Bloomberg Audio Studios, podcasts, radio news. Bloomberg Tech is a lie from coast to coast with Caroline Hide in New York and v lovel in San Francisco.
This is Bloomberg Tech coming up.
Mike Cron and Eske Heinez joined the one trillion dollar market Cap Club All about the Memory Plus.
Taiwanese prosecutors suspect three individuals of smuggling in video chips to China through Japan.
And we speak.
Bloomberg Tech is a live from coast to coast with Caroline Hide in New York and V lovel in San Francisco.
This is Bloomberg Tech coming up.
Mike Cron and Eske Heinez joined the one trillion dollar market Cap Club All about the Memory Plus.
Taiwanese prosecutors expect three individuals of smuggling in video chips to China through.
Japan, and we speak with the x price founder and early SpaceX investor Peter Demandis's SpaceX celebrates a major starship milestone.
And meanwhile, we still think about what space is doing to the public markets, but more broadly, the public markets just taking.
A bit of a breather.
Today we're looking at what howpillion terms of semiconductors in particular, it has been a fiery ride hire In the last five training days, we'd added thirteen percent of this benchmark. Today we pulled back a little bit, but there are notable players within the chip sector that you've got to keep your eyes trained on. It's all about the power the muscle of high bandwidth memory, and of course it's
sent Micron storing. We're still holding onto the day's games rock three tens of percent, but it was all about yesterday's market moves and those that happened in Asia as well ed.
That brings us to today's big number one trillion dollars in climbing memoryship giants s k Heinex and Micron have now joined the one trillion dollar market cap club and see continued momentum is investors pile into some of the companies the powering the AI boom. Let's get more with Bloomberg Equities reporter Ryan Vacelica. It's been a number of
days now where we've been trying to look at what's happening. Ryan, with the memory names of particular, Micron was the case study because of that ubs note which you can remind us of, but generally speaking, what is happening across memory, not just in US markets, but of course in Korean markets as well well.
There's been a huge and growing appreciation of how central high bandwidth memory is to the overall AI infrastructure build out. These companies have seen absolutely massive demand and absolutely massive growth. I think Micron's revenue nearly tripled last quarter, which was I think the fastest piece of growth going back to
I think the nineteen nineties. So absolutely just staggering levels demand for these types of chips, all of which are being used in the AI infrastructure, and that is really translating in a pretty direct way to the stock prices. I think Micron is up more than seventy percent in May by itself, which is the biggest one month jump since December nineteen eighty seven. Just to give you a sense of just how quickly these stocks are moving.
Up, let's look at about sk Heinex how much it moved up in the May number as well. They've almost moved in lockstep, both of them up seventy percent in the month of May. Sk high Inex Micron both at more than two hundred percent so far year to date, but what's notable is both of them are posting quarterly revenue increases of.
More than two hundred percent.
So in many ways, Ammas out there seemed to be saying, look, these are still actually fundamentally cheap.
Yeah, the levels of growth that these companies are seeing is absolutely astronomical. And I can also add in sand Disk, Western Digital, Seagate, all the memory in storage space are just seeing absolutely huge demand. The evaluation picture is a little bit tricky though. In fact, we have a story coming out tomorrow, but I'll give you a sneak preview. Micron multiple is trading under ten right now, ten times forward earnings. That is extremely low, especially for company growing
this quickly. However, because memory has historically been quite cyclical, there are some concerns that this low multiple might be an indication of peak earnings. I've actually had someone say to me that he would feel more comfortable if Micron was really expensive right now, because I would indicate maybe a trough level in earnings that are about to go ahead, leading the stock up with it. So right now you can certainly say that Micron is inexpensive. The question is
is AI changing the cyclical nature of memory overall? That we are in some kind of new paradigm where maybe Micron actually is as cheap as it looks, and this isn't some kind of contrarian warning light.
I'll answer that Ryan, Yes, because in high bandwidth memory and the design of the SEC the chiplet, it's directly embedded into the GPU.
And what's so crazy about all of this.
We're talking a lot about supply constraints, a shortage of GPUs, but it's not because of the GPU die. It's not because in video can't get enough of them necessarily. It's the corresponding high bandwidth memories. Not there unpack what UBS said in its note about that, particularly on the valuation. It said, one reason with doubling our price target is we think that people will start to appreciate that.
Yeah, so UBS came out, I think that actually tripled their price target. I think the estmate was that this would eventually be a one point eight trillion dollar company. So even though we've seen some huge gains in the stocks, UBS expects the upside. It sees a lot more additional
upside from here. What it said was that Micron really deserves to have a multiple that's on par with Nvidia, and I think it said it was looking for maybe fifteen times estimated earnings, whereasas historically it gave it a five times estimated earnings multiple. So they are saying that Micron should be worth three times what it used to consider a fair valuation for the stock. That is a really significant change here, and it really is causing people
to reevaluate. How should we be valuating these companies, How should we be considering the nature of the market, the nature of this demand, and what is that mean for growth going forward? Right now it's up in the air, But so far people seem extremely optimistic and extremely positive.
Well broadly, we're seeing at the moment ry not a single cell rating on either of these stocks. Despite that more than two hundred percent crescendo so far this year.
We appreciate you.
As investors are pushing memory chants like Skhinex, like Microntu record valuations. The race for advanced semiconductors also raises new concerns of export controls. So Taiwanese prosecutors suspect three individuals smuggled Nvidia AI chips to China through Japan. So, according to sources, some more we go to Blomboth Senior Tech editor Mike Sheppard to join us. This is once again about super micro computer. This is about servers that are somehow winding their way to China.
Well, that's right.
These three individuals are suspected of having falsified documents indicating that the final destination for these devices was Japan, when in fact, according to authorities, the ultimate buyer was in China.
So place.
Now, authorities managed to seize about fifty of these servers before they were actually shipped, but it looks like at least one shipment got through according to the reporting by our colleagues Mackenzie Hawkins and Debbie Wu, who have been all over this story from the start. Now, Caro, the
twist in this case is Japan. When we have been reporting on and talking about some of the other chip smuggling cases involving in Nvidia's hardware and products, they've typically gone through some of the Southeast Asian economies like Thailand and Singapore, and we haven't seen Japan really factor into this. Japan has actually seen it's more of a place where
Chinese companies can rent computing power legally. It's a practice that falls within the bounds permissible bounds of US export controls through data centers in Japan, Chinese companies can simply access all that AI compute simply by paying a rental fee chev real quick.
Nvidia didn't respond to a requests for comment on this story. But Jensenmong is Taiwan right now, and I believe somebody asked him broadly about this issue very quick.
What did he say, Well, he was asked over the weekend after those three individuals were detained whether super Micro needed to do more to rein in some of the concerns about chip smuggling, and he said, look, they need to perhaps tighten up the ship when it comes to
compliance and oversight of their customers. And this is a rare thing for the Nvidia CEO to say about one of the company's partners, super Micro, and neither company, of course, it's important to note, has been accused of any wrongdoing in these or other cases. But they insisted that their compliance is up up to par and that they are doing more to strengthen it in the wake of the charges that were filed against one of its co founders earlier this year.
An unfolding story one you're all across. We so appreciate it. Bloomberg's Mike Shephard. There, Look, we can continue to talk to your politics and how it impacts the market, and we continue to talk about US tech stocks that are holding onto me and record highs with a whole lot of enthusiasm for the AI trade.
But there's some pressure out of that.
Let's talk about it when that's Tegna CEO se I have left Tegna Investments and just pivoting back to how we started a story that Micron and sk Heinex and now in the trillon dollar club that we're still seeing AI infrastructure bottlenecks. Nancy, Is this just continuing to feed your optimism around the sector.
Yeah, I mean, Carolyn, we're we're ready for a correction because we get one about once a year, and this has been pretty frothy. But when you go back and look when there's been an eight week run like this, historically stocks are up over twelve around twelve percent a year later, and that's Bespoke's work, not mine.
So I think you have to be you have to be nimble.
And you know, we added to Micron at three sixty six just a few months ago and felt late and of course we weren't, at least not for the moment. What you have to do is be diligent about trimming things back, and that is what we've been doing, you know, taking some gains, sitting and waiting, reallocating to names that we think, you know, will continue to benefit. But yeah, this is a productivity driven ball market.
I think it will continue for some time.
We will probably get a correction that will be yet another opportunity like deep Seek was, like the first quarter of this year was. But you want to stick with the high quality names and you want it. You don't want to chase the latest because the hedge funds are going to pivot here pretty quickly, and we saw that from hardware to software and now back to hardware.
Nancy, I'm going to show you a chart, and for those that listen to Bloomberg Tech as a podcast, it's a squiggly line that shows a very sharp up with trajectory. From the end of March to present day. We've gone from about twenty two five hundred points to almost thirty thousand. Yes, if you tell anyone I quoted points on the I'll deny it. You just said that you're ready for a correction in the Micron story was about ubs almost tripling its price target.
How are you going to play Micron in a correction environment?
Yeah, I mean we are going to be trimming it here shortly and we will likely add back to it.
Beware of people that tell you, you know, old tech is dead or software is dead, because these companies find ways. I understand this is more of a demand issue, but they do find ways to pivot, and so I think what you want to be doing is taking some gains when you can and then looking for opportunities when the market pulls back. We called for a bottom on April fourth. I mean, that's not really our business, but we did. We were a little bit late, but that is how
this market has been moving. And so it's quickly and it's rapid, and it's violent to the upside.
And so don't.
Confuse that with you know, being a genius, because what the market giveth, it will also take it the way.
So you just want to remain nimble.
Remaining nimble and thinking about what your long term bet saw Like Nancy, you always bring us our twelve best ideas portfolio, and within that is some software is service now for example, and boy of those names been beaten up, but they have clawed their way back from some of the bottoms.
But tonight we get we get it called salesforce numbers.
Will that fundamentally show that these AI related disruptible names can hang on in there?
I think so, Caroline. It's an important report to be sure.
And you know what we've done is we've within our twelve best We have a six for twenty six that's up forty percent this year and includes names like CrowdStrike. Last year we had service Now in that five for twenty five and it did abysmally, but the portfolio is still outperformed. So that is what you have to think about, is that the right name is Microsoft going to be a winner?
I think so.
I'm more concerned about Salesforce, which we don't own.
We exited Adobe for obvious reasons a while back, and I think that's what you have to consider who will win. And I do think if you own the platform, you'll be a beneficiary of AI. And I do think service now is well positioned, despite how difficult of a name it has been to own s.
We're in this period now where we've gone through earning season, including through in video, So we're not now fixed on the Kapex figures. Maybe we still are in videos set its piece what happens in this interim period, like what is the catalyst in either direction for the market.
That's the right question ed. I think that is why we think we may be right for a correction.
What will happen is we will turn our attention the market's fickle, and then we'll start focusing on the Fed. We'll go back to is Horn moves open and there will be hand ringing, and I think that will drive the narrative because it's too early for the midterms. So we are putting in place some protection for our clients and it may be money not well spent if the market continues to accelerate, but it never does go straight up, So I think.
You want to pay attention to FED speak.
I hope there's less of it as we move forward from here. I mean that is a promise that Kevin worsh made, and let the FED do their job and then and there will also be on inflation watch, which which I think is actually not the bigger problem that we face from an economic standpoint.
But there are many who disagree.
With me on that what about it wrong.
Yeah, I mean I think I think that's problematic on so many levels. But you know, we're getting all the peace talk, but we haven't seen a lot of progress. So I think for most investors the question is when does when do the straits of straight of horror moves open.
When that happens, then I think you'll see a melt up.
But until then you'll see a lot of hand ringing and that you know, that is what markets do, and that is what the algos and the hedge funds exacts in the near term. So volatility, remember, friend of long term investor, use it to your advantage.
Because earnings growth has been amazing. We've also seen guidance raised.
We've been steadfast in the tech trade for the last three years while many wrote it off. So we're pretty happy with you know, the way our portfolios have performed as a result of that.
Nancy Tegler of Laffetango Investments back on v tech, Thank you very much. Now coming up, we're going to speak with the X Prize founder Peter Demandez as SpaceX celebrates a major starship milestone.
This is blombog Tech.
SpaceX's latest successful starship launch marks a major milestone for the company's next generation rocket program and a critical step towards the longtime growth strategy that was laid out in its recent IPO filing. Also crucial elon Musk, the S one made it absolutely clear with an outsized pay package, majority voting control for the CEO bloemeg intelligence writing. The document highlights some significant governance concerns that still might be
overshadowed by investor's fear of missing out. FOMO on the biggest IPO ever here to talk about it. The importance of Starship Musk the perspective of an early SpaceX investor. Peter diamandis founder of X Prize, and we're trying again. You're on the show Friday, and we had to cut our conversation short. But then on Friday night, Starship twelve Flight Test Mission V three. Your reaction to it, to how that went.
Well, Listen the fact that it was a brand new vehicle, brand new engines, the most advanced engineering ever built by humans. It was incredible success. You know what people, if do you realize, is that you can't test a rocket a little bit at a time. You have to test the entire system and the fact that the launch took off with its version three of its engines and the entire vehicle the largest, most powerful engineering system ever launched by humans.
It was great.
We're going to see Starship making incremental flights, getting better and better until the time where the entire vehicle is refliable, refuelable on orbit and becomes a platform to go to the Moon, go to Mars, go beyond.
Why Starship is also its capability and payload to orbit for first starlink and then orbital data center. I wrote at the beginning of the week about how clear the S one was about the limitation of SpaceX's current access to compute and how much it's going to need in the future. You then had an exchange with Elon Musk I think in the last twenty four hours, right or forty eight hours about that exact issue. What was the point you were trying to relate to Elon there?
The point is that he's not just built This is not just a rocket company, right, This is a vertically integrated satellite network, global broadband, sovereign communications, AI, compute and ultimately off planet infrastructure. You know, he is building a hyperscaler Uh, not just an AI system, So his ability to succeed as a company goes beyond just the you know, groc is an AI system. He's providing the infrastructure for
a large number of the frontier labs out there. You know, And what people need to realize about this company is it's the It isn't just a little bit ahead of
the entire launch industry on planet Earth. It's orders of magnitude ahead beyond anything else, and everything we hold a value on Earth, metals, minerals, energy, real estate is in near infinite quantities in space, and so what we're buying is the next you know, the global economy, you know, two dot oh and three dot O. As we look at into SpaceX.
Go global, Peter, because I know you can, and we're please that you do. There's a lot of hand ringing and almost people feeling that every turn, the US uses China as the excuse as to why we need less regulation, why we need to go full throttle, why we need to win quote unquote how much you seeing a space race US China and how much a SpaceX managed to dominate their visa v not just US competition, you.
Know, we humans love competition and it drives us forward. And so if you think about this, you know, it is the US versus the Soviets and the Russians in the you know, fifty years ago, we're doing that same thing again, but this time it's beyond just a political race. It's an economic race.
Again.
We're building out huge revenue engines as we're going forward, whether it's going to be mining the Moon for resources or mining the asteroids for resources. We're also going to be building out a level of global compute infrastructure in Earth orbit. And again, no one was talking about this a year ago, and today every major hyperscaler is we see anthropic buying into XAIS or space XAIS data centers on Earth. But they're also going to need that's going to be given for morbit.
Look, you're a man who's driven by trying to solve the world's most pressing problems. You're also a man who happened to have got into SpaceX and Google and other key companies very early.
Well at this moment where people.
Are questioning how it helps them, how is this being democratized in some way? How are we solving the world's greatest problems for everyone, not just the few? Peter, How are you thinking about that, particularly when we look at a listing of a company that is already so valuable.
Is that much left on the table in terms.
Of democratization and buying of shares by retail investors?
Sure, so let's hit the point you made first about solving the world's biggest problems. The single most powerful thing we humans have is our intelligence. We're about to see human intelligence increased by not just ten or one hundredfold, a millionfold, a billionfold, right, It's the ability for us to discover breakthroughs across physics and chemistry and biology, material sciences.
All of these things are going to be driven by the use of these AI systems, And so I don't think people understand how rapidly the economy is going to grow on the back of AI. We're going to be solving aging as a fundamental right. What is it worth if you can add thirty healthy years and then those extra thirty healthy years on your life by you the next fifty healthy years or one hundred healthy years, what
is that worth? Individuals? When we get to room superconducting, when we get to new ways of growing food, you know, twice as fast, five times faster, healthier, we're heading towards a world of abundance across everything, food, water, energy, healthcare, education, all of these elements. You know, when I interviewed Elon on my Moonshots podcast at the beginning of this year and then again in March, what he talked about, and I believe this is we're going to see double and
then triple digit GDP growth. And this is happening not because we're working harder or we humans are smarter. It's on the back of AI and humanoid robotics. One of the things he said was that we're entering a world where AI and robots are going to create so much, so much products, so much availability that we could not want enough. Now, this sounds like, you know, a techno utopian vision, but we have to realize that all of the progress that we've seen in humanity, we're living lives
today that are godlike. Compare it to our parents and grandparents.
Yes, let me jump into you just said thank you for vote January six for Elon at length and then again in March, and what you just said about robots. There is broad speculation right now about the prospect of a post IPO SpaceX merging with Tesla on January thirtieth.
January thirty, if I.
Reported they'd held talks prior to the Xai transaction, you say one hundred percent. I mean, what are you learning from those two conversations? What do you know about how real?
That is? Why it makes sense.
It makes sense because it consolidates Elon's control. Today as reporting the IPO in the S one, you know, there's super voting rights that he has and of course is inside owners. I think it's like, you know, eighty plus percent voting control. He doesn't have that in Tessla, and
by combining the companies, I think he'll have that. He'll have the ability to operate across all of this infrastructure and you know the a fleet of cybercabs, all of it Tesla vehicles out there that have compute capability on them, have power on them as well. We're creating a global infrastructure on the ground, in space. So it just consolidates control. It makes his ability to implement his vision a lot more efficient. So you know, I put it as not a matter of if and only a matter of when
those two companies come together. And also because there are valued public companies, you can make that merger happen a lot easier than if you're combining private companies and arguing about valuation.
Peter jmnis X Prize founder, fascinating to have some time with you and the vision of where spaces goes and long win Tesla.
We appreciate it.
Now coming up, we're going to be joined by the CEO of Cognition and discussing the AI startup's latest fundraise to power the first AI software engineer as they call it, Devon or many devans. There's a cracking, big valuation on the back of this. I'm excited about that conversation coming up next with San Francisco and New York and with stocks under pressure in the broader tech ecosystem.
This is blombg Tech.
Welcome back to Bloomberg Tech, and we pull back from some of our record highs today we turn our attention to what's happening with the Middle East. Will there be some sort of piece still between the US and Iran? There are conflicting we'll move music around it in newsflow, and the market just sells off a little bit in that.
I have a storm.
We're currently off by four ten percent on the nastat one hundred coming off of yesterday's record high. We're looking at the semiconductors, particularly under pressure video and the like, yanking it down too and a half percent, but remember it's up thirteen percent prior to this on a five.
Day winning streak. So we take a but we don't take a pause.
One key name, Micron, is still managing to cling into the green RUPs seven tenths of a percent. This is as they hit a trillion dollar figure.
Ed.
This is as we start to see the real focus on high bandwidth memory that saw also sk Heinex run up so far.
Let's get back to today's big number, one trillion dollars in climbing. That's the market cap memory chip giants sk Heenez and Micron have reached and his carriages outline.
Micron still has momentum.
Investors are piling into companies that are powering the AI boom. Let's get more Micron and sk Heiniez Boombozi and King, who leads our coverage of semiconductors.
You are so funny.
You and I go to all these conferences and speak to all these people in industry, and if you do it from their perspective, none of this is a surprise, like as Jensen one would put it at c Ovin VideA, he was trying to convince the memory makers years ago that this would happen. Now Here we are why do we need so much high bandwidth memory?
Yeah, I mean, if you look at the forward estimates for this company, Wall Street is bought in to this massive sense of we need so much more equipment. There's going to be trillions of dollars of spending happening on this, and memory is an absolutely fundamental base layer of the technology. If you believe that the industry and the economy is changing to the extent that people like Jensen we are saying it will.
And it seems like there's stereotypical oligopoly here. There's like three key players that are rushing to try and for the hole in high bandwidth memory.
Of course Dram sort of goes to the backseat a little bit here.
But are we seeing any competitive threats because at the moment, analysts all the community think keep buying these stocks. The bottlenecks are going to last through twenty twenty seven.
Yeah, I mean. The other that's a good point, Caroline, And the other way to look at that is why there are only three companies left. The answer is because this has been a horrible market for years. Right. We've had some good years, but we've had a lot of bad years as well. There's only three real providers because it's so expensive, the bets are so big, and it's so difficult to make us sustain living in this business.
There survivors, not necessarily thrivers, even though that's what's going on right now.
Could you just educate the Bloombert audience a bit on the history of memory. The idea was foom and bust where memory principally went before there was this great demand from the data center space.
Yeah, I mean, the key point is that this is a commodity. One chip from one company can be swapped out for another, so you effectively have a market, right, Prices go.
Up and down, supply and demand prices exactly.
It's a commodity, right. And in the past, you know, the big market was PCs and then it was smartphones, and of course these are consumer devices by in lige, so long term supply versus short term fluctuations in demand was a recipe for disaster.
In Boxie and King, Thank you very much. Carry some more news.
Yeah, it's time now for talking tech ed and first up and FTE. Dance is planning just sharply increased capital spending to lead the Chinese AI market.
Now.
The company is considering a seventy billion dollars.
This year to build out data centers, another AI infrastructure, and it may boost capex roughly one hundred billion dollars next year. As an update on the Samsung strike talks, the company's union members rooted in favor of a compensation deal that will hand Chip workers an average bonus about three hundred and forty thousand dollars now.
The agreement avoids a strike that had threatened.
To disrupt global chip supply, and similarly, TSMC chief cc Way told staff that they'll see more than a thirty percent jump and profit sharing payouts this year on average, according to a source. Whays comments come after some employees voiced concerns over their incentive plans online and followed the Samsung union deal.
There Okay, back to private markets, Cognition has raised over a billion dollars at a twenty six billion dollar valuation. Lux Capital, General, Catalyst and AVC led the new round. Cognition CEO Scott, who was here to discuss the news. We're also give an update on the company's AI codey agent, Devin, not just one agent, like I think Caroline made a good point, like it's an army of agents, and we'll get to that. Start with some basic Scott, A billion dollars, big valuation.
Why'd you do that?
Yeah, yeah, absolutely well, thank thank you so much for having me back. A few different reasons. First of all, you know, the growth that we've seen in the business has been incredible, and I think really across the board, what we're seeing is that AI.
Is doing real work at real companies everywhere.
And you know, every company in twenty twenty six is a software company, and so we work with, for example, the top five health insurance in the United States. We're seeing folks building way more tooling for all of their care providers. They're able to cut down on price and cover more people. We're working with banks on delivering software and making sure to give their customers access to what
they need right. We're seeing this with the Treasury and NASA as well, and so a couple of reasons for the raise.
I think, first of all, we.
Want to continue to grow aggressively and this really allows us to do that. It allows us to scale our compute, it allows us to grow the team and so on. Second of all, it allows us to stay independent and to really continue as an independent business, which is really really important for us.
It's going well as an independent business.
Like, what's interesting talking to you over let's say an aggregate over a period of year is to trap growth.
Yeah.
Right, So when you first start coming on the show, like beginning of twenty five to twenty four, the revenue.
Run rate was a few million with respect.
Then exactly a year ago, you're kind of in a run rate of about thirty seven million dollars. Where's your revenue run rate now?
Yeah, so we're getting close to five hundred million today. As you said, we've only been in business for about two years. And I think a lot of what it speaks to is just how much demand there is out there for all of this. And you know, there's about thirty thirty five million software engineers in the world today. We want to make all of them ten times more efficient, and then we think there is a lot more than ten times more software to build.
There is so much demand, Scott, but there's also a pretty crowded market when it thinks of startups. And admittedly you've been talking about how labs are sort of buying these startups, and we think about what deal Curse has just been doing over with SpaceX.
But I'm interested as to how you see.
The threat from the big labs in and of themselves.
Yeah, yeah, absolutely. For us, it's actually the other way around.
I think for us it's you know, being fully independent and fully neutral is actually the best way for us to be aligned with our customers. And so we work very closely with all these labs, Opening Eye Andthropic, but also Google, Xai and so on.
We have deep relationships with them.
We work with them on research, and it allows us to work with our customers and provide them the best model for every different use case. And so Devin is a compound system that works with all of these different models, and because of that, we're able to be the Switzerland in the equation here.
Switzerland that sometimes makes the most of the disruption when it comes to talent and the like. I think about what happened last year the Windsurf assets, the IP, the brand, the employees after Google took well, the key CEO and leadership of that company will more in a M and A happen. Is that where some of the new funding will come into perspective.
I'm sure there will be more, and you know there will be different things that come up. It's for us, you know, what we're personally most focused on is just continue to grow the business and.
As best as we can.
You said, well, a heavy emphasis on independence and you cooled yourselves the Switzerland of this space. Yes, you know, what do you think will happen if SpaceX does it acquire Cursor? Is the base question. But when I speak to say the engineering teams at Nvidia, the reason Laydight Cursor was the freedom to swap in and out the underlying model depending on what your your coding objective was.
I suppose that's one reason why they like Devin Right, But if Cursor becomes a part of space XAI, you know, how do you see that changing the field data is so critically important?
Right? And who you are beholden to? Yeah?
Absolutely, No, I mean it's a great question. I think in practice, I think there are a lot of great teams working on could, including of course the labs themselves. I think what we see is that the ecosystem is broad enough and vibrant enough that it makes sense for there to be different players in different positions, and so there are going to continue to be first party products
from the labs themselves. But as we're kind of saying, you know, to your point, I think having the ability to serve each of the different models, and not just the ability, but also the new vitality in terms of being incentivized to just serve whatever is the best model for each case rather than a single you know, a single series of models, I think is an important position in the.
Space as well.
You would say that Devin is the first AI software engineer, and we talked about the revenue run rate. Clearly that's evidence of momentum and success. But are you able to sort of give any data on how pervasive DEVN is, how it has changed the structure of different engineering orgs at software companies, at other technology companies for sure.
Yeah, I mean we're seeing this across the board. Where as teams are adopting devon and coding agents on mass that in practice they're able to.
Do much more and execute much more aggressively on road mass.
This is a new code or is it going back over old code bases.
It's both and so as you can imagine, the large majority of work is working on these existing codebases and continue to build and add new features and so on. But even internally at Cognition, for example, more than ninety percent of the code that we write is written by and so of course we're using devon all day when we're going and building devon itself.
And even with that productivity, you've been scaling the amount of people you hire.
Scott, just reflect on what.
This means in terms of how many software engineers we're going to need. Like this ongoing anxiety is the disruption that AI causes in all its ways and across all these industries.
Yeah, No, I think what we'll see is, of course the job will evolve over time, and we'll see some of the skill sets change, but I think we will have far more people doing this in building software and building products, not less. And you know my favorite sat on this personally is today there's about thirty or thirty
five million software engineers in the world. Just twenty twenty five years ago, at the start of the century, it was under one million, and that's clearly come with a massive rise in the amount of software that we've produced, and I think as we continue to make it more and more efficient, we're actually just going to produce even more software, not less.
Scope very very quick.
What's the goal you've set the team for the balance of the year one metric?
Yeah, yeah, Look, I think for this year we firmly intended crass a billion in revenue run rate. We want to keep going for anther even beyond that, and from there we just want to grow and share many of the companies in the world that we can.
Cognition CEO Scott Wu, thank you very much for joining us today. Now coming up, Salesforce Snowflake, both reporting earnings after the closing bell, would discuss what to expect how's AI disrupting their business models.
This is Bloomberg Tech.
We're watching shares of Salesforce up one point two percent, set to report quarterly results this afternoon. Wall Street's watching closely for science that the company's AI offerings can help reignite revenue growth. Bloomberg's Brady Ford, who covers Salesforce, is with us, and twenty four hours ago we were talking about the idea that right now the AI agent story
is more marketing than real in terms of revenues. That will be the test this evening, right is that kind of what you're looking for.
It's all these as apocalypse spheres, right, That's what everyone's been worried about for the last couple of quarters, and tonight the only real way Salesforce can beat pack all that skepticism, all the pessimism is showing that revenue is accelerating and it's coming from these new AI offerings. We haven't seen that yet. That's what folks are hoping for in the back half of this year.
We've had that Agent Force was sort of offering.
One an eight one hundred million dollar revenue stream thus far. But we're also seeing not bad revenue growth from Salesforce, or at least predicted, but a lot of that's coming from Informatica, right, from inorganic growth.
Yeah, that's been the classic debate with Salesforce for so long, which is you know how much revenue growth is from those organic products versus some of the acquisitions that's come back with Informatica.
And yeah, eight.
Hundred million a year on Agent Force nothing to scoff at. But at the end of the day, right now, if you're an application company, and your revenue is decelerating.
Market is really going to punish you.
Ryan Blastellica Inequities team who you know, we're all very close buddies, where they help us out a lot with their their sort of stock coverage. They frame this is actually, if this goes well, it could.
Change the story for the stock. What has the story been.
I mean, his salesforce been one of the I want to say victims, but those under pressure from the SaaS apocalypse kind of narrative.
I mean, it could change the narrative for the whole sector, right, because salesforce is these SaaS companies. So when folks think about the SaaS industry getting hurt by AI getting displaced, you know, the idea that innovation and technology is no longer happening at these SaaS companies like you know, an Adobe or a salesforce. If they can show that that's changing, that could be very meaningful.
But still, they hit a three year low on their stock last month. They haven't recovered much and they're down thirty percent year to date. So is there a narrative what anecdotal evidence to fight back as some of what you wrote Brody that it isn't actually working at agent thoughts with in compliance offices, yet they can't sign off on it.
For example, the big story I think is that the technology is real, but it takes a long time to implement at big corporations, right, I mean you or I can go and chat GPT and do some you know, whatever workflow you might do, but as a corporation, that's very difficult to implement it widely.
In many ways, that compliance stickiness, though, makes makes salesforce pretty sticky.
BLO makes pretty good. We appreciate you.
It's gonna be a busy evening after the bell. There's also other stocks reporting, for example Marvel and fascinating chip stock. Like this is in many ways about the focus on photonics, on optics, on networking, but it's also think about that two billion dollar investment in video made in but March and Marvel Technology, how.
Have they been performing within videos backing?
Yeah, it's interesting. The stock's down so much, you know, ahead of the earnings print. I don't see anything on the Bloomberg.
This year.
Yeah, exactly. It's a high flight.
So in our conversations about custom silicon, we super focus on broad com principally Marvel has XPU exactly the same idea. Rather than selling a chip, which is it, it partners with a hyperscala or a technology company and says we will do this with you custom silicon. There's also some examples that that might extend outside of the world the data centers.
We can talk about that at later date.
But yeah, if this is happening and we're in a compute deficit, Marvel's a likely winner, and.
We'll see whether they can live up to some of the expectations around their numbers, and there are some lofty ones out there. But all of this at the moment is around AI's obvious rapid growth, but it is actually pushing global energy demand eb A skyward two, forcing electrical grids, for example, to expand and modernized. Now companies from China to Nigeria investing in new tech to power the future. That's a focus of doing their primer this week.
Take a listen.
For a long time, rich countries haven't had to think about their grids all that much. Their electricity demand has been pretty much flat since the two thousands.
The times are changing.
Explosive growth of AI, the rapid build out of data centers.
They're consuming enormous amounts of energy.
A few have four companies now that are intending to spend over three hundred billion dollars this year.
With industries like AI and EV's growing fast, the world is predicted to use twice as much electricity by twenty fifty. That's roughly a whole new USA's worth of electricity every five years. To make all that power and get it to where it needs to go, the world's grids need to evolve. All that new infrastructure will cost billions of dollars, but so did broadband internet and that's ended up creating trillions of dollars of value. And some countries' grids are evolving faster than others.
In China, power generation has gone up seven times since two thousand.
The battle to build the best grid is a battle to win the future.
Here more from the team at Bloomberg Originals on today's episode of Primer that's tonight on Bloomberg at six pm Eastern and on Bloomberg Originals at apm Okay. Coming up, we hear from UBS Asia Pacific President ikbau Khan on how he sees AI impacting jobs.
This is Bloomberg Tech.
UBS Asia Pacific President ikbaw Khan says AI will free up capacity and improve productivity, but also have an impact on jobs. He spoke exclusively to Bloomberg Stephen Engel on the sidelines of the firm's Asian Investment conference in Hong Kong.
I think the opportunity that we're seeing now with technology and with AI is very much around simplifying, speeding up the processes, not cutting corners, but actually fundamentally improving the.
Process consistency of process.
Just think about documenting source of wealth of an individual.
It's a pretty complex task.
But if AI can help you contextualize that and help you actually do that and ensure that there's consistency, you're going to fundamentally improve the process in addition to what you're doing today.
Nice segue, AI. It's in everyone's discussion book right now. Obviously, we just had Jamie Diamond of JP Morgan talking about that. We've heard some comments that he got a little bit of blowback. Bill Winters at Standard Chartered saying it's going to have a significant impact obviously on maybe some of the rank and files in the banking industry. How do you see AI over the next couple of years changing the way you hire and who you hire.
So let's step back right.
I mean, clearly there's a lot of focus on AI. We see high valuations around AI. There's a lot of talk around the value chain, everything from actual lms to data centers to respective foundations infrastructure. For US, at UBS, we've been very focused on AI specifically, also led and
driven by Sanjo multiur Group CEO. We've implemented, for example, copilot across the board as just one example, and I have to tell you, in the last six months, I've been using it more myself and I've been using it personally as well as professionally and has actually made me more efficient, more effective, and I think over time everybody will become an AI native.
It comes down to adoption and application.
Fundamentally, we look at this as something that will really enhance and increase capacity. What does that mean as we become more productive, we can use that capacity to grow, we can use that capacity to serve our clients better. Imagine at UBS, when you come in through the door as a client, you get onboarded and if you're eligible to getting.
A solution or a service.
From a compliance and regulatory perspective, it comes down to is that valuable to you or not?
And we will serve you.
Now, all of that process is curated, semi automated, manual, people driven. You can aifi that if that's actually a word to day right, and it is. You do that, you can create a lot of capacity and that capacity can be used to serve clients even better and grow.
What does it mean about top line job growth? Do you cut back to get more efficient? How does it work and how do you communicate that?
Look at the end of the day, as I said, I think it's more about productivity capacity. Now, if we can use that capacity to serve our clients better, gain more share of wallet, grow faster, grow more, then the impact on costs and jobs is going to be less. Now, if we cannot, and this is an industry wide topic, then of course it will have ramifications and implications on costs and jobs.
Steven Angel that with the UBS focus. But we now turn our attention to the White House. President Trump is now speaking as a lot of his cabinet meetings it's just take a less
Minages and they've really been that way for a year