WEBVTT - Google, Blackstone to Create AI Cloud Firm 0:00:02.520 --> 0:00:13.399 Bloomberg Audio Studios, podcasts, radio news. Bloomberg Tech is alive 0:00:13.440 --> 0:00:17.239 from coast to coast with Caroline Hide in New York 0:00:17.560 --> 0:00:19.520 and ever Low in sentrancs go. 0:00:22.680 --> 0:00:25.480 This is Bloomberg Tech coming up. Google agrees to create 0:00:25.520 --> 0:00:28.640 an AI cloud business with Blackstone, which will run on 0:00:28.680 --> 0:00:30.520 Google's homegrown AI chips. 0:00:30.600 --> 0:00:32.880 Plus, the legal battle between Elon Musk and his Open 0:00:32.880 --> 0:00:36.120 Ai co founders ends with a jewelry ruling Musk waited 0:00:36.200 --> 0:00:38.600 too long to soon. We'll dive into the details of 0:00:38.640 --> 0:00:39.120 the verdict. 0:00:39.920 --> 0:00:43.560 A Meta Betts Big on Louisiana for the world's biggest 0:00:43.600 --> 0:00:48.080 AI facility. This is It plans AI related layoffs this week. 0:00:48.720 --> 0:00:52.400 AI related layoffs a theme for the market, but AI 0:00:53.159 --> 0:00:56.240 lack of exuberance pounding the market once again with actually 0:00:56.280 --> 0:00:59.160 yields pushing higher Riskill focused on geopolitics, said, but we're 0:00:59.160 --> 0:01:01.080 looking at a third straight day of losses on the 0:01:01.120 --> 0:01:03.000 laws that one hundred or off five out a percentage point. 0:01:03.000 --> 0:01:05.560 We're off about three percent in those three training days. 0:01:05.600 --> 0:01:07.399 As we just get a little bit of worry that 0:01:07.440 --> 0:01:09.440 we went too far, too farst on the hardware side 0:01:09.440 --> 0:01:11.800 of things, I shine a light on the Semiconductor Index. 0:01:12.040 --> 0:01:13.920 The SOCKS is coming off by two percent. In fact, 0:01:14.160 --> 0:01:16.560 we're down by eight nine percent. We're on the cusp 0:01:16.800 --> 0:01:19.680 of some sort of well recalibration of where we've been 0:01:19.720 --> 0:01:21.759 in this run up. Yes, it was about seventy percent 0:01:21.760 --> 0:01:23.480 from the end of March, but now are we on 0:01:23.520 --> 0:01:25.640 the verge of a correction. We're down almost eight or 0:01:25.720 --> 0:01:28.520 nine percent, as I say, but a rotation. Some money 0:01:28.520 --> 0:01:30.680 moving into the software stocks. You've got key earnings coming 0:01:30.680 --> 0:01:34.320 from software and hardware over the next few days. 0:01:34.440 --> 0:01:35.839 Neo clouds down. 0:01:36.080 --> 0:01:40.039 The news Blackstone and Google forming a new company to 0:01:40.120 --> 0:01:44.120 be a neocloud. Blackstone brings five billion dollars. There's debt financing. 0:01:44.400 --> 0:01:48.680 Five hundred megawats is the target. Google's TPUs tensor processing 0:01:48.760 --> 0:01:51.280 units are the chips. And look at those names that 0:01:51.320 --> 0:01:54.920 are down Core weave, Ironebuus in particular, all very Nvidia 0:01:55.000 --> 0:01:58.560 focused architectures. But neoclouds all the same. Competition seems to 0:01:58.560 --> 0:01:59.640 be coming it does. 0:01:59.680 --> 0:02:02.200 That's scuss the impact of Google blackstory. The partnership ed 0:02:02.360 --> 0:02:05.200 Brie Big Intelligence says this pact could be a major 0:02:05.200 --> 0:02:08.760 boost for the tech giants. Chips business Managsing, who wrote 0:02:08.840 --> 0:02:11.720 that React joins us now. And what's so interesting is, yes, 0:02:11.760 --> 0:02:13.799 the equity comes from Blackstone five billion, it could be 0:02:13.840 --> 0:02:15.920 leveraged up to twenty five billion dollars. But this is 0:02:15.960 --> 0:02:19.040 just another way for Alphabet to get its hardware out 0:02:19.040 --> 0:02:21.000 there but not actually have to build the data centers. 0:02:21.000 --> 0:02:21.600 It feels like. 0:02:22.800 --> 0:02:26.720 Yeah, and look, when it comes to Nvidia, their ecosystem 0:02:26.840 --> 0:02:29.680 of neo cloud players is pretty big. I mean, you 0:02:29.760 --> 0:02:32.120 had a few of those on the on your screen. 0:02:32.200 --> 0:02:34.360 And when it comes to Google, you know, you have 0:02:34.480 --> 0:02:37.880 got terrible cipher mining, but there aren't a lot of 0:02:38.800 --> 0:02:41.919 you know, TPU providers. And that's where you know, when 0:02:41.960 --> 0:02:45.920 Anthropics says they want to use almost two hundred billion 0:02:46.000 --> 0:02:50.400 dollars of capacity from Google, Google has to find a 0:02:50.440 --> 0:02:53.000 way to you know, wrap that up outside of their 0:02:53.080 --> 0:02:56.760 own cloud. And that's where I think this sort of 0:02:56.960 --> 0:03:01.040 a transaction is quite interesting. Where Blackstone is putting the 0:03:01.080 --> 0:03:04.280 money and also has got live capacity that they can 0:03:04.320 --> 0:03:07.640 bring online by twenty twenty seven, which is what makes 0:03:07.639 --> 0:03:09.880 this very attractive from a Google standpoint. 0:03:10.680 --> 0:03:13.919 Hey, Mandy, what's the Bloomberg intelligence thesis on the business 0:03:13.960 --> 0:03:16.560 model of setting up a neocloud for the TPU? 0:03:17.000 --> 0:03:17.080 Right? 0:03:17.120 --> 0:03:20.440 Because Google Cloud, you know, they have capacity that they 0:03:20.440 --> 0:03:23.120 could get to third party customers. Why set up a 0:03:23.160 --> 0:03:24.480 new standalone new code. 0:03:25.880 --> 0:03:26.120 Yeah. 0:03:26.160 --> 0:03:28.560 So in the case of Google Cloud, you know, they 0:03:28.600 --> 0:03:32.560 are serving that traffic from Google's data centers. I mean 0:03:32.600 --> 0:03:36.160 in this case, you know, having a neocloud provider use 0:03:36.680 --> 0:03:41.080 Google TPUs and then everything else is taken care of 0:03:41.480 --> 0:03:43.960 by a blackstown in this case, whether it's you know, 0:03:44.120 --> 0:03:47.120 the power or the cooling or the you know, even 0:03:47.160 --> 0:03:50.960 getting the land for that data center. So from that perspective, 0:03:51.200 --> 0:03:55.440 having that live capacity is what you know, makes this 0:03:55.600 --> 0:03:58.800 different from Google having to do it on their own, 0:03:58.840 --> 0:04:01.040 which will be a lot more aggreed in terms of 0:04:01.040 --> 0:04:04.040 how much gigawad they can bring online in twenty twenty six, 0:04:04.320 --> 0:04:07.440 twenty twenty seven and beyond, whereas this would come life 0:04:07.840 --> 0:04:10.760 much faster given the involvement of blacksture. 0:04:11.240 --> 0:04:14.240 Man deep Seeing, who leads technology research at Bloomberg Intelligence. 0:04:14.240 --> 0:04:16.280 Thank you very much. Let's take it today. A look 0:04:16.320 --> 0:04:19.960 at today's big number. Seven thousand. That's how many workers 0:04:20.520 --> 0:04:24.239 Medas resigning to new jobs related to AI. That's according 0:04:24.240 --> 0:04:26.440 to an internal memo. This comes on the eve of 0:04:26.440 --> 0:04:30.120 a ten percent staff cut tomorrow to improve efficiency and 0:04:30.160 --> 0:04:34.080 quote offset metas investments in AI. AAR related job cuts 0:04:34.160 --> 0:04:36.120 are the talk of Wall Street right now. Is Standard 0:04:36.200 --> 0:04:39.520 Chartered CEO Bill Winters also saying his company plans to 0:04:39.560 --> 0:04:42.720 cut more than fifteen percent of its supports staff by 0:04:42.760 --> 0:04:47.160 twenty thirty, aiming to replace quote lower value human capital 0:04:47.560 --> 0:04:51.080 with AI. Martin Norton is MPOWER Chief Investment Strategists, and 0:04:51.360 --> 0:04:54.320 I'm going to ask you straight up, it's an interesting 0:04:54.440 --> 0:04:59.160 balancing act cutting rolls and saying that this is not 0:04:59.200 --> 0:05:04.320 necessarily a trim downsize, it's a reallocation a priority to AI. 0:05:04.920 --> 0:05:07.640 As an investor, what signal does that give you? 0:05:08.640 --> 0:05:11.640 Well, it's not the only piece of data that we 0:05:11.760 --> 0:05:14.000 have as companies do that. We hear a lot of 0:05:14.040 --> 0:05:18.680 different announcements of companies making reductions to their workforce related 0:05:18.720 --> 0:05:22.720 to AI. I think a lot of these are idiosyncratic. 0:05:23.000 --> 0:05:25.320 When we're looking at the broad economy at the moment, 0:05:25.760 --> 0:05:29.479 we know that the broad economy added labor upon labor 0:05:29.640 --> 0:05:32.159 and the wake of the pandemic, and now there's a 0:05:32.200 --> 0:05:34.719 bit of right sizing, and so in some cases, I 0:05:34.800 --> 0:05:38.760 do think companies are adding the AI moniker to make 0:05:38.800 --> 0:05:42.839 it feel very forward looking to maybe AI whitewash it, 0:05:43.760 --> 0:05:46.960 but not necessarily a sign of the times in terms 0:05:46.960 --> 0:05:49.960 of very broad layoffs across the economy. In fact, one 0:05:49.960 --> 0:05:52.560 of the areas that I'm looking more closely for signs 0:05:52.600 --> 0:05:54.839 of an AI impact is more on the hiring side, 0:05:54.920 --> 0:05:57.279 where people are maybe slower to higher than they otherwise 0:05:57.320 --> 0:05:59.839 would because of AI. But yes, there are these idiosyncratic 0:06:00.120 --> 0:06:02.640 reports that suggest we are seeing a little bit of 0:06:02.640 --> 0:06:03.919 movement on that front. 0:06:04.160 --> 0:06:09.200 I'm sorry, language matters, lower value, human capital, the extraordinary 0:06:09.279 --> 0:06:13.239 choice of words, but Marta putting that to one side, 0:06:13.400 --> 0:06:16.640 there is anxiety out there of labor and people who 0:06:16.640 --> 0:06:21.279 are affected. But when you're thinking about how we see 0:06:21.600 --> 0:06:25.200 actual businesses lean into this moment, are we still seeing 0:06:25.240 --> 0:06:28.320 adoption taking root in the way you anticipated? Are we 0:06:28.400 --> 0:06:31.440 seeing the people the label that is able to use 0:06:31.480 --> 0:06:33.280 AI being more productive with it? 0:06:34.240 --> 0:06:36.760 You know, I think those early signs that are pretty 0:06:36.760 --> 0:06:40.400 positive in terms of adoption by current employees. So I 0:06:40.400 --> 0:06:43.240 think everyone at this point is using the chatbots on 0:06:43.279 --> 0:06:47.080 a regular basis to do, you know, low value research. 0:06:47.520 --> 0:06:51.239 But I think the bigger question is this AI agent 0:06:51.400 --> 0:06:54.799 deployment what that looks like? And you are seeing companies 0:06:55.360 --> 0:06:58.440 talking about that in real time and beginning to structure 0:06:58.520 --> 0:07:01.280 workflows around it. I think we're very early obviously on 0:07:01.320 --> 0:07:03.320 the adoption of that. But that is where I think 0:07:03.360 --> 0:07:10.040 a lot of this AI productivity can stem from. 0:07:08.640 --> 0:07:12.000 Modern the market's trading, in part in anticipation of Nvidia. 0:07:12.400 --> 0:07:16.200 After the closing bell Wednesday. Yesterday, I spoke to Nvidia 0:07:16.280 --> 0:07:19.040 CEO Jensen Wang. This is how he frames the state 0:07:19.080 --> 0:07:19.480 of play. 0:07:20.440 --> 0:07:23.000 We have the largest supply chain in the world. Our 0:07:23.040 --> 0:07:27.040 partners have done a great job securing supply for us, 0:07:27.360 --> 0:07:30.200 and so all of the pieces go together. The silicon, 0:07:30.240 --> 0:07:33.520 photonics is lined up, Everything is all lined up. It's 0:07:33.600 --> 0:07:38.080 just that the demand is much greater than the overall capacity. 0:07:37.560 --> 0:07:38.000 Of the world. 0:07:39.240 --> 0:07:43.600 The demand is far out pacing this industry's ability to supply. 0:07:43.680 --> 0:07:46.840 When you go into an earnings how do you decide 0:07:46.880 --> 0:07:49.800 therefore the bar in terms of what is positive and 0:07:49.800 --> 0:07:53.720 what is negative. If you're in a supply constraint world. 0:07:54.520 --> 0:07:56.680 Well, I think you do want to see signs that 0:07:56.720 --> 0:07:59.080 they are monetizing it. Right, If you're in a supply 0:07:59.240 --> 0:08:02.480 constrained world, that demand should be moving earnings and you 0:08:02.520 --> 0:08:05.240 want to see some evidence of that. But of course, 0:08:05.280 --> 0:08:07.920 when we're talking about the supply constraint. We're also talking 0:08:07.960 --> 0:08:12.040 about supply constraint and areas that would infect the likes 0:08:12.080 --> 0:08:14.280 of n videos so memory checks, for example, And what 0:08:14.320 --> 0:08:16.920 does that mean for gross margins and how are they 0:08:17.080 --> 0:08:19.960 navigating that? I think those are important questions at the moment, 0:08:20.200 --> 0:08:21.960 but I think the bigger picture, and I caught that 0:08:22.200 --> 0:08:25.000 that interview yesterday. I think the bigger picture that Jensen 0:08:25.040 --> 0:08:28.280 Wong is pointing to is this massive supply and demand 0:08:28.320 --> 0:08:31.600 mismatch that we'll carry through not just this earnings report, 0:08:31.720 --> 0:08:35.400 but for several quarters years to come as this all 0:08:35.400 --> 0:08:38.079 plays out. So I really agree with his comment that 0:08:38.160 --> 0:08:41.199 this is inning one, Inning two on the AI build. 0:08:42.040 --> 0:08:43.760 Luckily we all call that into you. It was a 0:08:43.760 --> 0:08:45.840 great one with Ed and Mountain Norton. We so appreciate 0:08:45.880 --> 0:08:47.920 having you on the show today of Empower. Thank you 0:08:47.960 --> 0:08:51.720 for your time. Coming up a bitter feud, personal grievances, 0:08:51.760 --> 0:08:54.520 billions of dollars. The decision comes down to timing. We'll 0:08:54.520 --> 0:08:57.120 discuss the jury's rejection of El Musk's claims against his 0:08:57.200 --> 0:09:14.680 Open AI co founders, and what comes next the decision 0:09:14.760 --> 0:09:17.080 in that closely watched legal battle between Ela Musk and 0:09:17.080 --> 0:09:20.480 his open ai co founders came down to timing. A 0:09:20.559 --> 0:09:22.800 jury rule that Musk waited too long to sue of 0:09:22.880 --> 0:09:26.080 open AI's transition to a for profit business. Mugs Madelon 0:09:26.080 --> 0:09:28.400 and Macklburg has been following the case every step of 0:09:28.440 --> 0:09:30.360 the way. You join us now, Madeline. What did the 0:09:30.440 --> 0:09:33.120 jury decide? What didn't it decide on? 0:09:34.840 --> 0:09:34.960 Right? 0:09:35.080 --> 0:09:37.480 It was a bit of an anti climactic verdict, if 0:09:37.480 --> 0:09:40.199 I can say that, because the jury didn't actually make 0:09:40.240 --> 0:09:43.360 a decision on any of the claims that Elon Musk 0:09:43.440 --> 0:09:46.720 presented over the course of this lawsuit. What they said 0:09:46.760 --> 0:09:50.400 instead was that he had not proven his case within 0:09:50.480 --> 0:09:53.600 the statute of limitations set by law for the specific 0:09:53.640 --> 0:09:57.240 claims that he brought. So essentially, they found that he 0:09:57.320 --> 0:10:00.160 had concerns about this conduct and should have known about 0:10:00.200 --> 0:10:04.000 this conduct years before he actually filed his lawsuit, and 0:10:04.040 --> 0:10:06.600 so therefore the case that the jury was hearing should 0:10:06.640 --> 0:10:08.520 be tossed out, which is what happened. 0:10:09.840 --> 0:10:12.440 We're going to get an appeal, according to mister Mask, 0:10:12.480 --> 0:10:14.520 who posted on x that he would appeal. I believe 0:10:14.640 --> 0:10:17.840 like his legal team outside the courtroom said that we 0:10:17.880 --> 0:10:20.800 would appeal. If that's the case, what do we know 0:10:21.120 --> 0:10:23.679 about what that appeal would be based on From a 0:10:23.760 --> 0:10:25.320 legal perspective. 0:10:26.280 --> 0:10:28.800 Right, it should come as no surprise to anybody who 0:10:28.960 --> 0:10:31.520 follows Elon Musk as a litigant that he plans to 0:10:31.640 --> 0:10:35.439 appeal this and fight this vigorously. We saw him celebrating 0:10:35.520 --> 0:10:39.160 on X afterwards saying that, you know, just because they 0:10:39.160 --> 0:10:42.040 decided on the statute of limitations, they never actually touched 0:10:42.080 --> 0:10:44.520 the merits of the case. I think an appeal on 0:10:44.800 --> 0:10:47.200 this ground is a lot more narrow than if the 0:10:47.280 --> 0:10:49.480 jury had actually ruled on the. 0:10:49.480 --> 0:10:50.920 Merits of his claims. 0:10:51.240 --> 0:10:53.040 So I think a lot It still kind of remains 0:10:53.040 --> 0:10:55.120 to be seen of how this is going to play out, 0:10:55.120 --> 0:10:56.720 but it's going to be an issue that the Ninth 0:10:56.720 --> 0:10:58.720 Circuit Court of Appeals is going to have to sort 0:10:58.760 --> 0:11:00.200 through when he eventually files that. 0:11:01.160 --> 0:11:04.480 Bloomberg's Madeline Meckelberg, thank you very much. Let's dive more 0:11:04.480 --> 0:11:07.480 into the legal considerations. Dorothy Lund is with us, a 0:11:07.520 --> 0:11:11.240 professor at Columbia Law School focused on corporate law, governance 0:11:11.280 --> 0:11:13.960 and contracts, returning to the show and has been covering 0:11:13.960 --> 0:11:17.000 this trial with us. In our Bloomberg News story, we 0:11:17.160 --> 0:11:21.559 write that mister Musk was too late in his suit, 0:11:21.679 --> 0:11:25.439 and we later down in the story explain the idea 0:11:25.559 --> 0:11:28.560 that he had knowledge in twenty twenty four and so 0:11:28.679 --> 0:11:31.400 basically should have filed his suit sooner. But I've got 0:11:31.440 --> 0:11:33.320 a lot of sympathy with people around the world that 0:11:33.360 --> 0:11:37.120 don't understand the legal technicality of that. The Statute of 0:11:37.120 --> 0:11:40.559 limitations seems to play a role here. Is this codified 0:11:41.360 --> 0:11:45.040 or is this the jury's discretion in deciding that mister 0:11:45.120 --> 0:11:46.200 Musk ran out of time? 0:11:47.679 --> 0:11:48.320 Yeah, the jury. 0:11:48.440 --> 0:11:52.720 So the jury had to basically answer this question of 0:11:52.760 --> 0:11:55.439 whether or not the Statute of limitations applied to the 0:11:55.480 --> 0:11:57.520 two claims that were brought here. So the two claims 0:11:57.559 --> 0:12:01.840 is to remind everybody or there was a violation that 0:12:02.000 --> 0:12:06.400 open ai, Brockman and Altman violated a promise that was 0:12:06.440 --> 0:12:08.880 made to Musk that it would have a permanent charitable 0:12:09.160 --> 0:12:13.040 charitable mission to develop safe open source AI technology, and 0:12:13.080 --> 0:12:17.000 that they received undeserved benefits because of those broken promises. 0:12:17.600 --> 0:12:20.760 And so then open ai said, well, you waited too 0:12:20.840 --> 0:12:25.080 long to sue. You know, you brought your case on 0:12:25.240 --> 0:12:29.440 August twenty twenty four, and the statute of limitations was 0:12:29.480 --> 0:12:32.719 only three It was three years, which meant the things 0:12:32.760 --> 0:12:35.160 that you're complaining about that took place in twenty eighteen, 0:12:35.320 --> 0:12:39.520 twenty nineteen, you're too late. 0:12:40.120 --> 0:12:40.880 Well, I think that is. 0:12:41.240 --> 0:12:44.840 Surely, did Elon Musk and his team of very expensive 0:12:44.920 --> 0:12:48.240 lawyers not understand the law or something when they followed 0:12:48.280 --> 0:12:48.640 the suit? 0:12:49.480 --> 0:12:50.120 Nosk? 0:12:50.200 --> 0:12:53.400 You know, so Musk of course had these fabulous lawyers, 0:12:53.960 --> 0:12:57.960 and you know that the statute of limitations is told 0:12:58.080 --> 0:13:01.440 if you've concealed, if harm has been concealed, or you 0:13:01.480 --> 0:13:05.240 haven't discovered the harm. And so his argument was, well, 0:13:05.280 --> 0:13:07.559 you know, Sam Altman and open Ai have been lying 0:13:07.600 --> 0:13:10.439 to me. They concealed that they were doing all this stuff. 0:13:10.440 --> 0:13:14.160 It wasn't until twenty twenty three that I really understood 0:13:14.280 --> 0:13:16.640 what was going on. And so this and this gets 0:13:16.640 --> 0:13:19.160 back to your original question, which is why did this 0:13:19.240 --> 0:13:20.000 go to a jury? 0:13:20.120 --> 0:13:20.360 Right? 0:13:20.760 --> 0:13:24.000 The jury had to sift through testimony from both parties, 0:13:24.520 --> 0:13:27.760 documents and to try to understand this timing question. Okay, 0:13:27.800 --> 0:13:31.120 so when was it that Musk knew or reasonably should 0:13:31.120 --> 0:13:34.400 have known about the harm that he was claiming happened. 0:13:36.200 --> 0:13:40.720 And instead we all get consumed with billionaire versus billionaire 0:13:41.040 --> 0:13:43.680 and how much equity they own in the underlying company 0:13:43.960 --> 0:13:46.240 and what they're like in terms of managers or how 0:13:46.280 --> 0:13:48.920 trustworthy they are at the end of the day, when 0:13:48.920 --> 0:13:52.000 we misguided to think that when they go back to 0:13:52.040 --> 0:13:54.600 an appeals court, will it be argued in a different way. 0:13:56.240 --> 0:14:00.199 Yeah, So it's a little disappointing that we didn't get 0:14:00.200 --> 0:14:03.360 an answer on the big question here, right, that the 0:14:04.200 --> 0:14:07.200 claims that I was telling you about just now are 0:14:07.400 --> 0:14:10.960 are unaddressed, and in this appeal we won't get any 0:14:11.000 --> 0:14:13.679 resolution on this either. So you know, this is going 0:14:13.720 --> 0:14:17.679 to be a narrow appeal on this decision of the 0:14:17.720 --> 0:14:21.440 statute of limitations. And because that statute of limitations question 0:14:22.000 --> 0:14:26.000 turned on a factual determination, you know that is going 0:14:26.040 --> 0:14:29.440 to be really hard to be overturned. Public courts rarely 0:14:29.560 --> 0:14:34.080 overturn a jury's fact intensive findings, especially in this case 0:14:34.120 --> 0:14:37.280 where the judge also reviewed the evidence and said I 0:14:37.320 --> 0:14:40.520 came to the same conclusion. And so the only path 0:14:40.560 --> 0:14:43.600 four from USK is going to be arguing a different 0:14:43.600 --> 0:14:46.280 set of things about you know, the argument would have 0:14:46.320 --> 0:14:54.080 to be some error that proud come here, improper instructions 0:14:54.240 --> 0:14:56.760 or improper evidentiary rulings. 0:14:57.080 --> 0:14:59.360 We're having some technical issues, or if you with your zoom. 0:14:59.440 --> 0:15:03.240 So give the Internet gremlins a second to reset. But 0:15:03.280 --> 0:15:05.160 my question to you is exactly that, I don't want 0:15:05.200 --> 0:15:06.040 you to speculate. 0:15:06.720 --> 0:15:09.240 But what options does mister. 0:15:09.080 --> 0:15:11.320 Musk have if he needs to go to the appeals 0:15:11.360 --> 0:15:14.600 court but argue something completely different to the original case. 0:15:16.360 --> 0:15:19.360 Yeah, his argument is going to have to be that 0:15:20.040 --> 0:15:25.120 there were significant legal or procedural errors that prejudice the outcome. 0:15:25.600 --> 0:15:29.080 You know that there the instructions to the jury were improper, 0:15:29.960 --> 0:15:33.840 their evidentiary rulings were wrongly decided. There won't be any 0:15:34.280 --> 0:15:39.880 discussion on the substance of this larger question of whether 0:15:39.960 --> 0:15:43.080 or not Opening I violated its charitable purpose when it 0:15:43.120 --> 0:15:45.920 restructured to create a for profit affiliate. 0:15:46.080 --> 0:15:49.400 Called a public opinion. Therefore, in many ways, Dorothy Lund, 0:15:49.480 --> 0:15:52.720 Professor at Columbia Law School, fantastic to have you. Thank you. 0:15:52.800 --> 0:15:56.240 Coming up, Parallel is moving past the era of flat 0:15:56.280 --> 0:16:00.160 f AI licensing deals. The CEO Arak Agrawal is me 0:16:00.280 --> 0:16:02.640 joining us. Next this is blue bag Tech. 0:16:16.760 --> 0:16:20.000 Parallel, a web API purpose built for AI and led 0:16:20.000 --> 0:16:23.400 by the FORMACYO of Twitter, is launching Index today, a 0:16:23.440 --> 0:16:27.600 new marketplace that pays publishers and data providers based on 0:16:27.640 --> 0:16:31.000 how much their content actually helps an AI agent completed task. 0:16:31.400 --> 0:16:34.960 Joining us is Parallel, founder and CEO Para Agrowl. So, 0:16:35.000 --> 0:16:37.200 in answer to the question what did you get done 0:16:37.240 --> 0:16:43.160 this week? You launched Index in this era of agentic right, 0:16:43.720 --> 0:16:48.120 the agents need the web data and the proprietary data sets. 0:16:49.200 --> 0:16:52.520 I guess saw a gap where the marketplace for that 0:16:52.600 --> 0:16:54.560 data doesn't exist. Then now you think it does. 0:16:55.520 --> 0:17:00.520 Yeah, we've been linked towards this content marketplace for the 0:17:00.560 --> 0:17:02.760 last two and a half years. We've started Parallel with 0:17:02.800 --> 0:17:05.919 the notion that agents will use all of the content 0:17:05.920 --> 0:17:09.280 on the web thousand x more than humans ever have. 0:17:09.840 --> 0:17:13.200 And when that happens, the technology that we've built over 0:17:13.240 --> 0:17:16.640 the last couple of decades isn't enough. But the business 0:17:16.680 --> 0:17:20.760 models also break down. Adds, subscriptions, these business models no 0:17:20.840 --> 0:17:25.199 longer work, and that I think is the opportunity for 0:17:25.280 --> 0:17:27.960 us to figure out how to align the interests of 0:17:28.280 --> 0:17:32.320 AI agents doing real work along with the content owners 0:17:32.359 --> 0:17:35.680 who are creating valuable content that feeds these agents. 0:17:36.400 --> 0:17:38.040 And that's what Index is all about. 0:17:37.920 --> 0:17:40.159 To help the Boomberg Tech audience out. The question what 0:17:40.240 --> 0:17:41.879 did you get done this week? Was posed to you 0:17:41.920 --> 0:17:45.919 by mister musk just before you left Twitter. How do 0:17:45.920 --> 0:17:48.560 you assign a value then to the data? You've called 0:17:48.560 --> 0:17:52.480 this index very interesting. Right on the one side you 0:17:52.520 --> 0:17:56.280 have the agent or the entity the agent is acting. 0:17:56.080 --> 0:17:56.760 On behalf of. 0:17:57.200 --> 0:17:58.640 And then on the other side you have the source 0:17:58.640 --> 0:18:01.439 of the data who sets up price or value. 0:18:01.640 --> 0:18:04.160 Yeah, So if you think about other deals that are 0:18:04.160 --> 0:18:08.520 being done, these deals are typically between a large lab 0:18:08.880 --> 0:18:13.639 or a hyperscaler or Google alongside a large publishing house. 0:18:13.680 --> 0:18:17.600 And these are flat feed deals where over time, I 0:18:17.640 --> 0:18:20.840 don't believe content owners get to participate in the growing 0:18:21.000 --> 0:18:24.760 economy of agents. Now, what's really happening with agents is 0:18:25.200 --> 0:18:29.040 they are starting to do real work, real knowledge work 0:18:29.080 --> 0:18:31.480 in the enterprise. So if you think about our customers, 0:18:31.480 --> 0:18:35.719 our customers span people building AI scientists. Our customers are 0:18:35.720 --> 0:18:40.200 building air lawyers, customers are building for finance using agents, 0:18:40.920 --> 0:18:44.159 and all of these agents that our customers built are 0:18:44.160 --> 0:18:48.920 doing real work when they produce value. Content owners don't 0:18:48.920 --> 0:18:52.720 get to participate in that economy of the value being. 0:18:52.600 --> 0:18:54.320 Created with index. 0:18:54.359 --> 0:18:57.280 What we're trying to do is buil a system where 0:18:57.840 --> 0:19:02.480 if very valuable work is done using somebody's data, they 0:19:03.040 --> 0:19:07.960 get more money. If someone has very high quality data, 0:19:08.359 --> 0:19:11.840 they get paid more. And we've built a model to 0:19:12.000 --> 0:19:15.320 figure out what this value is, what this marginal contribution is, 0:19:15.840 --> 0:19:18.920 and it draws from this game theorty concept called Shapi values. 0:19:19.600 --> 0:19:22.639 Let's talk about the shapely value. It's contribution to the 0:19:22.680 --> 0:19:25.159 work performed by an AI agent. So basically, instead of 0:19:25.680 --> 0:19:29.280 content access crawls or citations, you consider the value. I mean, 0:19:29.880 --> 0:19:32.560 but where's the monetary number coming from? What are you 0:19:32.640 --> 0:19:36.639 pegging it against? And how how do you work out what? 0:19:36.760 --> 0:19:39.720 Then ADII agent has eventually learned itself. 0:19:39.359 --> 0:19:41.040 Off for well. 0:19:41.119 --> 0:19:43.879 Shappie values has been my obsession for the last three 0:19:43.960 --> 0:19:46.719 years as we've been working on this. So just in 0:19:46.720 --> 0:19:49.200 a sentence, what shappily values is is about like if 0:19:49.560 --> 0:19:53.399 a few people decide to collaborate on building something in 0:19:53.440 --> 0:19:55.600 a positive some way where the whole is bigger than 0:19:55.640 --> 0:19:58.359 the sum of parts. The question is how do you 0:19:58.400 --> 0:20:01.080 divide up this value? And schappely value is the concept 0:20:01.119 --> 0:20:03.679 that tells us the way to divide this value so 0:20:03.720 --> 0:20:09.199 that everyone is incentivized to participate. The net outcome of 0:20:09.400 --> 0:20:13.920 using this concept two reward content owners. It creates three 0:20:14.000 --> 0:20:18.000 really nice properties. One high quality content gets paid more. 0:20:18.480 --> 0:20:22.399 Number two, content that gets used in high value work 0:20:22.640 --> 0:20:26.600 gets paid more. And most importantly, as agents do more 0:20:26.680 --> 0:20:31.320 work and create more value, content owners grow alongside them, 0:20:31.680 --> 0:20:35.800 which we believe allows content creation on the web to 0:20:35.800 --> 0:20:36.520 be sustainable. 0:20:37.280 --> 0:20:39.639 There are a lot of content creators out there right now, 0:20:40.320 --> 0:20:42.480 So how do they sign up to INEX? How do 0:20:42.520 --> 0:20:45.120 you make sure that it's not just quite large either? 0:20:45.200 --> 0:20:46.480 Individuals who are. 0:20:46.440 --> 0:20:48.720 Writing I think of alex Etho is actually those that 0:20:48.760 --> 0:20:51.560 have got a minor substack, but with really valuable content 0:20:51.560 --> 0:20:53.280 that somehow the agent has managed to access. 0:20:53.280 --> 0:20:54.199 How do they access that? 0:20:55.440 --> 0:20:55.720 Yeah? 0:20:55.760 --> 0:21:00.000 With Index Today we're announcing collection of launch partners. These 0:21:00.080 --> 0:21:04.480 span premium news publishers like The Atlantic and Fortune to 0:21:05.160 --> 0:21:09.080 really raw, factual, amazing content from the pr Newswire, business 0:21:09.160 --> 0:21:14.119 intelligence and data providers like pitch Bork and Traction and 0:21:14.200 --> 0:21:18.800 zoom Info, as well as independent creators like Alex Heath, 0:21:19.440 --> 0:21:23.879 Azimazer as several others. The really interesting part about this 0:21:24.160 --> 0:21:30.640 is that we want everyone, all content owners big and small, 0:21:31.280 --> 0:21:33.439 to be able to participate in this economy, and we 0:21:33.480 --> 0:21:36.640 want all agents, not just once built by large providers, 0:21:36.680 --> 0:21:38.560 to be able to get access to all of this 0:21:38.600 --> 0:21:39.280 amazing content. 0:21:39.359 --> 0:21:42.240 Correct, We just have sixty seconds, but can ask you 0:21:42.280 --> 0:21:45.040 about x the platform known as Twitter. Of course, still 0:21:45.119 --> 0:21:48.399 using it, your view of it and as part of 0:21:48.520 --> 0:21:49.800 XAI is part of SpaceX. 0:21:51.200 --> 0:21:53.240 I'm still using it. I'm still calling it Twitter. 0:21:54.280 --> 0:21:57.399 I think we built a product over a decade that 0:21:57.480 --> 0:22:03.680 I was there which endures in value forever, and I'm 0:22:03.720 --> 0:22:06.000