From Marhard where Innovation of Money and Power Collie in Silicon Valley, NBN.
This is Bloomberg Technology with Caroline Hyde and Ed.
Lud Love.
Live from New York and San Francisco. This is Bloomberg Technology coming up the DOJ's in video investigation. It heats up? Is it anti competitive? We discuss after yesterday's historic selloff.
And in markets tech whitsaws is investors question whether the AI euphoria has gone too far.
And Intel's frustration about accessing money from the Chips Act. We have the Bloomberg exclusive reporting, but first we check in on these markets. After a historic sell off, I shine a light on what is the Semiconductor index the socks indexes. It's colloquially known. It had its worst sell off since March twenty twenty. Yesterday we bounced back, but only some and we're still off by six point three percent. But Ed, the nitty gritty was pretty amazing.
Yeah, one of the biggest contributing factors is in Vidia. I go straight to the bigger picture. The stock is now in the moment at a session high one point six percent. It had been down as much as almost three and a half percent in the session, So whipsaw
is the right word. Yesterday it's biggest dropped since April of this year, but at times in the session its biggest drop since March of twenty twenty, as you alluded to, and it shed the most market cap in a single session in the S and P five hundreds history.
But it's weird.
Markets are weird, and they're kind of saguine sanguine. The news reporting is key here, Caroline.
It is.
Let's get into the micro, particularly when it comes to Nvidia. It receives subpoenas from the US Department of Justice along with other companies as part of an escalating anti trust investigation. Let's bring in Bloomberg's Ian King, Ian you along with colleagues reporting about this investigation since June. What has changed now? How is it ramped up?
Yeah, it's the subpoena element of this that makes it more significant. Basically, these are legally binding requests to companies for information that may be doermaine in an anti trust investigation. And really what that signals to everybody is that we are getting perhaps closer to a formal complaint if that's what the DOJ feels it has evidence for.
So sources tell us that in Vidia received a subpoena, and others in that industry receive subpoenas.
I'm assuming to ask them what they think about in Vidia.
What I don't understand is what is the accusation from an anti trust perspective against Nvidia. Anti trust basically means you have an abuse of market power or monopoly.
But in what sense?
Yeah, I mean, that's a good question, and we won't know that until we see if and when we see a formal complaint. But our reporting says that what the DOJ is looking at is the idea of bundling. As you know, in video makes networking, it makes software, it sells models, services, and of course the chips. If you just want its chips and you don't want the rest, maybe you don't get the right amount of supply that
you want, maybe you get a higher price. That kind of thing is what we believe that they've been looking at.
We're going to get a nuanced tape, particularly on what is anti competitive, what isn't ian. But from your perspective, this comes at possibly the worst time in terms of market dynamics for Nvidio and its share price. What do you think internally they're thinking right now, I.
Mean, all we can go upon is their statement. And their statement was fairly strong, wasn't it. It was like, look, we make the best stuff. That's why people buy it. If they don't like it, they can go on by other stuff. It's got nothing to do with anything else that's going on. So they are clearly not happy with this investigation and don't believe that it has merits.
And that there is the Nvidia's statement that was issued to bloom their boom Masie King, Thank you very much. There is more to understand about an antitrust room view of Nvidia. Joining us now is Rebecca Allensworth, an anti trust professor at Vanderbilt Law School. Elements Worth's research has been cited by the Supreme Court and in White House and congressional reports. And I guess procedurally, good morning to you.
What is the significance of the Justice Department issuing subpoenas to Nvidia but also their peers in industry.
Yeah, thanks for having me today. So the procedure here is that the dj is getting more serious about their investigation. They're looking to hear from Nvidia about their sales practices, about their industry, and they're probably looking at their peers, not because their peers are in trouble, but more because they want to hear about their experience in competing with and buying from Nvidia. And this is all, you know, a ramp up to a potential lawsuit, I think is the idea behind these subpoenas.
Can you go into precedence, Rebecca, of making it hard to switch suppliers or perhaps pushing customers for exclusive deals. Is that just part of the course of doing business if you've got someone wanting to remain with you vertically, or is that ever deemed anti competitive?
It can be deemed anti competitive? Excuse me, you know, this is not that different from some of the accusations in the Microsoft case, which is one of the biggest anti trust cases of the last twenty years twenty five years. Some of the ideas behind this, the theory of this case, which of course we don't really know what the theory of this case is, but some of the ideas behind this. The complaints so far have been similar to those from Microsoft.
Exclusive dealing was the problem in the Google case that Google just lost. So these things can be anti competitive. The question is a you know, are they undertaken by a monopolist.
So plenty of.
Things that you do that are perfectly fine if you're not a monopolist become unlawful if you're a monopolist. And then the other question is are they instead providing some value to their consumers? Is there some pro competitive reason to provide your your goods in this way?
Rebecca, I wondered if I might ask you a somewhat I guess, academic question or indeed a policy question that in Nvidia would tell you they are America's darling. They're an American company, they follow American laws, and they are putting America first in terms of their leadership of an industry. Why would the Justice Department go after them in that respect?
Yeah?
So, why would we kill our golden goose?
You know, like exactly so.
Yeah.
So part of the point here is that we're not trying to kill anything. We are trying to make this market, which is obviously crucial for the AI race, it's crucial for America's sort of you know, financial world leader status. We're trying to make it competitive because competitive markets work best, and you know, this president, even just a couple of years ago, for the agencies going after in Vidia for doing something they perceive as anti competitive, and they stopped it.
This was an acquisition from twenty two and in Nvidia went on to do all that it's done over the last two years. I think that's the idea. It's it's that we want. We can't expect companies to self regulate. We want them to be innovative, we want them to work hard. And yes, we love the success that in Vidia has had go out and do more, but we have to create the competitive environment for those incentives to work properly.
And I'm sure AMD Intel would love to feel that they can be and also run a little bit more than they're currently deemed by the market. But Rebecca, when this sort of policy action, if we do get a form of investigation, at the moment, we understand it's just supeneas, it's just the grounds work. If something does occur, how much of a destruction does it end up being for businesses At the moment, we have a problem of demand outstripping supply.
So that's true, and you know, legal stuff is definitely a distraction. It's something that Meta and Google have been used to do for much longer. They have larger infrastructure, larger sets of lawyers than in Vidia does. So you know, there's a cost to this, and there's the uncertain and there's the what happened to the market. You know, this regulation,
this sort of enforcement is not totally costless. At the same time, it's really the payoff I think is really significant, and I think there not being enough chips out there is part of the problem. So what we want is a competitive market because that maximizes output, that creates the most chips for the most people. And so that is actually what's at stake here. You know, and Video is going to say no if you come after us, that
means even fewer chips for the market. But I don't think it's a zero sum game in that way.
We want to thank you Rebecca Adamsworth, professor at Vanderbilt Law School, with a deep dive on what is anti competitive and competitive. We must check in on the slight market recovery after an extraordinary sell off. Yesterday we saw the worst chip set off in four years.
We get so many details.
Will Boom makes Emily graffeo and extraordinary it was March twenty twenty was the worst day since we can go back to for the silks do what were you hearing from Macapa disiplins.
Well, it's interesting that earlier this morning it did seem like the micro was still moving the market, that news about the subpoena over at in video seem to be driving the market lower, and now we're seeing that the macro is more important, at least for the last hour. After we saw the labor market data, the Joel's data coming out weaker than expected, bond yields moving lower, stocks
moving higher. That idea here that stocks are still focused on the idea of a rate cut and any data that they get that's going to signal that the feed is going to be more dubbish is still bullish for stocks. But really, the fact that stocks have run up so much this year means that it's a delicate balance, Caroline. We can't have data showing that the economy is too weak, but we also can't have data showing that the economy is too hot. So really, now market participants are watching
for Friday, that's when we'll get the payrolls. Before we've already seen how stocks reacted to the Juelz data. We'll have to see if that continues on Friday.
So Emily, here's the thing, right, irrespective of whether the driver in a given session is macro or micro tech, concentration is a massive focus and factor for all participants. That seems to be something reflected in our market's coverage daily at the moment.
Yeah, I just wrote about a new ETF filing from Invesco, and this is going to be an ETF that tracks forty five percent of the companies by weight in the
Nasdaq one hundred. So if you look at that right now, it's going to be an ETF that just owns about fifteen companies, you know, the names Apple, Microsoft, Nvidia, Broadcom, Amazon, Really speaking to the fact that investors and also issuers on the product side are trying to capture that hunger to just track just a few names, those megacap technology stocks that have really driven the market forward. We're still
seeing an appetite to bid those names up. The risk though, is that when you buy an ETF that's so concentrated in a few names, if those names pull back, if we see like the trading that we saw this morning in Nvidia, that will really drive the fund lower. So other investors want more of an equal weighted exposure. But we are seeing a ton of filings for these ultra concentrated ETFs, particularly in the tech space.
Bloomberg's Emily Graffeo all over the markets for us here on Bloomberg Technology.
Thank you for more on the markets.
We've got to epec, Oscar, dish Gaya, Swiss quote senior market analysts and strategists, and you have the difficult task, Epek, of explaining to me what's happening. You know, I look at each index and the biggest points booster is in Nvidia, and we say that they witsored in the session, having had its biggest drop since April in yesterday's session at times the biggest drop since March and twenty twenty, which dragged the socks to its biggest drop since March and twenty twenty.
What is happening, Well, there are two major explanations in the story. The first one is the AI fatigue. So we have seen last week that NVDA reported blow out results but investors were not satisfied. That's one part of
the story. And the second part of the story is unal macroeconomic setup where FED is expected to start cutting the interest rates from September this year, cuts a few times, and that is both seeing a rotation from technology company swarden non technology pockets off the market, and because Nvidia has been one of the stars of this technology rally, and it's normal that investors are going after Nvidia when the market conditions are no longer ideal for the company.
Effect to embrace ourselves for more days like yesterday.
Yes, we do.
We do expect to have more volatility moving forward, because, again, as I said, right now, the market narrative, the market's focus is shifting from the individual corporate news, which remain robust by the way, toward ideal macro It can't mix setup, which is expected to be a bearish for technology stocks and more polish for the cyclical and the non technology
pockets of of this market. And because IT stocks like NVDA have gone well to very lofty evaluation levels, we do expect to see a sizable downside correction if investors decide to but keep this rotation.
Going the epect I do think it's important that we continue to discuss the big news item, which is Bloomberg's reporting that Nvidia received a subpoena from the Department of Justice, and I simply look at the market reaction. The stock is up near session highs now, having been down almost four percent. What does that tell you about the market's attitude towards this story in and of itself.
Well, listen, the OJ allegations. Anti trust investigations are part of the daily life of a big technology stuck. They come and they go. Sometimes they lead to a deeper investigations, sometimes they lead to legal actions that sometimes they lead to finds. Yes, yet we have seen that in the past decade, and so these investigations or finds or actions taken against the big technology companies have not stopped them from becoming bigger, bitter, and more influential.
So I think from an.
Investor perspective, the dog investigation is not the major issue. I think a major issue is rather the rising competition, for example, which could naturally start leading into mviga's huge profit margins.
IPEC, You've done a beautiful job of telling us the wise. What then, how do investors protect to any more bouts of volatility? What is the market telling you is the way they're searching to protect on the downside.
But obviously you do have options, You do have all the kind of financial products. But what investors should right now really think of it's re shuffling their portfolios in a way to adopt them for a period of loosing when I tell a policy from the FED and potentially further economic slowdown. So this is I think very much important.
Now.
Last year or so, we have seen the technology stocks as a safe haven, a safe place where investors just went and seek refuge from the higher interest phrase, and now the opposite scenario is about to happen. So I think that in all cases the technology stocks will remain under a decent selling pressure. But now it does not mean that the AI value is over. It just means that we expect to see at least some downside correction in technology stocks and the portfolios have to be reshuffled.
Accordingly, Epek, Let's finish this discussion with the calendar.
There's still more to come, right.
Broadcom is top of mind for us on the show and then the FED because it's top of mind for everyone.
What happens next.
Well, Broadcom is expected to announce strong quarterly results, and part of it is due to the growing AI demands and part of it due to a rebound that we have seen in networking equipment sales life, so a part of it is due to the wan Weir's transition from
a perpetual sales to subscription model. So we do expect Broadcom to come up and announce probably better than expected to Easels now does, unfortunately not mean that great resolves will stop bleeding in Broadcom if the macroeconomic data well doesn't support in market mood.
As in video just learned impact oscar Desk. We thank you so much, Swiss Cote, senior analyst. Coming up what we joined by Ahmit Wallia, CEO of Informatica to discuss the impact of generative AI on business and how they're working. Within video two, this is really meg technology. Now for today's AI in Action segment, let's bring in Ahmet Waalia,
CEO of AI powered cloud data management provider Informatica. Basically, you're in the business of managing processing data, cleaning it up for your clients so that they can then harness the power of general to AI. For many we are questioning the hype on a generally VAI. How are your clients using it?
Well, Calin, thanks for hosting me today.
Well, I think we're in that phase where we just heard so much news about AI where everybody's ready for AI accept their data, and that's where we come in. We're helping customers truly convert all of that messy data estate to clean it up, put it to some structure govern it so it can go into an LM to actually then give output that they can actually use, So turn that data chaos into something actionable.
You have a relationship with NVIDEA, right, which is that your enterprise data management offering is improved or powered by Vidious Accelerated computing offering. How does that relationship work?
What's it like?
Well, look, I think for GPUs or that estate to be actually put to use by our customers are doing. Is that helping our customers take all of that data state and think of customers like SSM Health where they're basically looking at their orthopedic professionals and making sure they can get the right professionals at the right place to manage their end patients.
That drive seaside that drives care.
Or insurance companies to do claim processing a lot.
Faster and more efficiently. So those are the kind of use.
Cases we are powering that ultimately allows customers to use the underlying infrastructure powered by the Hyperscalers or Invidia to actually put to use and drive what I call ROI and productivity.
Just talking of that infrastructure when you offer data integration engines you use in video GPS, you're talking to clients. How much are they seeing in video as the only game in town? How much do you think that everyone just depends on them a little too much?
Well, I mean, I think you should step back. I cannot.
I said that a couple of times, even yesterday I was talking to a bunch of customers. JENNYI it's tremendously overhyped in the short term, but at the same time tremendously underestimated in the long term. So we are very much in the early stages of value creation.
The buildout has.
Happened, We've been talking about that one, but when it actually comes to converting that build out to applications or productive use cases, that's where.
Our customers are doing the work right now.
Yeah, but now are they depending on video for that infrastructure?
Look, they depending upon a lot of infrastucture providers.
In Vidia has obviously got a huge leg up, no doubt about that. With the Hyperscalers or the other ones. The llms, the vector databases, all of those are their powers, so needless to say. But now the work is being done to say, okay, how do I build that best in class customer sentiment analysis application to truly grow revenue. That's where the productivity comes into play, or ROI comes
into play. And I see that experiments happening right now, first productions going towards the fall or early next year. And I see that in the next two years is when operationalization will happen.
At more scale.
So the buildout will be converted to use. We just have to give it some productive time.
You use the word expe and you said that the value creations in the early innings, and this is the whole story.
We see all the.
Dollars go to cap X, we don't see the dollars yet coming out. Are you seeing something tangible in terms of value creation for all the investment in AI.
I think that's where we see our customers putting the work in. Like I said at the beginning, everybody's ready for AI. The buildout has happened, the roads have been built, except your data. So I talked about customers like Westcom, Credit Union NRMA, or owners or SSM Health.
They are all doing the work to.
Either figure out how do you serve your customers better, how do you make sure the insurance claims are a lot faster and they are much more cleaner. How do you make sure that when a customer call comes in that you can actually take the call with a lot faster and a lot more informed opinion.
Of a customer. Customers are doing the work, but that takes a little bit of time. So we're seeing.
If you took a baseball analogy, I think we're at best bottom of the first for a very long game, and I do see the next twelve eighteen months to be the ones where you will actually see the live.
Use cases come out at scale.
I see that, I see the usage of our clear GPT product, customers using it.
We see the.
Tracking of where the data is going, and I see those things scaling up, which the next twelve months will be extremely exciting.
Will the productivity gains be enough to vindicate the hundreds of billions being spent?
Well, look, I think I can give you some examples of our customers. So we looked at this insurance claim provider and the claim processing that Caroline took months was converted to a week. So yes, yes, Now what we need to also be careful is that when you take that experiment and operationalize that, you also have to think about governance and compliance.
Of those cases.
So I think there is productivity in that, there's absolutely we see that, but we have to give it some time because when you put it out in scale, especially in the regulated industries, it will take some phasing. And I think that's a good thing, because we just don't want will be completely the wild West in this and include.
A bigger to shoot in front.
Of usm it Weally, a CEO of Informatica, thank you very much.
Welcome back to Bloomberg Technology. I'm Caroline Hyde in.
New York and I'm Ed Ludlow in San Francisco.
The markets are moving in real time and changing direction. We're probably off this kind of high of the session rebound when it comes to the indices or indexes, as we say at Bloomberg and.
A's that one hundred basically flat.
The socks recovering up eight tens and one percent, having had its biggest drop yesterday since March of twenty twenty. There are some specific single name stories behind that. In video is top of mind for everyone right now. Just go to the Bloomberg's website or the terminal and you're seeing video just so much coverage. But the reporting on the DOJ, according to sources issuing subpoenas to that name, the market seems pretty calm about it. Intel a slight
different story. It continues to lag down one point seven percent. It's had a choppy session. Why some Bloomberg reporting is a part of it. The Biden Harris Administration's big bet on Intel to lead a US chip making renaissance is in trouble with the company's mounting financial struggles.
This could create a.
Potential damaging setback for the country's most ambitious industry policy in decades. Bloomberg's Mackenzie Hawkins is here with the reporting, and I think in simple terms, Mackenzie sources tell us Intel's frustrated because it wants access to the money it was granted through the Chips Act. On the government side, the White House and Commis Department are frustrated because Intel is not doing what they need to in order to get access to that money.
Explain what you've learned.
So thirty thousand foot perspective here, the US set aside thirty nine billion dollars in grants to bring chip making back to the United States to revitalize domestic production of these critical electronic components. Intel is supposed to be the biggest from that program. President Joe Biden traveled to the company's facility in Chandler, Arizona, earlier this year, said eight and a half billion dollars in grants eleven billion dollars
in loans. Intel's also in line for tens of billions more in tax credits if it fully builds out this one hundred billion dollar US manufacturing promise. But the company has been in dire financial straits for some time and posted a pretty disastrous earnings report at the beginning of last month. So now, while the chip maker says we need access to this money faster, the Biden administration is maintaining the stance it has the entire time, which is
that that's a preliminary agreement. It requires significant due diligence to get to a final term sheet, and even then Intel has to hit key milestones on all four of its US projects before receiving the money.
We are anticipating hearing from Pat Gelsing and the CEO of Intel later today at a city event. He's already said, We've been working hard to address certain issues as he's drawn criticism. What is Intel likely to do next to come together and decide whether they build, whether they don't, what they need to get in place to get the money.
So Intel has a board meeting in mid September, and that's when you'll see company executives considering a range of potential ops options, including possibly spinning off their manufacturing division or potentially pausing or scaling back some of their global factory projects.
Too early to.
Say whether that could include any of their US facilities, the biggest of which are an expansion of their Arizona foundry project and also this greenfield site in Ohio that Biden has called a quote unquote field of dreams. But it's pretty safe to say that if Intel were to pair back it's US manufacturing ambitions, it's chips act ward would changed to match.
There are some details in you're reporting, McKenzie, again citing sources that the commerce department is key here, and one of the things the commicce department tried to do was convince other chip names to go for Intel and their manufacturing business, and that that was not something that the industry was open to.
Who the name's involved here.
So one of the biggest metrics for the Biden administration is figuring out if these plants are going to be commercially viable, which requires customers. And Taiwan's TSMC is unambiguously the world leader in the types of chips that Intel is hoping to make in the US, including AI chips from companies like Nvidia and AMD. Commerce Secretary Gena Raimondo reached out to executives at both of those companies and
asked them to consider manufacturing it Intel's Ohio plant. Neither of them currently plans to do so, and the other major customers that Intel has announced for its global manufacturing operations, which include Broadcom, Media Tech and Microsoft, haven't gone into full production yet either.
It's all so extraordinary because you would have thought the hunger, the desire for AI related chips as such that Intel would manage to be there. Now. On the flip side, we clearly see that hunger and need for infrastructure when it comes to AI.
You've got a great.
Piece out we've sharing KAfari talking about how Sam Altman is also talking to the US government to come together with an infrastructure build out request. How is that project going?
So, this big tens of billions of dollar vision by open AI's Altman has been in the works for months now, and we're just starting to see some details take shape in what Altman pictures as some kind of a compliment actually to the chip sect building out data centers, energy infrastructure, and other US projects with a global alliance of investors to sort of leverage the resources needed to make AI possible.
If you think about the chips that Intel and TSMC make, those are kind of the inputs for artificial intelligence, and part of the plan involves receiving investment from investors in countries including Japan, South Korea, Canada, the US, and the United Arab Emirates, which has been a major national security focus for the United States recently, as the administration wants to partner more with countries in the Golf and throughout the Global South to encourage investors there to invest in
American firms than Chinese ones, but is also worried about the risks of technology sharing and other national security concerns in the region.
We've been tracking this one for some time as you said, and the new parts is also that it's gone beyond just semiconductors. I think Sam and his team thinking a lot about energy as well. What are the other areas they're trying to flesh out.
So energy, I think is a really big one, and that's a concern for the chips ac ROLD as well.
These fabs and data centers require massive, massive amounts of energy on the scale of like multiple nuclear reactors to power just a couple of chip making facilities, and so Altman's team is thinking about things like transmission lines, whether we can get clean energy resources devoted to these projects, and also building out data centers likely to include some of their existing major corporate partners to actually run and
house these operations on American soil. The plans span several US states and would start here first and then potentially expand to a more global vision.
Mackenzie Hawkins across all the elements of AI infrastructure for US today, Thank you now. Also out of Washington, the Justice Department says it won't pressure social media companies to remove or block content when the government shares information about foreign threats to national security or elections. Now, of course, that follows a recent letter from two Congress from CEO of Meta, Mark Zuckerberg, which alleges that Facebook was in
fact pressioned by the Biden administration to censor COVID related content. Now, of course, the idea here ed was that Meta took its own decisions around certain content and decided to pull back, but Mark Zuckerberg saying he regretted some of the action being taken. Is this being the next answer to that from the DOJ?
It seems?
You know what's so interesting is, you know, it wasn't that long ago that Meta's position was it shouldn't end with us, It shouldn't be our decision. We need to like oversight, and oversight is still something that I don't think anyone's really come up with a framework for.
Meanwhile, we look at what's happening in terms of oversight of those that don't do any content administration or very little like Telegram and the CEO currently held up in France. Now coming up, we're going to be discussing AI in
many other ways. We're joined by Richard Socher, CEO and founder of you dot Com to discuss the company's brand new funding around that's next as a blue big technology time now for Talking tech first up Lift, It plans to sell some of its bike and scuter business and cut one percent of its workforce as the ride hailing
company still focuses on turning profitable. Plus, Ali Baba will finally let shoppers check out via ten cents we Chat pay, and in one of the biggest barriers between the Chinese tech chants, it will allow hundreds of millions of shoppers to select we Chat pay. As the Chinese economy struggles with a downturn and the billion dollar venture capital John andrewsen Horowitz ditching Miami just two years after opening an office in that city. A sixteen z left the space
in May. We understand because employees just weren't using it enough. It's acording to people familiar with the matter.
Ed Okay a piece of news in startups and bench capital. U dot Com, which makes an AI powered search engine and productivity tools, has raised fifty million dollars in a new funding round that includes backing from Nvidia and Duck dot Go, among others. Here for more in the news is You dot Com CEO Richard Socher himself, I have to go straight to Nvidia, you know, sizeable contribution in the round. What do you think in Video's interest in backing you dot Com.
Is thanks for having me ed. Yeah, we're very excited about this round. Led by Georgia and Videos, our second largest investor. They have seen that the future usually first comes on you dot Com, and then a lot of other folks follow. We've been the first to connect LM's to the web. We've been the first to make LA be multimodeal and have more than just text as outputs, like if you asked for stock price, you see a
stock ticker. Because as much as I love natural language processing in LMS, that's not.
Always loving terminal of course, if anything, market stats are related, but please continue.
That's right, and so they want to partner with the folks that bring the future, bring LMS into new markets, and we're of course very excited to partner with them to also see and use their technology that's moving up to stack away from just GPUs.
You've pivoted a little. When we first welcomed you on the show, Richard, we were all about AI search. Now there's some other key players getting in on AI search. I think a perplexity in the private side. You think of Google getting it backed together in the public side, what else are you trying to add into your business model so that you're not so crunched here.
Yeah, people can still use you dot Com as Google replacement, so it's not quite a pivot. But what we've realized is that when you really want to look for a tenex better experience, it's hard when people make very simple queries, you know, when you just ask like how old is Obama?
Or what's the weather tomorrow?
It's hard to do ten x better than a Google on those kinds of queries. But where we have seen a lot of traction is when companies actually start using you dot Com to be more productive for an account executive to go in and get the company brief and hence have an AI agent that then writes that brief for her or him. That's where we've seen tenx improvements, and that's where we're seeing a lot of customer traction with big companies and small startups all relying on you
dot Com store. We're leaning into that enterprise space and that productivity engine.
Bloomberg's piles Very, who leads interviews on the show here, wrote up the details of the report, and I was reading it and then listening to what you're saying just then, and what I'm hearing is business model.
I think is that fair? Richie?
You kind of looked at sir and said, I think we've got to work out what our business model is long term here?
Would you just explain that a bit?
Yeah, So we looked at how can we really differentiate from a pretty busy space, and number one is accuracy, but number two is actually looking at the customer's success that's possible, and then hearing from a lot of companies they bought some large seat licenses from some of the large vendors, or they even build their own prototypes, but then they don't see the adoption within their organization, and
so we wanted to align incentives. And now actually for our larger enterprise customers do a consumption based model rather than a per seat license model, and we're helping them actually adjust all their workflows to this new GENI world.
I have to say, it makes me think of your old employer, Richard. You achieve scientist and EVP over at Salesforce. Salesforce also simply trying to work out the revenue generation from AI and they're going all in on agents well, that's kind of exactly where you're going all in on as well. So are you working alongside Salesforce?
Are you working for Salesforce?
Will ultimately there'll be some sort of purchase.
We're very excited to be partnering with Salesforce also on this round.
They're also an investor.
We combine external data from the Internet together with company internal data that can be in a lot of different places, and then make that useful for knowledge workers. So if you want to build a company brief, or you want to start in your marketing campaign, or you want to research on a protein as a biotech engineer or a hedge fund manager, you can do all of those things
on you dot com. And then the agents essentially help you automate certain kinds of workflows that you'd otherwise manually do, And in those cases we end up actually replacing ten types of Google searches and hence provide a ten ex better experience and work that used to take you an hour or two will now take you minutes.
Richard, I want to go back to your answer on the relationship with Nvidia. You said you wanted to work with them up the stat not just with GPUs and the big story today is bloombog reporting on doj issuing subpoenas to video. But our reporter and king and explaining what's going on reflected on something that I spoke to Jensen one about last week. They don't want to just sell GPUs. They want to be the systems vendor that
includes Envy link, Ethernet and now the software. Could you just explain on the other side of the table how that works. You know, you buy a big package from video they gifted to you through the investment, what was the mechanics.
So at a very high level, you can kind of think of Jevins paradox here where a lot of things are getting more efficient. But as things get more efficient, as was invented with steam engines, but now it's with intelligence, you might think we use less of that. But actually I think as intelligence gets cheaper and cheaper, we will actually be using more intelligence and we'll realize that we
can automate even more workflows. And so from video it's just a natural evolution to move up the stack from GPUs to also offer more and more capabilities on top of that, and we as a startup who are using a lot of GPUs, are very excited to partner with them on developing these kinds.
Of technologies you don't come see Richis Searcher, Thank you very much.
In video shares, they're fluctuating this morning, coming off of its biggest one day decline in market cap on record. It shared more than two hundred and fifty billion dollars, more than now megs ABAGAIL do little. Take us through what's happening now?
Yeah, the volatility is really pretty interesting today we have had this stack. It opened sharply lower, well sharply as relative for the shares of in video because yesterday, of course down almost about ten percent, so today's earlier two to three percent decline not so much, but then higher. Let's take a look at the last five days though, because you can see a pretty steep decline here Dan fifteen point nine percent.
These are the worst five.
Days for Video going all the way back to September of twenty twenty two.
And what's interesting.
About this, They of course reported earnings last week.
Yes, yes they boosted, but.
It was a lot less than what it might normally be a bit of a disappointment for a stock that at one point year to date this year was up more than one hundred and eighty percent. That's a way of saying too far, too fast. So this week, especially yesterday, there's that big decline about ten percent, the big market cap shed that you were just talking about, Caroline. In fact, let's put that into context. And one thing I will say here it is on a split adjusted basis, yesterday
was the worst, say since April. So if we had not had the I think it was a ten for one split back in June, this market cap loss would not have been as great. But the split did happen, so it is as great. Take a look at that decline, two hundred and seventy nine billion dollars worth of market cap loss yesterday alone. We put that into the perspective, that is more than the value of Chevron today. That
is more than the value of Adobe today. And if you think about Boeing that has a market cap now around one hundred billion dollars, it's more than two Boeings. That is a big, big decline on yesterday's trading session.
For in video.
The question is whether or not this selling is done.
I suspect not.
If we take a look at the technicals, this chart is a little bit wonky. But some people are talking about whether or not in video is going to go sub one hundred or hit one hundred. I would say one hundred's generous based on this chart. This is a one year chart. You can see last year the stock had been on a split adjusted basis closer to forty. We are now seeing a bit of reversal here. This
entire area. Caroline likely to take this stock down closer to seventy, possibly even forty ed That would be quite declined, something to stay tuned for if it does happen.
Bin Bag exact, bigail do little with some of the best size and scope I've seen out there for a long time. Another big story, Elon must satellite internet provider Starling reverse course and said it would comply with an order by Brazil's top court to block access to X, the billionaire's social network Bloombus. Kirt Wagner has no shortage of things to write about. He's already written the book.
On Twitter you track it daily. But you know this is a concession by Elon Inc. In the country of Brazil.
Absolutely, I mean, this is his internet service provider choosing to block you know, his social network. Right, So one part of his empire is sort of working against another, and that is because of the pressure applied by you know, Brazilian regulators and the judge there to block those bank accounts. So we're seeing the leverage that they used on Starlink actually impact x here.
I like the way you call it Musk's constellation of businesses, and it can in many ways be deeply advantageous and in others make it far.
More complex to do business in certain countries.
Absolutely, I mean, I think the benefits right. We've seen investors flock towards anything Elon does in hopes of perhaps you know, getting on board with the spacexipo that will eventually come. We've seen him share resources among these companies.
Right.
The fact he has all these businesses has helped him to become the rich, powerful person that he is. At the same time, we're seeing right now live in Brazil. The downside of this, right, which is that a company or excuse me, a country or regulators can apply pressure on one part of his empire to get to another, right, And I don't think Brazil is the only country where certainly this could happen. And now that we're seeing it
play out and sort of work a little bit. It's the kind of thing that I could see happening more regularly going forward.
You right in the piece that basically the more companies you must associated with, almost the more vulnerable he is, and they have different points of entry. I just go further with that, explain it.
Yeah, Well, I mean good example would be, you know Tesla in China, for example, X does not operate in China right now, but you could imagine a world in which the Chinese government, you know, trying to get to X or another part of Elon's business is going after Tesla, right. So, just the more touch points he has, the more ways people can get after him.
Now, Court Wadna, we appreciate it so much. Thank you on all things X. But meanwhile, that does it for this edition of Bloomberg Technology. CO follow us on X and plenty of other ways.
Ed Is.
The story continues on video.
Yeah for a short week because of the US holiday Monday. It is astonishingly pat with news. Recap all of that on the podcast Apple, Spotify, iHeart wherever you get your podcasts of course, on the Bloomberg platforms as well. From New York City and here in San Francisco, Happy Wednesday, Keep it going.
This is Bloomberg Technology.