All of 0:03:17.790 --> 0:03:19.570 a sudden, they had to kind of plan for a 0:03:19.750 --> 0:03:22.870 rapid acceleration in computing capacity far beyond what I think 0:03:22.889 --> 0:03:25.110 they had anticipated earlier this year. And while you can 0:03:25.169 --> 0:03:27.210 do that by trying to launch these kind of Stargate 0:03:27.230 --> 0:03:29.660 initiatives like OpenAI has done, I think you're seeing Anthropic 0:03:29.680 --> 0:03:30.840 move more in the direction of trying to get a 0:03:30.900 --> 0:03:33.800 lot of these kind of bespoke partnerships off the ground 0:03:33.900 --> 0:03:34.640 and as quickly. 0:03:34.419 --> 0:03:36.360 As they can. On this idea of warrants, I mean, 0:03:36.380 --> 0:03:38.690 I think increasingly we're seeing this, not just from companies 0:03:38.710 --> 0:03:41.890 like Anthropic, but also companies like NVIDIA that are in 0:03:41.930 --> 0:03:44.200 this space, this idea that They can become owners or 0:03:44.240 --> 0:03:46.860 investors in other publicly traded companies. How do you look 0:03:46.880 --> 0:03:48.320 at this in terms of the cycle? 0:03:48.540 --> 0:03:50.820 Yeah, it certainly feels very much a piece of the 0:03:50.880 --> 0:03:54.030 circular financing criticism that we've all heard. AMD did stock 0:03:54.050 --> 0:03:56.970 warrants as part of a larger computing deal with OpenAI. 0:03:57.130 --> 0:03:59.390 So it's certainly not unprecedented here. But I think it 0:03:59.570 --> 0:04:02.900 adds to that concern or anxiety that these suppliers and 0:04:02.920 --> 0:04:06.650 developers are taking stakes in one another. potentially inflating the 0:04:06.690 --> 0:04:10.110 demand and also raising questions about the financial viability of 0:04:10.150 --> 0:04:11.850 these deals. 0:04:11.950 --> 0:04:14.070 This demand isn't coming from everywhere. Let's take the news 0:04:14.090 --> 0:04:16.400 from Oracle yesterday, for example, and this idea of a 0:04:16.420 --> 0:04:19.140 force majeure on a data center in the Southwest. How 0:04:19.180 --> 0:04:21.040 can we look at what we heard from Anthropic and 0:04:21.080 --> 0:04:23.599 Akamai yesterday and also what we heard from Oracle? 0:04:23.620 --> 0:04:25.260 I think two things are happening at the same time, 0:04:25.279 --> 0:04:27.700 that we are seeing a surge in business and consumer 0:04:27.720 --> 0:04:30.000 adoption of these products as they get better. And we're 0:04:30.020 --> 0:04:33.630 also seeing a lot of groundswell of criticism and pushback 0:04:33.710 --> 0:04:37.350 on the ground from lawmakers, public officials, and communities. And 0:04:37.390 --> 0:04:38.670 so I think these firms have to figure out how 0:04:38.690 --> 0:04:42.100 to navigate doing these massive data center build-outs that raise 0:04:42.120 --> 0:04:44.550 the concerns with other projects that might come to fruition 0:04:44.570 --> 0:04:45.089 more quickly. 0:04:45.150 --> 0:04:47.170 Well, Seth, you and your team help us navigate it all, 0:04:47.190 --> 0:04:49.450 so we appreciate you joining us on Bloomberg Tech. Bloomberg's 0:04:49.470 --> 0:04:52.390 Seth Figerman joining us here in New York. Shares of 0:04:52.410 --> 0:04:55.340 NVIDIA are higher today as Elon Musk races to scale 0:04:55.400 --> 0:04:59.260 up XAI's computing power. Musk says its Colossus 2 cluster 0:04:59.300 --> 0:05:05.430 in Memphis now has 550,000 NVIDIA GB200 and GB300 chips, 0:05:05.850 --> 0:05:09.229 with hundreds of thousands more expected online in the coming months. 0:05:09.650 --> 0:05:12.130 Now if all goes according to plan, the cluster could 0:05:12.170 --> 0:05:14.870 more than double its current chip count by the end 0:05:15.270 --> 0:05:18.609 of the year. The AI boom may be showing up 0:05:18.650 --> 0:05:20.990 in the bond market, too. The 10-year treasury yield jumped 0:05:21.029 --> 0:05:24.789 again yesterday. This is investors' way stronger growth, stubborn inflation, 0:05:24.810 --> 0:05:27.510 and a more hawkish Federal Reserve. At the same time, 0:05:27.550 --> 0:05:30.050 the massive AI build-out is flooding the market with new 0:05:30.110 --> 0:05:34.070 corporate debt. Ana Rathbun, Grenadilla Advisory Founder, joins us now 0:05:34.150 --> 0:05:36.410 for more. Ana, good to have you on the program. 0:05:36.510 --> 0:05:39.250 I want to talk about the relationship between tech stocks. 0:05:39.920 --> 0:05:43.200 and the 10-year because yesterday's increase in the 10-year I 0:05:43.220 --> 0:05:45.920 think was surprising to a lot of people with a 0:05:45.940 --> 0:05:50.110 yield higher than 5.2%. Yet here we are in a 0:05:50.150 --> 0:05:53.750 week where the NASDAQ 100 hit a new all-time high. Typically, 0:05:54.010 --> 0:05:55.789 tech stocks don't really like higher yields. 0:05:57.370 --> 0:05:57.590 Yes. 0:05:58.690 --> 0:06:01.150 No, they don't because they're long duration assets, right? So 0:06:01.170 --> 0:06:05.090 if we're talking bond speak, the duration here matters. But, 0:06:05.130 --> 0:06:08.250 you know, actually, it doesn't always move with a 10 year. 0:06:08.310 --> 0:06:10.920 And I think that's what we're seeing today. And if 0:06:10.960 --> 0:06:13.240 we go back a little bit, you know, tech stocks 0:06:13.260 --> 0:06:15.900 have been pressured for many, many weeks. And so there's 0:06:15.920 --> 0:06:18.460 some a little bit of a selling fatigue that I 0:06:18.500 --> 0:06:19.160 think we've seen. 0:06:19.220 --> 0:06:22.240 So we see that bouncing. But over the long term. 0:06:22.700 --> 0:06:26.239 There has been repricing of AI stocks and tech stocks 0:06:26.520 --> 0:06:29.409 that will match what we're expecting to see in the 0:06:29.450 --> 0:06:33.549 longer term, especially in the 10-year yield for the foreseeable future. 0:06:33.570 --> 0:06:35.560 And I do think that we are in a higher 0:06:35.600 --> 0:06:37.160 for longer environment. 0:06:37.940 --> 0:06:39.800 Ana, I'm so sorry to do this. I want to 0:06:39.839 --> 0:06:42.339 get to President Trump now. He's with the First Lady. 0:06:42.380 --> 0:06:45.120 They're now doing a site visit. This is the National 0:06:45.180 --> 0:06:50.719 Archives with Chinese President Xi Jinping and Madam Pang. You 0:06:50.760 --> 0:06:54.990 can see the live pictures coming from Washington, D.C. 0:06:55.050 --> 0:06:55.430 Right now. 0:06:55.890 --> 0:06:58.130 This is after the U.S. and China extended a trade 0:06:58.170 --> 0:07:01.490 truce for another two months following President Trump and Xi 0:07:01.510 --> 0:07:04.849 Jinping's meeting yesterday. We're going to continue to monitor this. 0:07:05.270 --> 0:07:09.150 the situation. You can also see it there at LiveGo 0:07:09.440 --> 0:07:11.760 on the Bloomberg Terminal. We'll bring you any headlines as 0:07:11.800 --> 0:07:14.620 we continue to get them. Ana Rathbun joins us from 0:07:14.660 --> 0:07:17.680 Grenadilla Advisory. I want to get right back to Ana. Ana, 0:07:18.340 --> 0:07:20.200 I want to bring you in here. Thanks for your 0:07:20.340 --> 0:07:21.910 patience with that, by the way. I appreciate it. We're 0:07:21.930 --> 0:07:24.450 watching everything happening in Washington at this, I think it's 0:07:24.470 --> 0:07:26.830 fair to say, historic visit. I do want to bring 0:07:26.870 --> 0:07:30.870 in geopolitics, though, to this conversation because this is really important, Ana, 0:07:31.090 --> 0:07:34.240 and it's the question of You know, the relationship between 0:07:34.250 --> 0:07:38.020 the U.S. and China and really what it means for 0:07:38.160 --> 0:07:39.880 not just tech companies. I mean, look at who you 0:07:39.920 --> 0:07:44.110 had there last night. It was absolutely a who's who of, 0:07:44.500 --> 0:07:46.580 I think it's fair to say, the most important people 0:07:47.300 --> 0:07:50.179 in Silicon Valley, like not just from the venture capital 0:07:50.220 --> 0:07:54.000 community and the technology community, but from business leaders all 0:07:54.060 --> 0:07:55.840 over the country. How did you look at that? 0:07:56.970 --> 0:07:59.250 Well, that tells you where the priorities are, right? 0:07:59.330 --> 0:08:03.580 Not only from the United States standpoint, but also in 0:08:03.720 --> 0:08:05.280 our relationship with China. 0:08:05.600 --> 0:08:08.740 And I think AI is at the center of a lot. 0:08:08.540 --> 0:08:10.840 Of these talks, even though we are talking about soybeans 0:08:10.900 --> 0:08:14.020 and we are talking about tariffs. AI is front and center, 0:08:14.040 --> 0:08:18.500 and everyone's been talking about this potential global governing body 0:08:18.760 --> 0:08:20.960 so that we can put some safety rules. 0:08:20.700 --> 0:08:24.090 Around AI development. But the truth is, no one's going 0:08:24.110 --> 0:08:24.830 to slow it down. 0:08:25.110 --> 0:08:28.190 Even between competitors in the United States, there is no 0:08:28.230 --> 0:08:28.870 slowing down. 0:08:28.910 --> 0:08:31.190 I mean, we just talked about Anthropy getting more compute. 0:08:32.070 --> 0:08:33.450 And definitely between U.S. 0:08:33.510 --> 0:08:36.630 And China, I mean, we can't expect China to slow down, 0:08:36.670 --> 0:08:38.309 and they're not expecting us to slow down. 0:08:38.350 --> 0:08:40.610 So all of these talks... We're saying that. 0:08:40.450 --> 0:08:43.860 These things are very important to the two parties. But 0:08:43.970 --> 0:08:46.030 at the end of the day, the reality is that 0:08:46.090 --> 0:08:47.270 it's going to be more of the same. 0:08:47.690 --> 0:08:52.130 You know, I'm wondering about the different messages that we're 0:08:52.170 --> 0:08:57.179 getting from the president and his view of a so-called 0:08:57.260 --> 0:09:01.400 slowdown in this technology versus, you know, Dario Amadei addressing 0:09:01.420 --> 0:09:05.209 the UN just a couple days ago. Sam Altman of 0:09:05.270 --> 0:09:08.829 OpenAI addressing the UN, talking about the dangers and the 0:09:08.870 --> 0:09:14.510 importance of sort of international cooperation with this technology. I'm 0:09:14.550 --> 0:09:17.720 curious about your view on the different messages that we're 0:09:17.740 --> 0:09:19.820 getting from the president versus the people who are actually 0:09:19.860 --> 0:09:21.059 developing the frontier models. 0:09:22.020 --> 0:09:25.980 Well, I think that there are reasons to be concerned. 0:09:26.040 --> 0:09:29.270 The reality is, I mean, if we have an international 0:09:29.309 --> 0:09:32.190 governing body, who's going to enforce it if somebody breaks the. 0:09:32.130 --> 0:09:32.650 Rules, right? 0:09:33.050 --> 0:09:35.390 And I think President Trump, what he is saying, it 0:09:35.450 --> 0:09:37.390 is different, but I think it's a little bit more 0:09:37.410 --> 0:09:39.030 realistic from my perspective. 0:09:39.100 --> 0:09:40.240 Nobody's slowing down. 0:09:40.280 --> 0:09:42.660 I think this is just talk at this point in 0:09:42.700 --> 0:09:44.579 time because there's no way to enforce it. 0:09:45.300 --> 0:09:49.680 But do you think there needs to be some sort 0:09:49.700 --> 0:09:52.610 of international standard? I'm going to try not to talk 0:09:52.650 --> 0:09:55.470 to you about the AI apocalypse because we've been talking 0:09:55.510 --> 0:09:58.270 about that so much. But the fact of the matter 0:09:58.290 --> 0:10:00.840 is there's a growing chorus of people who do think 0:10:00.880 --> 0:10:04.420 that this technology could be dangerous to humankind. I mean, 0:10:04.720 --> 0:10:06.360 we're talking about something existential here. 0:10:07.360 --> 0:10:07.720 Yeah. 0:10:07.860 --> 0:10:12.840 And I think humans have the key to producing guardrails 0:10:12.920 --> 0:10:15.679 as well as putting safety rules around it. I mean, 0:10:15.720 --> 0:10:19.000 for example, if we're worried about bad actors getting access 0:10:19.040 --> 0:10:22.280 to powerful AI tools, that's not an AI problem. That 0:10:22.340 --> 0:10:25.780 is a bad actor problem. So I think we do 0:10:25.840 --> 0:10:29.160 need guardrails for sure, because this is new technology and 0:10:29.210 --> 0:10:31.910 these AI agents are going to places we don't intend. 0:10:32.010 --> 0:10:32.210 Right. 0:10:32.270 --> 0:10:34.810 So we need to learn from it and put guardrails 0:10:34.870 --> 0:10:38.869 around it. But I am skeptical about the international nature of. 0:10:39.410 --> 0:10:44.230 potential international nature of anything that will actually govern AI development. 0:10:44.270 --> 0:10:46.180 We only have about a minute left with you, and 0:10:46.200 --> 0:10:48.120 I do want to continue to go around the world 0:10:48.179 --> 0:10:51.860 and kind of go back to geopolitics and the relationship 0:10:51.980 --> 0:10:54.700 between bond yields and rising oil prices, because that was 0:10:54.760 --> 0:10:57.110 a major story this week, just in the context of 0:10:57.450 --> 0:11:00.429 these technology companies and what's happening in Washington and here 0:11:00.450 --> 0:11:02.670 in New York at the UN. How are you looking 0:11:02.750 --> 0:11:05.770 at what's happening in the Strait of Hormuz affecting the 0:11:05.809 --> 0:11:07.050 way you look at the markets? 0:11:08.170 --> 0:11:10.470 Yeah, I mean, straight of the straight of Hormuz is 0:11:10.510 --> 0:11:14.790 what holds the key to oil prices, diesel prices, bond 0:11:14.830 --> 0:11:18.350 deals at the end for the short term. So I'm 0:11:18.470 --> 0:11:21.360 looking I'm hoping for it, actually. But we've been hoping 0:11:21.420 --> 0:11:24.160 for some kind of an end to this crisis since. 0:11:24.860 --> 0:11:25.280 March. 0:11:25.940 --> 0:11:29.400 So I'm not holding my breath, but it is good 0:11:29.460 --> 0:11:31.640 to hear that there is work sort of going back 0:11:31.700 --> 0:11:34.800 to the MOU from June and that we're actually talking 0:11:34.860 --> 0:11:37.059 again instead of like, you know, bombing each other or 0:11:37.100 --> 0:11:40.969 bombing the pipelines. So, you know, it's a little bit 0:11:41.010 --> 0:11:43.410 of a roller coaster, but I am hoping that this 0:11:43.490 --> 0:11:46.950 is productive because, as I said before, the Strait of 0:11:46.990 --> 0:11:49.449 Hormuz holds the key to a lot of things that 0:11:49.530 --> 0:11:51.430 have been a domino effect in the markets. 0:11:51.890 --> 0:11:54.450 Ana Rathbun, the perfect guest for the moment. Appreciate your 0:11:54.470 --> 0:11:59.780 time this morning of Grenadilla Advisory. Joining us from New York. Thanks, Ana. Well, 0:11:59.840 --> 0:12:04.100 AI data center startup Nscale has raised $ 3. 36 billion. This 0:12:04.140 --> 0:12:07.560 as it prepares to go public. The pre-IPO financing was 0:12:07.600 --> 0:12:10.760 led by Third Point and includes a $ 1 billion commitment 0:12:11.059 --> 0:12:14.540 from NVIDIA. Nscale filed for an IPO last week that 0:12:14.559 --> 0:12:19.140 could raise as much as $ 3 billion. Coming up, President 0:12:19.160 --> 0:12:22.120 Trump calls his meeting with Xi Jinping, quote, very productive, 0:12:22.500 --> 0:12:26.020 but offers few details on the future for trade, for tech, 0:12:26.380 --> 0:12:29.340 and the AI race. More on that next. This is Bloomberg. 0:12:42.050 --> 0:12:45.510 President Trump calling his meeting with Xi Jinping very productive, 0:12:45.570 --> 0:12:48.150 but so far there's little to show for it. After 0:12:48.190 --> 0:12:50.830 a lavish state dinner packed with some of America's biggest 0:12:50.870 --> 0:12:55.290 tech CEOs, the president offered no details on breakthroughs over trade, technology, 0:12:55.570 --> 0:12:59.650 or AI. Bloomberg's Mike Shepard joins us now. Shep, I 0:12:59.700 --> 0:13:02.179 want to talk to you about what we learned last night, 0:13:02.200 --> 0:13:04.260 but I do want to note that as we just 0:13:04.300 --> 0:13:07.780 saw a few minutes ago, the president and Xi Jinping 0:13:08.240 --> 0:13:12.920 wrapping up at least comments at a tee in Washington, D.C. 0:13:12.940 --> 0:13:16.620 The president said we've made great strides, positive for both countries, 0:13:17.100 --> 0:13:19.820 made tremendous and great strides for China and the U.S., 0:13:19.860 --> 0:13:23.400 and we've had a tremendous visit from Xi. Apart from 0:13:23.440 --> 0:13:26.070 the superlatives, do we have any details? 0:13:27.350 --> 0:13:30.330 We really don't have very much crunch behind that or 0:13:30.410 --> 0:13:33.810 substance of what that all means. The president went on 0:13:33.870 --> 0:13:36.469 to say that American farmers would be very happy to 0:13:36.790 --> 0:13:39.440 and that the American people were also very. 0:13:39.250 --> 0:13:40.300 Happy with the visit. 0:13:40.679 --> 0:13:44.200 Unclear what exactly he's basing that on, but certainly the 0:13:44.240 --> 0:13:48.020 pageantry last night was the kind of thing that he 0:13:48.059 --> 0:13:51.100 has really come to embrace, not only in his first 0:13:51.140 --> 0:13:54.349 term in office, but really very much so as a 0:13:54.390 --> 0:13:58.140 defining feature of his second term as president. We heard 0:13:58.179 --> 0:14:01.040 the Chinese leader also speak during that same appearance in 0:14:01.059 --> 0:14:03.440 the Red Room at the White House for tea, where 0:14:03.820 --> 0:14:07.160 he talked a little bit more about strategic stability between 0:14:07.179 --> 0:14:09.280 the two sides. And this is a key phrase, and 0:14:09.320 --> 0:14:11.700 it's one that he brought up in May when the 0:14:11.760 --> 0:14:14.199 two leaders met in Beijing. And it really does seem 0:14:14.240 --> 0:14:18.109 that China is looking to ensure that relations between the 0:14:18.150 --> 0:14:22.530 world's two largest economies remain on track, and unbuffeted by 0:14:22.610 --> 0:14:23.970 turbulence or tension. 0:14:24.290 --> 0:14:26.150 On the tech side of this, I mean, this was 0:14:26.190 --> 0:14:29.990 a who's who of tech and finance heavyweights. I mean, 0:14:30.330 --> 0:14:35.210 we're talking Apple executive chair, Tim Cook, Jeff Bezos, David Solomon, 0:14:35.250 --> 0:14:39.270 Jamie Dimon, Jane Fraser, Satya Nadella, Sunder Pichai, Jensen Wong, 0:14:39.310 --> 0:14:44.430 Lisa Su, Sam Altman, Mark Zuckerberg. We're seeing Sam Altman 0:14:44.450 --> 0:14:47.510 and Mark Zuckerberg right there speaking to one another. What 0:14:47.530 --> 0:14:50.740 does it tell you? about the state dinner, about the 0:14:50.780 --> 0:14:53.360 priorities of the president, about the priorities of this administration, 0:14:53.380 --> 0:14:54.440 that we saw all these people there. 0:14:55.780 --> 0:14:59.170 Well, Tim, I've been diving through the photos from last night, 0:14:59.250 --> 0:15:01.550 as you have been too, and as we're seeing now, 0:15:03.350 --> 0:15:07.410 showing now our audience from last night's festivities. And what 0:15:07.450 --> 0:15:09.620 it tells me is that the president really has thrown 0:15:09.670 --> 0:15:14.160 his arms around Silicon Valley as an engine of investment 0:15:14.340 --> 0:15:15.240 and job creation. 0:15:15.640 --> 0:15:18.130 Principally in this area of artificial intelligence. 0:15:18.170 --> 0:15:20.330 And that is something that he has no desire to 0:15:20.370 --> 0:15:23.430 slow down right now. And we have heard him articulate 0:15:23.470 --> 0:15:27.580 this repeatedly, but he has escalated his defense of continuing 0:15:27.620 --> 0:15:31.720 the AI build-out against all these growing safety concerns that 0:15:31.780 --> 0:15:35.030 we have seen expressed recently. by Silicon Valley leaders. As 0:15:35.050 --> 0:15:37.670 you noted earlier in the program, we heard from Dario 0:15:37.730 --> 0:15:40.010 Amadei and even Sam Altman, who was at the dinner 0:15:40.150 --> 0:15:44.290 last night, expressing their concerns about AI risks, that it 0:15:44.330 --> 0:15:47.680 could be used by malicious actors. It also carries a 0:15:47.760 --> 0:15:53.260 risk of perhaps even existential proportions if the technology self-improves 0:15:53.360 --> 0:15:57.440 to an extent that it slips beyond human control. And 0:15:57.620 --> 0:16:00.370 that is something that worries the industry, is fueling calls 0:16:00.410 --> 0:16:03.130 for some standards, but it's one that Trump has rejected. 0:16:03.520 --> 0:16:06.180 On the other hand, during remarks yesterday, we did hear 0:16:06.240 --> 0:16:09.910 Xi Jinping express his desire to see that the technology 0:16:09.970 --> 0:16:13.670 be managed safely. And he also expressed his desire that 0:16:13.770 --> 0:16:17.050 it remain under human control, signaling that there is perhaps 0:16:17.100 --> 0:16:19.400 a bit of difference between the two sides on this 0:16:19.460 --> 0:16:21.300 question of guardrails and standards. 0:16:21.760 --> 0:16:24.500 Mike Shepard down there in Washington, D.C. Shep, it's always 0:16:24.520 --> 0:16:27.300 great to check in with you. Do appreciate your time. 0:16:27.780 --> 0:16:31.020 I want to continue the discussion now with Julian Goetz, 0:16:31.280 --> 0:16:33.870 fellow at the Harvard Belfort Center, who served as senior 0:16:33.890 --> 0:16:36.609 director for China and Taiwan affairs at the White House 0:16:36.690 --> 0:16:39.990 National Security Council during the Biden administration. Julian, good to 0:16:40.050 --> 0:16:43.690 have you on Bloomberg Tech this morning. Just curious about 0:16:44.990 --> 0:16:49.740 what you make of any sort of actual progress between 0:16:49.760 --> 0:16:53.130 the relationship between the US and China. Is there any 0:16:53.150 --> 0:16:57.280 indication that it's gotten better after yesterday's dinner and today's festivities? 0:16:58.560 --> 0:17:01.040 Well, I think it's important to remember that for President Trump, 0:17:01.100 --> 0:17:04.560 the point of this visit is the pageantry. It is 0:17:04.619 --> 0:17:08.160 the optics. He has not entered this time that he's 0:17:08.180 --> 0:17:11.379 spending with Xi Jinping with a big, long list of 0:17:11.480 --> 0:17:14.060 substantive deliverables that he's hoping to secure. 0:17:14.460 --> 0:17:16.330 He is really hoping to project. 0:17:16.010 --> 0:17:20.270 This image of his relationship with Xi Jinping being positive. 0:17:20.609 --> 0:17:23.930 And I think He has succeeded by that metric. But 0:17:24.730 --> 0:17:26.970 the reality is that for the United States and China, 0:17:27.010 --> 0:17:30.540 there is a long list of substantive challenges to work 0:17:30.580 --> 0:17:34.980 through from trade to technology to Taiwan, and that there 0:17:35.000 --> 0:17:38.379 has been very little time even in their schedule for 0:17:38.420 --> 0:17:42.490 them to sit down and do substantive work on those topics. 0:17:42.570 --> 0:17:46.600 And instead, they've been showing off the helipad, touring the 0:17:46.640 --> 0:17:50.160 White House, touring the National Archives. And so the net 0:17:50.200 --> 0:17:52.310 effect of all of this is that the United States 0:17:52.440 --> 0:17:57.810 is giving Beijing these tremendous optics. But in exchange, it's 0:17:57.850 --> 0:18:00.510 not clear to me that we're getting much of substance that, 0:18:00.570 --> 0:18:02.050 you know, sort of for core U.S. 0:18:02.109 --> 0:18:02.510 Interests. 0:18:02.630 --> 0:18:04.570 Julian, I'm a little surprised you didn't mention Iran there 0:18:04.850 --> 0:18:09.710 in that long list, because China's role in aiding Iran 0:18:09.850 --> 0:18:13.290 during this conflict is seems to be, I don't know, 0:18:13.310 --> 0:18:18.030 a very big part of this relationship and an idea 0:18:18.090 --> 0:18:21.600 that maybe something the US and the president need to 0:18:21.630 --> 0:18:23.360 address with Xi Jinping. 0:18:24.619 --> 0:18:26.980 Absolutely, and it should be on the list. And, you know, 0:18:27.020 --> 0:18:29.840 it's quite surprising because in the lead up to this summit, 0:18:30.560 --> 0:18:34.420 as you know, the focus from the Treasury Department, other 0:18:34.460 --> 0:18:37.240 parts of the US government was on trying to see 0:18:37.260 --> 0:18:38.919 what they could do to turn up the pressure on 0:18:38.960 --> 0:18:42.500 Beijing a bit. without going for sanctions or some of 0:18:42.520 --> 0:18:45.460 the steps that might disrupt the trade truce to try 0:18:45.480 --> 0:18:49.930 to put pressure on Beijing's close relationship with Iran. But 0:18:49.950 --> 0:18:53.389 it does not seem, at least from we can tell publicly, 0:18:53.770 --> 0:18:57.359 like President Trump has made that a priority during this visit. 0:18:57.480 --> 0:19:00.780 So it's surprising that it hasn't been more salient throughout Beijing. 0:19:01.000 --> 0:19:03.500 And candidly, you know, with a different leader, you might 0:19:03.600 --> 0:19:06.600 see more pressure. But I get the sense that President 0:19:06.640 --> 0:19:09.140 Trump does not want to do anything that might, you know, 0:19:09.160 --> 0:19:12.010 kill the vibe, so to speak, between him and Xi Jinping. 0:19:12.050 --> 0:19:15.330 Could you say apply that same logic to Russia and 0:19:15.830 --> 0:19:17.290 Russia's invasion of Ukraine? 0:19:18.410 --> 0:19:18.870 Absolutely. 0:19:18.890 --> 0:19:20.990 I mean, I think we've seen very little interest from 0:19:21.060 --> 0:19:24.219 Beijing in doing more to help bring that conflict to 0:19:24.260 --> 0:19:27.740 an end over many years now. And President Trump certainly 0:19:27.800 --> 0:19:30.119 does not seem to want to put the pressure on 0:19:30.140 --> 0:19:33.150 Xi Jinping. There have even been reports that if President 0:19:33.190 --> 0:19:35.869 Trump travels to China a second time this year to 0:19:35.890 --> 0:19:38.810 go to Shenzhen for a multilateral meeting later that Xi 0:19:38.830 --> 0:19:41.110 Jinping will host, which they very much hope he will, 0:19:41.410 --> 0:19:43.800 that one of the things Xi Jinping will dangle is 0:19:43.880 --> 0:19:48.850 a trilateral meeting between him, Trump, and President Putin of Russia, 0:19:49.190 --> 0:19:52.080 which I think would be an extraordinary misstep, though it's 0:19:52.119 --> 0:19:54.919 exactly the kind of thing that President Trump seems inclined 0:19:54.960 --> 0:19:58.000 to do with these two leaders who he regards as 0:19:58.140 --> 0:20:00.260 peers of his on the world stage. 0:20:00.420 --> 0:20:03.379 Julian, digging into tech and AI a little bit, the 0:20:03.420 --> 0:20:10.109 Taiwan crisis, challenge for this administration. Given TSMC's role in 0:20:10.170 --> 0:20:13.300 the build out of AI infrastructure, its relationship with Nvidia 0:20:13.760 --> 0:20:17.859 and many other companies, what does the president need to 0:20:17.920 --> 0:20:20.260 do or I guess the better question is what leverage 0:20:20.280 --> 0:20:24.379 does China have with regard to the AI ecosystem? 0:20:25.970 --> 0:20:28.810 Well, it's really important to remember that when it comes 0:20:28.869 --> 0:20:33.970 to broader U.S.-China AI competitions, this is an intense competitive dynamic. 0:20:34.230 --> 0:20:37.380 The United States has the world's leading frontier labs, but 0:20:37.440 --> 0:20:41.020 China is working intently to catch up, both in diffusion 0:20:41.060 --> 0:20:44.899