WEBVTT - Apple Picks Gemini to Run AI-Powered Siri 0:00:02.520 --> 0:00:13.400 Bloomberg Audio Studios, podcasts, radio news. Bloomberg Tech is alive 0:00:13.480 --> 0:00:17.279 from coast to coast with Caroline Hide in New York 0:00:17.560 --> 0:00:19.560 and Eva Low in San Francisco. 0:00:22.400 --> 0:00:24.880 This is Bloomberg Tech coming up in video. Plans to 0:00:24.920 --> 0:00:27.920 invest one billion dollars over five years in a new 0:00:28.000 --> 0:00:30.280 lab with Eli Lilly to speed up the use of 0:00:30.320 --> 0:00:32.680 AI in the pharmaceutical industry. 0:00:32.360 --> 0:00:35.479 Plus paramount sues Warner Brothers Discovery, and plans to nominate 0:00:35.520 --> 0:00:39.400 directors to the board as it's increasingly hostile. Takeover bid 0:00:39.640 --> 0:00:41.720 takes aim to upset Netflix. 0:00:42.040 --> 0:00:45.960 An alphabet breaches four trillion dollars in market value. This 0:00:46.040 --> 0:00:50.400 is CNBC reports Apple has picked Gemini to power Siri 0:00:50.760 --> 0:00:52.160 this year, and. 0:00:52.040 --> 0:00:56.200 Then those moves fall again, and really we've got anxiety 0:00:56.240 --> 0:00:58.640 across the market's more broadly aired digging into the market 0:00:58.680 --> 0:01:01.480 implications of what seems to be a dialing up. 0:01:01.600 --> 0:01:03.360 In the administration's fight with a FED. 0:01:03.440 --> 0:01:05.280 That is what the market tries to digest and why 0:01:05.319 --> 0:01:07.040 we see a bit of a sell America theme at 0:01:07.040 --> 0:01:09.640 the moment. Yes, they're small moves in terms of benchmarks 0:01:09.640 --> 0:01:11.080 for stocks were down by a tenth of a percent 0:01:11.120 --> 0:01:13.720 on anastat one hundred, but we're seeing bigger moves when 0:01:13.760 --> 0:01:16.000 you're looking what's happening in the bomb market. In the dollar, 0:01:16.000 --> 0:01:18.640 for example, we're seeing thirty year years once again, pushing 0:01:18.720 --> 0:01:21.480 up two basis points, seeing the dollar on the downside. 0:01:21.640 --> 0:01:23.120 All of this as we try and bake in the 0:01:23.160 --> 0:01:26.200 macro implications. We look to CPI tomorrow, but you're looking 0:01:26.200 --> 0:01:27.600 at what's underneath the hood right now. 0:01:28.160 --> 0:01:30.480 Yeah, the big breaking news of the last hour is 0:01:30.520 --> 0:01:34.720 that Apple has selected Google's Gemini model to power serri 0:01:34.840 --> 0:01:38.720 this year. Reporting from CNBC citing an Apple's statement, and 0:01:38.760 --> 0:01:41.200 you can see a big reaction in the stock the 0:01:41.240 --> 0:01:43.080 moment the headlines hit, it has faded. 0:01:43.480 --> 0:01:44.240 I'll take you back to. 0:01:44.280 --> 0:01:47.720 November when Bloomberg's Mark German reported that a deal was 0:01:47.800 --> 0:01:50.960 very close, that Apple would pay Google one billion dollars 0:01:51.000 --> 0:01:54.080 per year and that this was just an interim until 0:01:54.120 --> 0:01:56.880 Apple could get its own foundation models together, but or 0:01:56.920 --> 0:02:00.400 continue to track that story. Another interesting trends action in 0:02:00.400 --> 0:02:04.520 the market this morning Nvidia co investing one billion dollars 0:02:04.520 --> 0:02:08.560 with Eli Lilly into a Silicon Valley based lab split 0:02:08.639 --> 0:02:13.360 over five years. AI Pharmaceuticals for the details are interesting here, 0:02:13.440 --> 0:02:14.160 Cara and we. 0:02:14.120 --> 0:02:15.920 Want to get those details, and we can do exactly 0:02:15.919 --> 0:02:17.639 that with Katie Greifeld, which is over at the JP 0:02:17.720 --> 0:02:22.040 Morgan Healthcare Conference, indeed looking at Silicon Valley, getting ever 0:02:22.160 --> 0:02:24.000 closer to the world of health. 0:02:24.080 --> 0:02:25.239 Katie, what'd you make of this deal? 0:02:28.080 --> 0:02:31.160 Yeah, this was a really interesting headline to wake up to, 0:02:31.280 --> 0:02:34.840 especially when you remember that Eli Lilly partnered with Nvivia 0:02:34.880 --> 0:02:38.840 back in October to develop an AI supercomputer. So the 0:02:38.880 --> 0:02:41.839 details on this specific deals as you guys run through them. 0:02:42.160 --> 0:02:45.640 One billion dollars OVERFI fears to build a facility in 0:02:45.720 --> 0:02:49.200 Silicon Valley, basically a new lab. We don't have much 0:02:49.280 --> 0:02:52.160 details when it comes to the terms beyond that, but 0:02:52.240 --> 0:02:55.640 both companies have described this as a joint investment. And 0:02:55.680 --> 0:02:58.360 I just actually spoke to the CFO of Eli Lilly, 0:02:58.400 --> 0:03:01.959 Lucas Montarge, about what the ultimate vision is here when 0:03:01.960 --> 0:03:05.360 it comes to AI and healthcare and this specific partnership, 0:03:05.360 --> 0:03:08.480 and he said, this is really about drug discovery. It's 0:03:08.520 --> 0:03:11.120 still early days, of course, they haven't built the facility yet, 0:03:11.160 --> 0:03:14.320 but the hope is that this will help basically in 0:03:14.440 --> 0:03:18.640 discovering drugs to really target and treat those really hard 0:03:18.680 --> 0:03:22.680 to treat diseases. Think gene therapy for example. So again 0:03:22.840 --> 0:03:25.400 early days, but some big ambitions. 0:03:24.840 --> 0:03:27.840 Here Bloomber's Katie Greefelk just down the road at the 0:03:27.919 --> 0:03:30.880 JP Morgan Healthcare conference all day long here in San Francisco, 0:03:30.960 --> 0:03:33.040 less than two weeks into the new year, and in 0:03:33.160 --> 0:03:35.200 Video is already dominating the news flow. I want to 0:03:35.200 --> 0:03:38.320 get to Denny Fish, portfolio manager on the Global Technology 0:03:38.480 --> 0:03:41.960 an innovation team at Janis Henderson and the Frank Reality visit. 0:03:42.000 --> 0:03:44.240 In video is the top holding across a number of 0:03:44.280 --> 0:03:46.600 funds that you manage. What do you make of that 0:03:46.880 --> 0:03:48.480 relationship between in video and ELI. 0:03:48.840 --> 0:03:51.000 Yeah, well it actually it makes a ton of sense 0:03:51.040 --> 0:03:51.240 to me. 0:03:51.480 --> 0:03:53.600 I mean, if we think about what's grabbed all the 0:03:53.640 --> 0:03:56.960 attention over the last three years since the chat ChiPT moment, 0:03:57.320 --> 0:04:01.360 it's effectively been the idea of the digital manifestation of AI. 0:04:01.880 --> 0:04:04.119 But the reality is, if we think about the profoundness 0:04:04.120 --> 0:04:06.640 of how this will impact the economy over time and 0:04:06.760 --> 0:04:12.560 society more broadly, the physical manifestation of AI is as big, 0:04:12.600 --> 0:04:15.400 if not a bigger opportunity. And if you think about 0:04:16.200 --> 0:04:19.760 drug discovery and what that means for society, and then 0:04:19.880 --> 0:04:22.280 just think about how far we've come with autonomous driving, 0:04:22.279 --> 0:04:24.760 where we're going with robotics. They are all these different 0:04:24.839 --> 0:04:28.239 verticals where it makes a lot of sense for Nvidia 0:04:28.320 --> 0:04:30.960 to be leaning into. And frankly, in Vidia has got 0:04:31.120 --> 0:04:33.640 some of the most the best talent in the world 0:04:34.120 --> 0:04:36.800 and to see these partnerships with market leaders makes a 0:04:36.839 --> 0:04:37.440 ton of sense. 0:04:37.480 --> 0:04:38.720 And you know, I bring it back. 0:04:38.760 --> 0:04:42.920 Sometimes you get this argument of circularity, but the reality is, 0:04:42.960 --> 0:04:46.880 if you're Jensen and you have a firm belief at 0:04:46.920 --> 0:04:49.960 just how big these market opportunities are, you should be 0:04:50.080 --> 0:04:53.440 leaning and hard to this intment co investments into labs 0:04:53.640 --> 0:04:56.440 like they're doing with Lily, or direct investments into companies 0:04:56.800 --> 0:04:58.680 like they're potentially doing with Open AI. 0:04:58.800 --> 0:05:00.760 We a let's one of miss driving. A lot of 0:05:00.800 --> 0:05:03.360 people get it. I wrote about it extensively this weekend. 0:05:03.400 --> 0:05:05.320 We're going to talk about it later in the program. 0:05:05.360 --> 0:05:07.320 The go to market in the first instance is for 0:05:07.440 --> 0:05:11.800 Mercedes that we use hardware and software. But with pharmaceuticals 0:05:11.960 --> 0:05:15.280 drug discovery research, what is the go to market that 0:05:15.320 --> 0:05:17.240 then video is going to pull off in the future. 0:05:18.360 --> 0:05:20.440 Well to the so so far. 0:05:20.600 --> 0:05:24.080 If we think about pharmaceuticals, where we've seen the early 0:05:24.120 --> 0:05:28.400 benefits of AI have been in things like the clinical process, 0:05:29.200 --> 0:05:32.280 you know, getting drugs that have already been discovered through 0:05:32.360 --> 0:05:36.679 that process, internal efficiencies, marketing, things like that. The real, 0:05:37.080 --> 0:05:41.320 you know, the big prize out there is drug discovery. 0:05:41.560 --> 0:05:45.160 It's finding solutions to problems that we just couldn't get 0:05:45.160 --> 0:05:48.080 there without AI or would have taken much much longer. 0:05:48.360 --> 0:05:51.200 So the go to market is actually direct partnerships with 0:05:51.279 --> 0:05:56.479 these companies to fast forward the discovery process. 0:05:57.000 --> 0:05:57.320 Denny. 0:05:57.560 --> 0:05:59.760 That speaks to the broadening out that we're starting to 0:05:59.760 --> 0:06:01.600 see the market. We're also going to talk about that 0:06:01.640 --> 0:06:03.839 the fact that we are seeing the S and P 0:06:03.920 --> 0:06:05.880 five hundred actually doing a bit of a challenge in 0:06:05.920 --> 0:06:08.440 terms of returns versus some of the mags seven names. 0:06:08.839 --> 0:06:11.560 How much does it mean that we've already juiced the 0:06:11.640 --> 0:06:14.159 valuation side of the equation for in video or is 0:06:14.279 --> 0:06:17.400 AI infrastructure still going to be the key watchword for twenty. 0:06:17.160 --> 0:06:22.599 Six Well, so far, all the data points would suggest 0:06:22.800 --> 0:06:27.000 that the investments into AI infrastructure are going to continue 0:06:27.040 --> 0:06:32.240 to be quite robust, and we would fully expect that, 0:06:32.480 --> 0:06:35.240 and you know, and so far, you know, I just 0:06:35.400 --> 0:06:37.760 looking back, you know, I mean Ed was talking about 0:06:37.800 --> 0:06:41.919 alphabet earlier in the relationship with Apple. Just watching the 0:06:42.000 --> 0:06:44.920 leap frogging effects you know with you know, open Ai 0:06:45.600 --> 0:06:49.560 and Xai and Gemini, that's very healthy and as long 0:06:49.560 --> 0:06:52.280 as we continue to see that. And then bringing back 0:06:52.320 --> 0:06:54.520 to you know, in Vidia and VideA now is you 0:06:54.520 --> 0:06:57.799 know they announced last week Ruben is fully in production 0:06:58.720 --> 0:07:01.080 that is going to have materi aerial impacts as it 0:07:01.080 --> 0:07:05.800 relates to inference. Productivity versus training and inference is the 0:07:05.800 --> 0:07:08.880 next big handoff. And so if you're a believer that 0:07:08.920 --> 0:07:11.520 we are now in this period where we're going to 0:07:11.520 --> 0:07:17.640 start seeing increasing deployment of these applications, inference is going 0:07:17.680 --> 0:07:19.440 to go through the roof and as a result, that's 0:07:19.480 --> 0:07:21.880 going to continue to sustain the investment profile. 0:07:22.200 --> 0:07:23.680 So we expected to be pretty healthy. 0:07:23.960 --> 0:07:26.520 Now, I will say when you're talking about the broader 0:07:26.600 --> 0:07:28.800 SMP five hundred, I mean one of the thesis that 0:07:28.840 --> 0:07:32.120 we've had for the last three years as AI has 0:07:32.160 --> 0:07:35.400 really started to take hold, is this is going to 0:07:35.440 --> 0:07:39.080 be very impactful across all industries. And if you believe that, 0:07:39.240 --> 0:07:41.760 and the companies that are able to get revenue lift 0:07:41.840 --> 0:07:44.240 because of deployment of AI, but then at the same 0:07:44.280 --> 0:07:47.320 time are able to drive cost efficiencies that should be 0:07:47.360 --> 0:07:51.080 pretty good for multiples. So I'm not surprised to see 0:07:51.400 --> 0:07:55.240 the market, you know, sort of anticipating of sorts, some 0:07:55.280 --> 0:07:56.800 of the benefits we might get from AI. 0:07:57.000 --> 0:07:59.560 The market's all allo got more discerning on who has 0:07:59.600 --> 0:08:01.360 to pay for the AI of structure and what that 0:08:01.400 --> 0:08:04.080 costs in terms of debt. Denny, I'm looking at again 0:08:04.200 --> 0:08:07.000 yet another number, three trillion dollars, this time coming from 0:08:07.000 --> 0:08:10.360 Moody saying how much the data center infrastructure build out 0:08:10.400 --> 0:08:12.520 is going to cost up until twenty thirty. I mean, 0:08:12.560 --> 0:08:15.720 McKinsey had already put seven trillion dollars on that number. 0:08:16.040 --> 0:08:18.120 But is there a worry as to who has to 0:08:18.160 --> 0:08:18.920 actually pay for it? 0:08:19.000 --> 0:08:21.720 Denny, Well, of course there is. 0:08:21.840 --> 0:08:24.200 You know, the good news is the majority of this 0:08:24.400 --> 0:08:27.760 is being funded by the hyperscalers, and they generate the 0:08:27.800 --> 0:08:31.720 free cash flows to actually be able to fulfill the 0:08:31.720 --> 0:08:34.440 commitments that they've made. And they're all playing the twenty 0:08:34.520 --> 0:08:36.560 year game. I mean, we've just seen that with you know, 0:08:36.600 --> 0:08:39.480 the number of you know, power agreements they're being struck 0:08:39.559 --> 0:08:44.439 between you know, Meta and Google, Microsoft and others, and 0:08:44.640 --> 0:08:47.560 so we feel pretty good there now. Now clearly, you know, 0:08:47.600 --> 0:08:49.439 the debt markets are going to have to remain healthy. 0:08:49.440 --> 0:08:52.400 We're going to have to continue to see a. 0:08:52.360 --> 0:08:56.040 Progress as it relates to you know, you know the 0:08:56.080 --> 0:08:59.280 ambitions of these companies, but there's also something else too. 0:08:59.360 --> 0:09:01.520 You know, when you look those numbers, there's also a 0:09:01.520 --> 0:09:03.960 lot of double counting in it. You really have to 0:09:04.000 --> 0:09:05.800 parse it back to get to the real meat of 0:09:06.160 --> 0:09:10.120 what's necessary. And then you know, also there are natural 0:09:10.160 --> 0:09:12.960 governors at just how fast this can get deployed. So 0:09:13.400 --> 0:09:16.160 the deployment of the capital is going to be pretty measured. 0:09:16.520 --> 0:09:18.760 And that's what makes this very different than say the 0:09:18.840 --> 0:09:22.120 dot com era, where so much capital came into areas 0:09:22.160 --> 0:09:25.800 like telecom infrastructure so fast and it was very easy 0:09:25.840 --> 0:09:27.000 to get that stuff deployed. 0:09:27.360 --> 0:09:28.120 This is hard. 0:09:28.400 --> 0:09:30.800 You look at what's going on at Stargate and Abilene, 0:09:30.880 --> 0:09:33.880 Texas and trying to put up ten gigawatts, the labor 0:09:33.960 --> 0:09:38.120 that's needed, the power, just the expertise, and you know, 0:09:38.160 --> 0:09:41.360 it's a heavy lift, and so that it'selfs a natural 0:09:41.400 --> 0:09:43.320 governor on just how fast it can get deployed. 0:09:43.440 --> 0:09:45.520 This is what we were trying to pass at CES 0:09:45.679 --> 0:09:49.520 last week. The baseline assumption very quickly, Denny, is that 0:09:49.640 --> 0:09:54.079 demand still outstrips in video and AMD's ability to supply. 0:09:54.520 --> 0:09:55.920 How does that set us up for this year? 0:09:56.880 --> 0:10:00.240 Well, that's good for infrastructure, you know, because you're in 0:10:00.280 --> 0:10:03.280 a situation like that, and even a TSMC for example, 0:10:03.600 --> 0:10:04.559 we're second topic. 0:10:04.840 --> 0:10:06.319 Yeah, yeah, where you. 0:10:06.240 --> 0:10:08.600 Know, you only have so much capacity and you can 0:10:08.640 --> 0:10:11.800 only bring up incremental founderies so quickly, and so as 0:10:11.800 --> 0:10:16.080 a result, that's really good from a pricing standpoint, particularly 0:10:16.080 --> 0:10:18.600 when you're the only game in town, and for certain 0:10:18.640 --> 0:10:21.200 aspects Nvidia is still the only game in town too. 0:10:22.120 --> 0:10:25.880 Denny Fish Janis Henderson Investors. Great to have you on 0:10:25.960 --> 0:10:28.560 all the games we are coming up. Paramount in ram 0:10:28.600 --> 0:10:30.960 sub efforts to caught a plan merger between Warner Brothers 0:10:31.000 --> 0:10:32.040 Discovery and Netflix. 0:10:32.400 --> 0:10:35.080 The latest in the media merger saga. That's next. There's 0:10:35.080 --> 0:10:45.920 a bloom Beg Tech Paramounts guidance. 0:10:45.960 --> 0:10:48.160 Well, it says it plans to nominate directors to Warner 0:10:48.160 --> 0:10:51.120 Brothers Discoveries board to vote against the approval of a 0:10:51.120 --> 0:10:53.560 merger with Netflix, of course, and the company has filed 0:10:53.600 --> 0:10:56.760 a suit to force Warner Brothers to disclose information about 0:10:56.800 --> 0:11:00.400 the proposed Netflix Warner Brothers tie ups entertained a reporter 0:11:00.440 --> 0:11:02.960 Pholix Jillette joins us on what is an ever more 0:11:02.960 --> 0:11:05.240 hostile potential takeover? 0:11:05.679 --> 0:11:08.480 So suing for more information? What do they want to 0:11:08.520 --> 0:11:10.040 really demonstrate to invest us here? 0:11:10.480 --> 0:11:12.520 I mean, at this point, you know, they're choosing to 0:11:12.640 --> 0:11:16.200 raise the hostility versus raising their bit Yes, and I 0:11:16.240 --> 0:11:20.200 think you know they've been campaigning shareholders join our tender offer. 0:11:20.360 --> 0:11:22.560 You know, they've said for weeks now our offer was 0:11:22.600 --> 0:11:26.839 better than the Netflix offer, accusing you know, the Warner 0:11:26.840 --> 0:11:30.240 Brothers board of not being transparent during the proceedings. And 0:11:30.280 --> 0:11:32.960 now they're making this move to essentially say, you know, 0:11:33.040 --> 0:11:36.200 you have to provide the information to shareholders so they 0:11:36.240 --> 0:11:39.120 can judge which is a better offer. And to do that, 0:11:39.200 --> 0:11:42.719 you know, the big most contentious issue for months now 0:11:42.760 --> 0:11:46.520 has been how do you value the cable assets that 0:11:46.559 --> 0:11:49.320 Warner Brothers has planned to spin off? Prior to selling 0:11:49.760 --> 0:11:52.800 the studios and the streaming business to Netflix. 0:11:52.960 --> 0:11:55.240 The Netflix offered twenty seven dollars a share for the 0:11:55.280 --> 0:11:59.120 studios and streaming spin off the cable networks paramount thirty 0:11:59.160 --> 0:12:01.839 dollars a share all of it, but basically assigning a 0:12:01.920 --> 0:12:05.800 value to the networks of zero dollars. This is a saga. 0:12:05.880 --> 0:12:08.560 It will play out over time. It's a really difficult 0:12:08.640 --> 0:12:11.880 question for you as editor, hear Felix, But like what 0:12:11.960 --> 0:12:15.559 happens next? Is there a real chance that Paramount can 0:12:15.640 --> 0:12:19.160 force a hostile takeover through this mechanism. 0:12:19.000 --> 0:12:19.839 They're going to try? 0:12:20.280 --> 0:12:20.520 You know? 0:12:20.679 --> 0:12:23.199 I mean I think, like you know, the shareholders at 0:12:23.200 --> 0:12:24.839 this point, a lot of them seem to be sitting 0:12:24.880 --> 0:12:27.520 on the fence, you know, in terms of who they're 0:12:27.520 --> 0:12:30.000 going to support. I think a lot of them are 0:12:30.040 --> 0:12:32.960 hoping that Paramount would come in with a higher bid, 0:12:33.080 --> 0:12:35.840 go over the thirty dollars to share. That hasn't happened yet, 0:12:36.920 --> 0:12:41.040 you know, I think again, you know, David Ellison and 0:12:41.120 --> 0:12:44.319 Paramount have been trying to advocate for this notion that 0:12:44.400 --> 0:12:46.960 like the Netflix deal isn't even going to happen. They 0:12:46.960 --> 0:12:50.960 have too many regulatory issues, you know, putting aside the 0:12:51.000 --> 0:12:53.760 fact that if Paramount offer was except that they'd have 0:12:53.840 --> 0:12:58.040 their own issues in Europe with the States. So yeah, 0:12:58.120 --> 0:13:01.360 I mean, this continues to just get more and more hostile, 0:13:01.920 --> 0:13:05.160 and this is you know, the next step is this 0:13:05.240 --> 0:13:08.400 lawsuit and a proxy fight over the director's. 0:13:08.760 --> 0:13:11.360 Bloomberg's Feenix to that, thank you very much. Another top 0:13:11.400 --> 0:13:15.240 story deep seek founder liangwen Feng's hedge fund surge more 0:13:15.280 --> 0:13:18.120 than fifty percent last year, making it the second best 0:13:18.120 --> 0:13:20.400 performer among China quant funds. 0:13:20.640 --> 0:13:21.960 The strong performance. 0:13:21.480 --> 0:13:24.360 Helped boost the cash available for deep seek itself. Let's 0:13:24.360 --> 0:13:28.200 get out to Bloomberg Tech executive editor Peter Elstrom. You know, two, 0:13:28.280 --> 0:13:30.920 there's the Bloomberg story, right, which is about a quant 0:13:30.920 --> 0:13:34.600 funds you know, amazing performance. But in the reporting it's 0:13:34.600 --> 0:13:37.600 super clear the read through that people are making is 0:13:37.640 --> 0:13:40.640 that that the returns on that the earnings from that 0:13:40.679 --> 0:13:43.600 fund's performance can be taken by deep Seak's founder and 0:13:43.760 --> 0:13:45.880 used to give deep seek a bigger budget for research 0:13:45.920 --> 0:13:47.400 and deploying its models. 0:13:48.520 --> 0:13:50.960 Right right, That's exactly true. Yeah, there, So there's a 0:13:50.960 --> 0:13:53.160 finance story here, but there's also a tech story, and 0:13:53.160 --> 0:13:56.600 the finance story is very impressive. The hedge fund high Fire, 0:13:56.679 --> 0:13:58.840 that was really the place where deep seek was born, 0:13:59.160 --> 0:14:01.800 had a great year. Returns were fifty seven percent on 0:14:01.880 --> 0:14:05.400 average across their various funds. That is very good for 0:14:05.640 --> 0:14:08.679 the founder early on one thing, as you mentioned, it's 0:14:08.720 --> 0:14:11.319 good performance compared with the other quant funds out there, 0:14:11.360 --> 0:14:14.400 but it also gives him a much bigger checkbook to 0:14:14.440 --> 0:14:16.600 be able to go out there and make new investments 0:14:16.640 --> 0:14:18.000 now that probably. 0:14:17.559 --> 0:14:19.680 Would be in Deep Seek. Deep Seek has. 0:14:19.520 --> 0:14:22.600 Gained a lot of momentum since a year ago when 0:14:22.600 --> 0:14:25.200 they had their big breakthrough dropped the bombshell that they 0:14:25.240 --> 0:14:27.880 had invented this LLM for a fraction of the cost 0:14:27.960 --> 0:14:31.400 of Opening Eyes with very similar performance. So now you 0:14:31.400 --> 0:14:34.480 could see the founder invest more money into Deep Seek, 0:14:34.800 --> 0:14:37.960 perhaps chase after a little bit of the capital intensive 0:14:37.960 --> 0:14:40.440 projects that we've seen in the US in particular, we 0:14:40.520 --> 0:14:42.960 haven't seen those kind of investments in China so far, 0:14:43.400 --> 0:14:46.040 or he could invest in other kinds of AI initiatives 0:14:46.040 --> 0:14:48.640 that may not be directly the same as Deep Seek. 0:14:48.640 --> 0:14:50.920 Because the whole claim about Deep Sek has it only 0:14:50.920 --> 0:14:53.000 cost them six million dollars, And the theory in this 0:14:53.080 --> 0:14:55.920 particular reporting is that maybe they're getting revenues of some 0:14:56.000 --> 0:14:58.960 seven hundred million dollars That could go a long way 0:14:59.080 --> 0:15:01.920 if you're thinking about the costs of doing the business here. 0:15:02.160 --> 0:15:05.680 But talk to us about why they're able to outperform 0:15:05.760 --> 0:15:07.880 so much. What is it they're betting on? How these 0:15:07.960 --> 0:15:10.560 quant funds managing to be so superior. 0:15:11.680 --> 0:15:14.720 Well, partly with Deep Seek early on. They were able 0:15:14.760 --> 0:15:16.800 to do this because there were a lot of constraints 0:15:16.840 --> 0:15:20.560 around the AI initiatives in particular, and we've written a 0:15:20.600 --> 0:15:23.160 fair bit about High Flyer in the past. They have 0:15:23.280 --> 0:15:27.160 these quantitative techniques to be able to outstrip their peers 0:15:27.200 --> 0:15:29.320 within the China market. They were the number two fund 0:15:29.320 --> 0:15:31.720 as you mentioned. Also, the China market has been on fire. 0:15:31.800 --> 0:15:34.080 There have been a lot of investments in other kinds 0:15:34.080 --> 0:15:36.520 of AI models. Deep Seak hasn't gone public yet, but 0:15:36.840 --> 0:15:39.200 two of the other AI models within the country have 0:15:39.320 --> 0:15:41.640 gone public, giving them a lot of momentum here. So 0:15:41.680 --> 0:15:43.640 the market's been very strong. They've been able to take 0:15:43.640 --> 0:15:45.800 advantage of that. But again, as you say, they now 0:15:45.840 --> 0:15:47.760 have one hundred times as much money as they used 0:15:47.800 --> 0:15:50.160 to start up Deep Seek in the early days. What 0:15:50.200 --> 0:15:51.640 are they going to do with that money? How are 0:15:51.640 --> 0:15:53.560 they going to invest it? They're not talking about this 0:15:53.560 --> 0:15:55.760 publicly of course at this point, but we would expect 0:15:55.760 --> 0:15:58.160 more things to come from High Flyer and Deep Seek 0:15:58.200 --> 0:15:58.720 in the future. 0:15:58.960 --> 0:16:00.960 I mean, just think many Max went public on Friday. 0:16:00.960 --> 0:16:02.640 It was a ut one hundred percent. It's up fifteen 0:16:02.640 --> 0:16:05.280 percent again today. Peter Elstrom on all things Asia. We 0:16:05.320 --> 0:16:08.440 so appreciate it, thank you. Coming up a bit more 0:16:08.440 --> 0:16:11.640 on Asia. Could Chinese evs actually dodge tariffs in the EU. 0:16:11.840 --> 0:16:14.520 We're going to discuss the global electric vehicle wars the 0:16:14.600 --> 0:16:18.360 Stephanie Valdez Street from Cox of Automotive that's next as. 0:16:18.200 --> 0:16:18.920 A BlueBag tech. 0:16:26.280 --> 0:16:29.920 The European Union is weighing minimum prices for evs exported 0:16:29.960 --> 0:16:31.840 to the block from China that actually. 0:16:31.640 --> 0:16:33.280 Would replace steep tariffs. 0:16:33.640 --> 0:16:35.600 Is it a sign that we've got some easy trade 0:16:35.640 --> 0:16:38.520 tensions even as the US presses Europe to. 0:16:38.440 --> 0:16:39.200 Take it off line? 0:16:39.200 --> 0:16:42.160 On Beijing here to discuss this is Stephanie Valdez Street, 0:16:42.600 --> 0:16:45.320 director of Industry Insights for Cox Automotive. That was sort 0:16:45.320 --> 0:16:48.240 of the latest news headline that we get that maybe 0:16:48.280 --> 0:16:50.520 BYD will be able to sell into Europe and not 0:16:50.560 --> 0:16:53.080 have to have huge tarifts that eats its margin. Are 0:16:53.120 --> 0:16:55.440 we going to see them able to tackle the European 0:16:55.480 --> 0:16:56.840 market and continue to scale. 0:16:58.200 --> 0:17:01.440 Yeah, you know, we look at the electric vehicle market, right, 0:17:01.560 --> 0:17:04.560 China definitely is dominating and you think about their strategy 0:17:04.720 --> 0:17:08.800 is to export, Right, They've exported millions of vehicles and 0:17:08.880 --> 0:17:11.439 they've had a lot of luck in increasing share and 0:17:11.480 --> 0:17:15.000 so the UK, it's basically, you know, it's a compromise, right, 0:17:15.119 --> 0:17:19.840 the EU, Europe needs to meet those climate goals, right, 0:17:19.880 --> 0:17:24.000 they need to increase their EV adoption. They rely on 0:17:24.119 --> 0:17:27.159 China for minerals, right, batteries. So I think it's a 0:17:27.200 --> 0:17:31.320 good compromise, and I think China will continue to increase 0:17:31.359 --> 0:17:33.800 their share in those markets, not only UK, but some 0:17:33.840 --> 0:17:35.359 of the emerging markets as well. 0:17:35.600 --> 0:17:38.399 Meanwhile, one of the biggest feedback I often get with 0:17:38.440 --> 0:17:42.040 my social is about the lack of BYD or Chinese 0:17:42.080 --> 0:17:43.680 offerings here in the United States. 0:17:43.840 --> 0:17:44.960 Is that ever going to come. 0:17:44.880 --> 0:17:46.560 To bat or they're just forever going to be shut 0:17:46.600 --> 0:17:49.879 out because it's really bets on TESLA and also pushing 0:17:49.880 --> 0:17:51.399 away against evs more broadly. 0:17:52.520 --> 0:17:54.200 Yeah, you know, I think if you look at the US, 0:17:54.280 --> 0:17:56.760 that's one thing I think both political parties can agree 0:17:56.800 --> 0:17:58.960 on is to keep China out. And so I think 0:17:59.000 --> 0:18:01.400 that's kind of where we're at now. Out in the future, 0:18:01.520 --> 0:18:03.760 you know, maybe we could see something that similar happened 0:18:03.760 --> 0:18:06.560 to the Japanese and the Koreans. They come, they build 0:18:06.600 --> 0:18:09.920 a partner and they build manufacturing plants. But for now 0:18:10.000 --> 0:18:12.400 the near term, we're not going to have those other 0:18:12.440 --> 0:18:14.720 than the ones that we already have under Geele, which 0:18:14.760 --> 0:18:17.720 is the Vulgo and pole Star. But definitely, I think 0:18:18.280 --> 0:18:21.840 consumers are seeing evs when they travel abroad and getting 0:18:21.920 --> 0:18:25.400 the experience of writing in a Chinese ozem and discovering 0:18:25.440 --> 0:18:29.200 how good quality high tech, and so I think that 0:18:29.240 --> 0:18:31.320 it would resonate with US consumers at the right. 0:18:31.200 --> 0:18:32.840 Price, Stephanie. 0:18:32.920 --> 0:18:36.720 Last week at CS Nvidia outlined a full stack hardware 0:18:36.800 --> 0:18:40.320 software solution for autonomous driving, and it very quickly raised 0:18:40.400 --> 0:18:41.359 questions about Tesla. 0:18:41.840 --> 0:18:43.919 So I asked Jensen Wong about it. Listen to this. 0:18:44.680 --> 0:18:47.640 I think the Tesla stack is the most advanced AV 0:18:47.760 --> 0:18:52.360 stack in the world, and I think the Tesla AV 0:18:52.480 --> 0:18:55.680 operations is the most advanced in the world, and I'm 0:18:56.440 --> 0:19:02.000 I'm fairly certain that they were already using them to 0:19:02.160 --> 0:19:02.800 end AI. 0:19:04.040 --> 0:19:04.720 So here's the thing. 0:19:05.000 --> 0:19:08.440 Mercedes will ship a vehicle this quarter capable of point 0:19:08.480 --> 0:19:12.520 to point hands free using in video technology. Does Tesla 0:19:12.600 --> 0:19:15.080 finally have something real in the market to worry about? 0:19:16.320 --> 0:19:17.040 I think they do. 0:19:17.119 --> 0:19:19.520 You think about some of the even Rivian's another player, 0:19:19.600 --> 0:19:21.880 right that they at their aidea they announce their plan. 0:19:22.000 --> 0:19:24.639 I think definitely there's a lot of competition, and I 0:19:24.640 --> 0:19:27.439 still think, you know, we're making some tremendous progress, but 0:19:27.480 --> 0:19:31.400 I think one of the challenges continues to be regulation safety. 0:19:31.440 --> 0:19:34.920 But I think in a video announcement with that technology 0:19:34.960 --> 0:19:38.480 to really identify those edge cases which will help with 0:19:38.680 --> 0:19:43.360 regulations more credibility. So definitely Tesla has some challenges out 0:19:43.359 --> 0:19:43.920 there for sure. 0:19:44.600 --> 0:19:46.359 The reason I find it fascinating is that if you 0:19:46.400 --> 0:19:48.480