# AI Spending, Mobileye's CEO Exit & IBM CEO on Results

An episode of [Bloomberg Tech](https://invisiblenetwork.ai/podcast/1611767434), 23 Jul 2026, 47 min.

Audio: https://podtrac.com/pts/redirect.mp3/traffic.omny.fm/d/clips/e73c998e-6e60-432f-8610-ae210140c5b1/41764a4f-fc64-4e11-89ba-ae7c0030ab5e/4b643b39-2166-41f4-a6cc-b4900139f9c8/audio.mp3?utm_source=Podcast&in_playlist=9caafc41-289c-4115-995d-ae7c0030ab75

Bloomberg’s Ed Ludlow breaks down Alphabet and Tesla's earnings as the Google parent company boosts its investment plans, while Tesla burns through cash to fund Elon Musk's AI ambitions. Plus, Mobileye founder and CEO Amnon Shashua will be stepping down after 27 years. And we speak with IBM CEO Arvind Krishna after a steep drop in mainframe sales weighed on the company's results. See omnystudio.com/listener for privacy information.

## Transcript

[0:02] Speaker 1: Bloomberg Audio Studios, Podcasts, radio news. Bloomberg Tech is live
[0:15] Speaker 1: from the heart of Silicon Valley with ed La Throw
[0:19] Speaker 1: in San Francisco.
[0:22] Speaker 2: This is Bloomberg Tech coming up. The AI spending races
[0:25] Speaker 2: ramping up, with Alphabet boosting its investment plans, while Tesla's
[0:29] Speaker 2: burning through cash to fund Elon Musk's AI ambitions. Plus
[0:33] Speaker 2: mobil I founder and CEO and non Shashua will be
[0:36] Speaker 2: setting down after twenty seven years. He joins us for
[0:39] Speaker 2: an exclusive interview to explain why, and we speak with
[0:42] Speaker 2: IBM CEO Arvin Krishna.
[0:44] Speaker 3: After a sleep drop.
[0:45] Speaker 2: In mainframe sales wide on the company's results. This is
[0:50] Speaker 2: a technology earning story about AI spending, how much, how fast,
[0:55] Speaker 2: with what returns. We're looking at Alphabet, the parent of Google.
[0:58] Speaker 2: We're looking at Tesla in Alphabet place on track for
[1:01] Speaker 2: his biggest drop at one point since May of last year.
[1:04] Speaker 2: Tesla a very steep decline now of fourteen percent, biggest
[1:08] Speaker 2: drop in a long long time. Tesla had a strong
[1:11] Speaker 2: quarter for EV deliveries, but profits still tumbled. Shares, as
[1:15] Speaker 2: I said, down fourteen percent. This is partially due to
[1:19] Speaker 2: ambitious spending. Was Tesla's first cash burn or negative free cash.
[1:23] Speaker 3: Flow in two years.
[1:24] Speaker 2: Here's what CEO Elon Musk had to say on the
[1:27] Speaker 2: company's capex plans.
[1:29] Speaker 4: We should be spending on capex as fast as we can,
[1:32] Speaker 4: spend as fast as we can without without it being
[1:36] Speaker 4: too wasteful.
[1:37] Speaker 5: So I'm not trying to aim for like some.
[1:41] Speaker 4: Extremely high efficient efficiency capital spend because that would slow
[1:46] Speaker 4: things down. So it's trying it's a balance between how
[1:51] Speaker 4: much so capital efficiency versus time.
[1:56] Speaker 2: Joining us now, Ivan find Sef Tigres Financial Partners, CEO
[1:59] Speaker 2: and partner. It's been a long time since you've been
[2:02] Speaker 2: with us on Bloomberg Tech. Welcome back. The stocks down
[2:05] Speaker 2: a lot, biggest drop since June of last year. We
[2:09] Speaker 2: went into this with Wall Street saying we want to
[2:11] Speaker 2: see Tesla spend. They are why the negative reaction, Well,
[2:17] Speaker 2: they like.
[2:17] Speaker 6: To see the companies. Wall Street likes to see companies spend,
[2:21] Speaker 6: but they don't like companies that spend. And look, the
[2:24] Speaker 6: Tesla has never been a car story. It's always been
[2:27] Speaker 6: a technology company story and AI story, and uh, you know,
[2:32] Speaker 6: Wall Street has to continue to go back and forth
[2:34] Speaker 6: with evaluating that and the whole you know, the the
[2:37] Speaker 6: upcoming drivers for the company are its autonomous technology. It's Robotaxi, uh,
[2:45] Speaker 6: the Optimist robots and evolving from to a physical AI
[2:51] Speaker 6: company and how you evaluate that. And it also takes
[2:54] Speaker 6: a lot of capital and it probably is going to
[2:57] Speaker 6: take And like Elon just said, they want to spend
[3:01] Speaker 6: as much as they can smartly.
[3:02] Speaker 5: Yes, but they are also.
[3:05] Speaker 6: Competing against a lot of other competitors. And this company
[3:11] Speaker 6: is viewed as a tech company, and the profitability from
[3:15] Speaker 6: that is still a ways off.
[3:18] Speaker 2: Again, I'm going to acknowledge the stop biggest drop since
[3:21] Speaker 2: June of last year, down fourteen percent, but at its
[3:23] Speaker 2: lowest level since August of twenty twenty five. Let me
[3:27] Speaker 2: just go through the sort of non financials, right. Cybercab
[3:30] Speaker 2: production has begun, They've been doing employee rides in those
[3:34] Speaker 2: cyber cabs.
[3:35] Speaker 3: They've expanded the.
[3:37] Speaker 2: Robotaxi areas to new cities, but also expansion in Austin,
[3:42] Speaker 2: and the Optimus humanoid robot lines have gone in in Fremont,
[3:45] Speaker 2: production expected to start later this year. That doesn't seem
[3:49] Speaker 2: like enough of an update on those business lines to
[3:51] Speaker 2: convince investors.
[3:54] Speaker 6: Well, everybody wants to see when the optimist robots will
[3:58] Speaker 6: be available, what would be the functionality, how will they
[4:01] Speaker 6: be deployed? So that's still a weighs out. So it's
[4:03] Speaker 6: still as shown me company. And right now the market
[4:07] Speaker 6: is going through a difficult time. I mean, it's held
[4:10] Speaker 6: up phenomenally well with what's going on in the world,
[4:13] Speaker 6: but and it's been driven by tech. The market strength
[4:17] Speaker 6: for the past several years. And you know, this year
[4:19] Speaker 6: so far has been driven by tech and that price
[4:22] Speaker 6: for perfection. And when you see, you know, a slight
[4:25] Speaker 6: pause of concern, the stocks are going to get hit hard.
[4:29] Speaker 6: I mean, Tesla has always been a volatile stock and
[4:32] Speaker 6: if you look back in its history, buying the dips
[4:36] Speaker 6: have paid off. So I feel that this sell off
[4:38] Speaker 6: today is a buying opportunity. It's shown that to be
[4:41] Speaker 6: the case in the past, and I still believe that
[4:44] Speaker 6: to be the case now. And the drivers of its
[4:47] Speaker 6: future growth are still ahead of it. And the concern
[4:53] Speaker 6: is that we're not seeing currently or real tangible results
[4:57] Speaker 6: or not on the near term horizon.
[5:01] Speaker 2: Record ev deliveries in the quarter gone, but lower asps,
[5:05] Speaker 2: higher interest rates, rising commodity prices, stock based compensation, and
[5:09] Speaker 2: then spending AI spending. And what I find so interesting
[5:12] Speaker 2: right is capex is still twenty five billion dollars for
[5:15] Speaker 2: this year, which is you know, for Tesla, like it's
[5:18] Speaker 2: just unprecedented, But they're not even tracking to that right now.
[5:22] Speaker 2: They're going to have to really accelerate in the second
[5:24] Speaker 2: half of this year going forward, Ivan, what's the story
[5:27] Speaker 2: going to be. What is the metric that you'll track
[5:30] Speaker 2: as evidence that Elon Musk and the Tesla team are
[5:32] Speaker 2: making progress on those future business lines?
[5:35] Speaker 6: Well, our number one measure of performance is economic profit.
[5:40] Speaker 6: It's an increasing return on capital and these will eventually
[5:44] Speaker 6: be high margin businesses with higher returns, but it takes
[5:48] Speaker 6: a lot of capital invested now, so that's diluting the
[5:51] Speaker 6: near term returns and that's the primary issue. I'm also
[5:55] Speaker 6: impressed that they did sell the number of cars that
[5:58] Speaker 6: they did. The demand for their EV's remained strong because
[6:02] Speaker 6: the overall EV demand in the industry has been solved
[6:07] Speaker 6: for some time, and Tesla has sold a lot of
[6:10] Speaker 6: cars since inception, and the demand for the cars remained strong.
[6:14] Speaker 6: But the future of the company is physical AI, it's robotics,
[6:20] Speaker 6: and it's also the cybercab and Robotaxi, and those are
[6:29] Speaker 6: still a little ways out. So anytime that things are
[6:33] Speaker 6: not perfect, the stock sells off.
[6:36] Speaker 2: I find sepe Tirest Financial partners back on the show,
[6:39] Speaker 2: thank you very much. Indeed, another piece of news from Tesla,
[6:42] Speaker 2: and at a time when memory chips are hard to
[6:44] Speaker 2: come by and prices are high, Micron's carving out capacity
[6:48] Speaker 2: for Tesla. During the company's owning school, Elon Musk praise
[6:51] Speaker 2: Micron for setting aside what he called a significant allocation
[6:55] Speaker 2: of memory chips, and he also called them under reasonable terms.
[6:59] Speaker 2: It's interesting to see Micron up almost three percent on
[7:02] Speaker 2: a day where a lot of chip stocks are down
[7:04] Speaker 2: generally speaking, and a lot of tech stocks are lower.
[7:07] Speaker 2: Let's turn to the other big, big story. Alphabet, the
[7:09] Speaker 2: Google parent, raised the top end of its CAPEX plan
[7:12] Speaker 2: for this year to two hundred and five billion dollars.
[7:16] Speaker 2: There's cloud growth, there's Gemini engagement. All there, but the
[7:19] Speaker 2: discipline on spending is a bit of a concern. Joining
[7:22] Speaker 2: us is Eric Sheridan Goldman Sachs Co Business Unit leader
[7:25] Speaker 2: of the Technology, Media and Telecommunications Group in Global investment Research.
[7:30] Speaker 2: He says Alphabet's well positioned to benefit from the growing
[7:34] Speaker 2: demand for AI across both consumer and enterprise markets and
[7:37] Speaker 2: reiterates a by rating will lowering is twelve twelve month
[7:41] Speaker 2: price target to four hundred and thirty five dollars from
[7:45] Speaker 2: four hundred and forty Eric, welcome to the program.
[7:49] Speaker 3: Not a surprise really.
[7:50] Speaker 2: That they would raise the capex expectation for this year,
[7:54] Speaker 2: but the reaction to that seems a bit severe. They
[7:58] Speaker 2: did swing to negative free cash flow for the first
[8:01] Speaker 2: time as a public company.
[8:04] Speaker 7: Was that it They did swing to negative free cash flow,
[8:07] Speaker 7: And I think there's a mixture of signals versus noise
[8:10] Speaker 7: in this print. The long term signals are Search is
[8:13] Speaker 7: to stable business, YouTube continues to gain momentum across the
[8:17] Speaker 7: broader media landscape, and Google Cloud revenue continues to re
[8:21] Speaker 7: accelerate and will likely reaccelerate in an outsized way for
[8:25] Speaker 7: most of the next one to two years. They made
[8:28] Speaker 7: some decisions short term to raise cap X and strike
[8:32] Speaker 7: deals for third party compute that are impacting op X
[8:36] Speaker 7: that are all about closing some of the demand versus
[8:39] Speaker 7: supply gap that exists around compute today because they didn't
[8:43] Speaker 7: want to slow growth and disappoint external clients. Now we
[8:47] Speaker 7: certainly are cognizant that in this market environment, over indexing
[8:50] Speaker 7: to investment and under indexing to short term return isn't
[8:54] Speaker 7: being rewarded. But we think Alphabet is making the right
[8:57] Speaker 7: long term decisions when angling again still larger market opportunity
[9:01] Speaker 7: for AI over the next couple of years.
[9:04] Speaker 2: We've got a lot of stats that's about Gemini Stat's
[9:07] Speaker 2: about enterprise adoption, the cloud units growing eighty two percent
[9:11] Speaker 2: year on year. For me, the really simple question is
[9:15] Speaker 2: is Google doing well at AI?
[9:19] Speaker 7: They are still an AI winner in our view. The
[9:23] Speaker 7: market took a step back from that view overnight. The
[9:27] Speaker 7: delays around three point five pro and the fact that
[9:30] Speaker 7: they no longer have a foundational model that sits right
[9:33] Speaker 7: at the frontier of performance and benchmarking has definitely taken
[9:38] Speaker 7: a little bit of the shine off the AI winner theme.
[9:41] Speaker 7: What sound upper I talked about last night is that
[9:44] Speaker 7: they're likely going to have to wait for Gemini for
[9:47] Speaker 7: to be back at the frontier of performance with AI models.
[9:51] Speaker 5: Two points.
[9:52] Speaker 7: I think generally, when you look at access to chips data,
[9:56] Speaker 7: the ability to train these models, we think Alphabet is
[9:59] Speaker 7: as well positioned as anyone. But there can be short
[10:02] Speaker 7: term gaps that open up between performance and training runs
[10:05] Speaker 7: around these models. More importantly, we think the world is
[10:08] Speaker 7: broadly shifting from token maxing to token optimizing. And some
[10:13] Speaker 7: of these other models that are around speed and efficiency,
[10:17] Speaker 7: including some of the flash models that they released, will
[10:20] Speaker 7: allow them to remain very competitive for incremental workloads. But
[10:24] Speaker 7: investors want to see companies spending this amount of money,
[10:27] Speaker 7: then they want them at the frontier of model performance.
[10:30] Speaker 7: They might have to wait a few months for that
[10:32] Speaker 7: with Alphabet.
[10:33] Speaker 2: That is a conversation I've been having with CEOs all
[10:36] Speaker 2: across the stack recently, the difference between token maxing and
[10:40] Speaker 2: token optimizing. If Google nails that, where does it show up?
[10:44] Speaker 5: Right?
[10:44] Speaker 2: I think you right at the top of your note
[10:46] Speaker 2: that cloud revenue estimates now revised even higher.
[10:49] Speaker 3: Is that still the metric to follow.
[10:51] Speaker 2: On how they are being used out in the real world.
[10:55] Speaker 7: Yes, and we believe companies like Alphabet and next week
[10:58] Speaker 7: we'll hear this from Amazon that are going into enterprise
[11:01] Speaker 7: customers and saying we're going to help you optimize your spend.
[11:05] Speaker 7: It's not going to be about just buying tokens no
[11:08] Speaker 7: matter what the cost from a single model, but buying
[11:11] Speaker 7: a wider array of tokens from a wider array of
[11:14] Speaker 7: models is generally where this landscape is going. We wrote
[11:17] Speaker 7: a note a couple of months ago about where the
[11:19] Speaker 7: AI economy would go over the longer term, and I
[11:22] Speaker 7: think what got lost in that note ed would be
[11:24] Speaker 7: the fact that to drive utility and to drive token growth,
[11:29] Speaker 7: you need deflation. Every technology compute shift i've ever covered
[11:33] Speaker 7: an analyzed has unit growth that comes with deflation because
[11:37] Speaker 7: you have to incent adoption rates. And we don't think
[11:40] Speaker 7: the AI economy is going to be any different than that.
[11:43] Speaker 3: We don't have time for this.
[11:43] Speaker 2: But China's focused on lowering dollar p token, America's focused
[11:47] Speaker 2: on the quality of the token. I just note very
[11:49] Speaker 2: quickly that the other hyperscaler is as by association, markedly
[11:53] Speaker 2: lower today. Eric Sheridan of Goldman Sachs really enjoyed having
[11:55] Speaker 2: you on the program. Thank you very much. One more
[11:58] Speaker 2: nugget from Alpha BET's earnings. Its early bets are paying off.
[12:03] Speaker 2: Google says it's sitting on ninety four billion dollars worth
[12:07] Speaker 2: of SpaceX shares after SpaceX's IPO. Together with its anthropic steake,
[12:12] Speaker 2: those holdings delivered nearly one hundred billion dollars in gains
[12:16] Speaker 2: last quarter alone, but for now, most of those SpaceX shares,
[12:19] Speaker 2: of course, remain under lock up restrictions, limiting when alphabet
[12:24] Speaker 2: can cash in cash out. Coming up, a White House
[12:28] Speaker 2: official accuses China's Moonshot of improperly using USAI models and
[12:33] Speaker 2: in video chips to create the Commune K three system.
[12:36] Speaker 2: Details next, this is bloomboog Tech. Yesterday, Open Ai said
[12:47] Speaker 2: its advanced AI models mistakenly breached Hugging Faces systems during
[12:52] Speaker 2: a controlled cybersecurity test. Bloomberg has now learned the models
[12:55] Speaker 2: completed the attack in just hours, a task that would
[12:58] Speaker 2: typically take skill hackers weeks.
[13:01] Speaker 3: That's according to sources.
[13:02] Speaker 2: Open Ai says it's continuing a joint investigation with Hugging Face.
[13:08] Speaker 2: Sticking with AI, China's Moonshot improperly used USAI models and
[13:13] Speaker 2: in video chips to create the Kimmy K three system.
[13:17] Speaker 2: That's according to White House Office of Science and Technology
[13:20] Speaker 2: Policy Director Michael Kratzios, making the case in a social
[13:24] Speaker 2: media post yesterday. Nvidia and Moonshot didn't respond to requests
[13:28] Speaker 2: for comment. Bloomberg's AI reporter in DC, Maggie Easton joins
[13:33] Speaker 2: us for more. It is a big accusation by director Kratzios.
[13:38] Speaker 2: It's a detailed post. Give us the reporting, the specifics
[13:43] Speaker 2: of the accusation, what we need to know.
[13:46] Speaker 8: There are two key points here that this White House official,
[13:49] Speaker 8: Michael Kratzios is making. The first is that Moonshot, again
[13:53] Speaker 8: the maker of Kimmy K three access Grace Blackwell's so
[13:57] Speaker 8: those are in Nvidia servers in Vidiaga, inside servers that
[14:02] Speaker 8: Chinese companies are not allowed to purchase. So that was
[14:04] Speaker 8: kind of the first accusation.
[14:05] Speaker 9: He also said that they access those chips in Thailand
[14:09] Speaker 9: as well as acquired them. And then the second accusation
[14:11] Speaker 9: here is that Moonshot actually distilled from US models. Now
[14:16] Speaker 9: this is kind of like an emerging technique that Washington
[14:18] Speaker 9: has been quite worried about, in which the Chinese companies
[14:22] Speaker 9: are using outputs from the US companies to then feed
[14:25] Speaker 9: them into sort of copycat versions of the US products.
[14:30] Speaker 2: Bloomberg Tech has made every effort to invite mister Kratzios
[14:34] Speaker 2: on the show because people have very simple questions. For example,
[14:38] Speaker 2: they look at when Andthropics fable was released and then
[14:42] Speaker 2: controlled by the US government and when K three was
[14:44] Speaker 2: released and say, well, how is that possible on the
[14:47] Speaker 2: distillation side, give us, I guess the context, then the
[14:51] Speaker 2: size and scope of K three, and then the distillation
[14:54] Speaker 2: accusation a little bit more based on what Misscratsjos is saying.
[15:00] Speaker 9: Yeah, I think there's a lot of questions in d
[15:02] Speaker 9: C and beyond around this uh, you know, question of
[15:07] Speaker 9: distillation and whether this is a real, uh a real
[15:11] Speaker 9: thing that Washington needs to be concerned about, right because
[15:14] Speaker 9: Anthropic and Opening Eye have been raising these concerns for
[15:16] Speaker 9: the better part of a.
[15:18] Speaker 8: Year, and essentially some would still say that this distillation
[15:22] Speaker 8: technique is perfectly fine. And as you point out, the
[15:25] Speaker 8: time between you know, Anthropic's latest release and the release
[15:29] Speaker 8: of Kidney K three's is not that long. So some
[15:32] Speaker 8: have pointed to that timeline and question, you know, where's
[15:34] Speaker 8: the evidence that this distillation is really happening. You know,
[15:37] Speaker 8: China obviously has a lot of you know, talent in
[15:40] Speaker 8: the AI industry as well, so there's still you know,
[15:42] Speaker 8: this lingering question of how much did Moonshot rely on
[15:46] Speaker 8: US technology and how much is their own innovation.
[15:50] Speaker 2: Bloomberg's mag Eastlam with a critically important report on Bloomberg today.
[15:53] Speaker 3: Thank you very much.
[15:54] Speaker 2: Now coming up, Mobilized founder and CEO Amnon Shasher announced
[15:58] Speaker 2: today that he'll be stepping down after twenty seven years.
[16:02] Speaker 2: Shashua joins us next for an exclusive interview.
[16:04] Speaker 3: Stay tuned. This is Bloomberg Tech.
[16:15] Speaker 2: It's a big dave for Mobili, whose founder is stepping
[16:18] Speaker 2: down as CEO after twenty seven years. The news comes
[16:22] Speaker 2: as Mobili released its second quarter earnings earlier today, beating
[16:25] Speaker 2: analyst estimates with a reported revenue of five hundred and
[16:28] Speaker 2: eight million dollars, just above its strongest quarter of twenty
[16:31] Speaker 2: twenty five. Mobile I founder and CEO and Non Shasha
[16:35] Speaker 2: joins us now for an exclusive interview.
[16:38] Speaker 3: Welcome back to the show.
[16:40] Speaker 2: And I think the easiest place to start is why
[16:43] Speaker 2: why you're stepping down? Why you think now is the
[16:45] Speaker 2: right time to hand the reins over to someone else?
[16:49] Speaker 5: Well, good morning, Ed.
[16:51] Speaker 10: I think that there's never a good time, but at
[16:54] Speaker 10: the same time, this is the best time because you know,
[16:56] Speaker 10: Mobili built a great foundation, logical foundation going forward, and
[17:02] Speaker 10: we are at an inflection point in which now there
[17:05] Speaker 10: is no open scientific problem in the tax and the
[17:09] Speaker 10: software tax that we are developing. Everything is running either
[17:13] Speaker 10: offline or online, and it's ready both for Robotaxi and
[17:17] Speaker 10: for everything that we are developing. On the other hand,
[17:20] Speaker 10: there is huge expansion, operational expansion, go to market expansion.
[17:24] Speaker 10: For example, in the Robotaxi, we want to go B
[17:28] Speaker 10: two C to explore B two C not only be
[17:30] Speaker 10: to be humanoids. We want to it's a new thing
[17:33] Speaker 10: for us B two C as well. And at the
[17:36] Speaker 10: same time AI is moving very very fast, as you've
[17:39] Speaker 10: just mentioned in your previous articles, and as a scientist,
[17:45] Speaker 10: this is really my strongest point in my contribution to MOBILI.
[17:50] Speaker 5: So I think it's the best time.
[17:51] Speaker 10: To bring a new CEO that will take care of
[17:53] Speaker 10: the growth and me focus on the long horizon thinking.
[17:59] Speaker 2: You stay posts until a success is found. How active
[18:03] Speaker 2: will you be in the process and on finding your replacement.
[18:08] Speaker 10: Well, the board has a search cheer committee. I'll of
[18:10] Speaker 10: course be very contributing to it and also our management team.
[18:18] Speaker 10: We want to cast a very wide net. We want
[18:20] Speaker 10: to bring the best CEO we are looking. I'm looking
[18:26] Speaker 10: many years into the future and at this point in time,
[18:31] Speaker 10: if you find an excellent to CEO, the growth potential
[18:34] Speaker 10: of Mobilizer is huge, It's really huge.
[18:38] Speaker 2: The stock is down significantly, right, fifteen percent, biggest drop
[18:42] Speaker 2: since August of twenty twenty four. And you know, the
[18:44] Speaker 2: cell side acknowledge is that your your decision to drop
[18:50] Speaker 2: to step down is overshadowing a pretty strong set of results.
[18:56] Speaker 2: You know, has that surprised you the reaction to this, Well.
[19:01] Speaker 10: Actually I thought the stock would go up, but I
[19:04] Speaker 10: think that the markets they don't like uncertainty. And you know,
[19:08] Speaker 10: once it will be internalized that I'm here to stay.
[19:12] Speaker 10: I'm not going anywhere, just you know, releasing myself from
[19:16] Speaker 10: the day to day management and focusing on what really matters,
[19:20] Speaker 10: which is the future, technology of the future, science of
[19:23] Speaker 10: the future. You know, Mobilize is one of the few,
[19:26] Speaker 10: really very few companies in the physical a space that
[19:30] Speaker 10: does both autonomous cars and humanoid the robotics. This is
[19:35] Speaker 10: this is really this is really huge and there are
[19:37] Speaker 10: lots of technological ideas going forward, and this is what
[19:41] Speaker 10: I want to focus on.
[19:43] Speaker 2: So this is also a fundamental question for the company, right,
[19:45] Speaker 2: you know, whoever comes in as the next leader of mobili,
[19:50] Speaker 2: what is your expectation that they just completely shift the
[19:53] Speaker 2: company's focus to robotaxi as a domain as opposed to
[19:58] Speaker 2: being a supplier of SEC for AIDAS. Just going deeper
[20:02] Speaker 2: into that segment.
[20:04] Speaker 5: Uh No.
[20:05] Speaker 10: Look, AIDAS is contributing today about two billion dollars of
[20:09] Speaker 10: revenue per year four hundred million.
[20:11] Speaker 5: Dollars of a profit.
[20:15] Speaker 10: Uh there is at least a single digit growth in
[20:18] Speaker 10: the year on year on AIDAS. It is, it is
[20:23] Speaker 10: really a cash cow. Nobody wants to remove the focus
[20:26] Speaker 10: from from AIDAS. The challenge is to increase not instead,
[20:31] Speaker 10: but increase focus on robotaxi and the humanoid on first robotaxi.
[20:36] Speaker 5: This is this is really around the corner.
[20:39] Speaker 10: We have you know, cooperation with the Folkswagen with the Moya,
[20:44] Speaker 10: our first City Orlando and Lake Nona. Just last week
[20:49] Speaker 10: there was a very big demonstration of end to end
[20:53] Speaker 10: including tele operation.
[20:55] Speaker 5: The KPIs are are on.
[20:57] Speaker 10: Track commercialization, commercial deployment by the end of the year.
[21:02] Speaker 5: Everything looks it looks looks really good.
[21:06] Speaker 10: And now that you know the challenges of operational go to market,
[21:11] Speaker 10: building the operations and it has it's really orthogonal to
[21:16] Speaker 10: aid US.
[21:17] Speaker 2: So I suppose you envisage the next lead that being
[21:20] Speaker 2: more operational rather than from a science, science and sort
[21:25] Speaker 2: of entrepreneur background like you have, right, I think, just
[21:28] Speaker 2: to end this, I'm known to be clear, this was
[21:31] Speaker 2: your idea. This was you wanting to move to something new,
[21:35] Speaker 2: as opposed to the board or anyone else saying, you
[21:38] Speaker 2: know what, we should make a change.
[21:40] Speaker 5: Well, it's my idea.
[21:42] Speaker 10: And also the board has offered me the chairman position,
[21:46] Speaker 10: so it shows confidence, it shows confidence in me, it
[21:49] Speaker 10: shows confidence in my future contribution. And it's my idea
[21:54] Speaker 10: because I think it is the right time not to
[21:56] Speaker 10: wait too long, because it's an inflation point and that
[22:01] Speaker 10: was really the right time to bring someone that can
[22:03] Speaker 10: help with the growth.
[22:05] Speaker 3: Just very quick.
[22:06] Speaker 2: We have fifteen seconds for which I apologize, But do
[22:09] Speaker 2: you have a target timeline date?
[22:13] Speaker 5: No, No, I'm here to stay.
[22:14] Speaker 10: Well, it'll take as much time it'll take to find
[22:16] Speaker 10: a success or a CEO, And as I said, we're
[22:20] Speaker 10: casting a white net to find to find the best
[22:23] Speaker 10: CEO possible.
[22:24] Speaker 2: Right Mobile Ie founder and CEO Amnon Shashua, thank you
[22:29] Speaker 2: very much. Coming up, IBM cuts, it's for your sales
[22:32] Speaker 2: outlook after weakness in its mainframe demand. We speak with
[22:35] Speaker 2: IBM CEO Arvin Krishna, it's half time, will be right back.
[22:39] Speaker 3: This is what the markets look like.
[22:41] Speaker 2: Alphabet and its raising of CAPEX is weighing down quite
[22:46] Speaker 2: a lot of the tech sector Tesla too. This is
[22:48] Speaker 2: Bloomberg Tech. Welcome back to Bloomberg Tech, chez VIBM trading
[23:03] Speaker 2: NEI le lowist level since November twenty four. September twenty
[23:06] Speaker 2: twenty four, the company dialed back its full year sales
[23:09] Speaker 2: forecast after a pretty steep drop in mainframe sales, which
[23:13] Speaker 2: weighed on results.
[23:14] Speaker 3: Joining us from New.
[23:15] Speaker 2: York is the co host of Bloomberg's The Clothes Remain
[23:18] Speaker 2: Bustic alongside IBM CEO of In Krishna remain Arvan.
[23:22] Speaker 11: You've seen the reaction amongst investors. Here are some concerns
[23:25] Speaker 11: here about that lowered sales forecast overall, as well as
[23:28] Speaker 11: softness in software. You've characterized this as basically a shortfall
[23:34] Speaker 11: for one quarter that is limited to CAPEX sensitive areas
[23:37] Speaker 11: of the portfolio, but that's still a meaningful area of
[23:40] Speaker 11: your portfolio. Were sales coming into that quarter? Was that
[23:43] Speaker 11: pipeline overstated?
[23:46] Speaker 12: I don't believe so, because when we look at all
[23:49] Speaker 12: the deals that didn't close. I think we have done
[23:52] Speaker 12: enough verification, including with the clients, to know that they
[23:57] Speaker 12: were very real and it was a reproductization of the
[24:00] Speaker 12: CAPEC spend at the end of the quarter. This was
[24:03] Speaker 12: pretty confined to I'll call it the fortune one hundred
[24:07] Speaker 12: the deals that we're within a subset of those. Now
[24:10] Speaker 12: one third of what didn't happen has already come back,
[24:14] Speaker 12: So that tells us that this was a reprioritization and
[24:17] Speaker 12: those deals are very real as opposed to us being
[24:20] Speaker 12: optimistic in our projections and romain. I would also add,
[24:25] Speaker 12: I think that maintaining a free cash flow tells us
[24:28] Speaker 12: that we have levels around productivity and conviction and confidence
[24:32] Speaker 12: in the business, and that is also I think going
[24:34] Speaker 12: to serve on investor as well.
[24:36] Speaker 13: But the next few months will tell us that well.
[24:38] Speaker 11: On that cash flow figure, yes, and that certainly pleased
[24:41] Speaker 11: a lot of analysts and investors out there, that billion
[24:43] Speaker 11: dollar number of protective free cash flow growth. You maintaining
[24:47] Speaker 11: the dividend as well, but you're largely done that so
[24:49] Speaker 11: far by cutting costs. So that raises the question that
[24:52] Speaker 11: if we are anticipating slower growth on the revenue side,
[24:57] Speaker 11: does that mean more cost cuts are in store.
[25:00] Speaker 12: So the bulk of our cash flow growth over the
[25:04] Speaker 12: last four years has actually been on a just adbida,
[25:08] Speaker 12: So that tells you that this is mostly through revenue growth.
[25:11] Speaker 12: And our model has always been that we are the
[25:14] Speaker 12: last dollar is more productive and more profitable than the
[25:17] Speaker 12: first dollar. So we've been growing revenue four five percent
[25:21] Speaker 12: and we've been growing cash flow up in the seven
[25:24] Speaker 12: eight nine percent. So that's kind of our model and
[25:26] Speaker 12: we intend to keep maintaining that now. Right now, if
[25:30] Speaker 12: we drop revenue by one point because we said four
[25:33] Speaker 12: to five instead of five plus, we can absolutely make
[25:37] Speaker 12: it through productivity. Cost cuts is an interesting question. Cost
[25:41] Speaker 12: cuts doesn't always come down to people reduction in headcount.
[25:45] Speaker 12: Our headcount has been more or less flat over the
[25:48] Speaker 12: last many years. I think there's a lot of third
[25:50] Speaker 12: party spend where we are going to get a lot
[25:53] Speaker 12: more efficient with that third party spend than we have
[25:56] Speaker 12: been always.
[25:59] Speaker 2: In BEG Techs Live on Bloomberg Television and Bloomberg Radio,
[26:02] Speaker 2: and we're speaking with the IBM CEO of In krishnav
[26:04] Speaker 2: in good morning. You want to focus on accelerating revenue
[26:09] Speaker 2: growth and accelerating profitability and just really simply I'd love
[26:12] Speaker 2: to hear what you're asking the team to do differently
[26:15] Speaker 2: now in response to all of the factors that you outlined.
[26:19] Speaker 12: So ed really, so, if I look at our software business,
[26:24] Speaker 12: eighty percent of it is already an annuity consumption opics
[26:28] Speaker 12: bused business. Twenty percent of it is a CAPEX business.
[26:32] Speaker 12: If we think that the CAPEX headwinds are going to continue,
[26:36] Speaker 12: but eighty percent is already growing at about eight percent,
[26:40] Speaker 12: we want to put a lot more focus. So we
[26:42] Speaker 12: are going to direct a lot of the team with
[26:44] Speaker 12: forward deployment engineers, with people who are focused on deploying
[26:47] Speaker 12: the software at clients much more technical help and make
[26:51] Speaker 12: that eighty percent grow even faster. Products like red Hat, Confluent,
[26:55] Speaker 12: Harshi all fit that model. On the CAPEX side, we
[27:00] Speaker 12: have to make sure that while we can continue to
[27:02] Speaker 12: do it, don't depend upon outsized growth on that side
[27:06] Speaker 12: to go there. Then on the supply chain, can we
[27:09] Speaker 12: leverage all of our capability and supply chain to make
[27:12] Speaker 12: sure we have enough distributed infrastructure in storage in Unix
[27:17] Speaker 12: systems that people can fulfill all of the demand. Because
[27:20] Speaker 12: we came out of the second quarter with half a
[27:22] Speaker 12: billion dollars of backlog in that part of the portfolio.
[27:26] Speaker 12: So those give you an idea of the kind of
[27:28] Speaker 12: changes that we're making already, not just for the rest
[27:32] Speaker 12: of the half.
[27:33] Speaker 2: You summarize the state of the world beautifully. You said
[27:37] Speaker 2: that customers shifted spending towards servers, storage and memory in
[27:42] Speaker 2: late June, and when I posted on social media You're
[27:45] Speaker 2: coming on the show that the question from the audience
[27:47] Speaker 2: really simple. Did that trend continue from June into July?
[27:51] Speaker 2: And for how long do you expect it to last?
[27:55] Speaker 12: We have not seen it in July, but I would
[27:58] Speaker 12: tell you, look, conductor pricing memory is up what three
[28:02] Speaker 12: to four times over the last eighteen months. Networking infrastructure
[28:07] Speaker 12: is up sixty to eighty percent, Fiber is up, connectors
[28:12] Speaker 12: are up. I do think that some of these trends
[28:14] Speaker 12: on the underlying components are going to carry on for
[28:17] Speaker 12: some more time. That works its way up into those things.
[28:21] Speaker 12: So people are going to have those prices. Now, the
[28:24] Speaker 12: question is is that going to maintain a reprioritization of
[28:27] Speaker 12: the capex spent? If I take the signal that a
[28:30] Speaker 12: third of our deals closed, then that says no. People
[28:32] Speaker 12: are going to get more careful about how to spend
[28:35] Speaker 12: their capex. If that goes into September, because a lot
[28:39] Speaker 12: of capex does get committed at the end of a quarter,
[28:42] Speaker 12: not always through then it means that that carries on
[28:46] Speaker 12: for some more time. And that is why we gave
[28:49] Speaker 12: a guide of four to five percent. Depending upon if
[28:52] Speaker 12: that carries on, we'll be at the lower end of
[28:54] Speaker 12: the range, but some of it comes back, then we'll
[28:56] Speaker 12: be at the high end of the range.
[28:57] Speaker 11: So, Alvin, I know there's a big focus right now
[28:59] Speaker 11: on on sort of the clients that sort of did
[29:03] Speaker 11: not necessarily materialize in the quarter. With regards to the
[29:05] Speaker 11: existing clients that you have, and I'm primarily referring to
[29:08] Speaker 11: your mainframe business, I mean, can you share, like sort
[29:11] Speaker 11: of what percentage of those clients are actually increasing their spending?
[29:15] Speaker 11: Is that going up still?
[29:16] Speaker 13: Absolutely?
[29:17] Speaker 5: So.
[29:18] Speaker 13: The current machine is called the Z seventeen.
[29:21] Speaker 12: The Z seventeen has been from the beginning, which was
[29:25] Speaker 12: May of twenty twenty five to today at one hundred and
[29:28] Speaker 12: thirty percent in terms of the capacity growth compared to
[29:31] Speaker 12: the prior machine. That's one very strong signal, and that's
[29:35] Speaker 12: an aggregate. Then you can say, hey, is everybody in
[29:38] Speaker 12: there or only a few eighty five percent of the
[29:41] Speaker 12: clients are increasing their capacity as opposed to the fifteen
[29:45] Speaker 12: who are not. That optimization goes on all the time,
[29:48] Speaker 12: and I would tell you this is probably the best
[29:50] Speaker 12: that we have seen.
[29:51] Speaker 13: In a long time of what is going on there.
[29:54] Speaker 13: I'm a software going to.
[29:55] Speaker 11: Lag, Yeah, software is going to I mean, look, we
[29:58] Speaker 11: know the hardware capacity is there, softw where's lack? And
[30:00] Speaker 11: I understand why it's lagging in the moment. And I'm
[30:03] Speaker 11: going to ask you a question, and I forgive me
[30:05] Speaker 11: if it's a bit unfair, but I look back to
[30:07] Speaker 11: the nineteen nineties and whether there are some parallels to
[30:10] Speaker 11: some of the mainframe issues that IBM went through back
[30:12] Speaker 11: then under a predecessor two or three times removed from you,
[30:16] Speaker 11: John Akers, and this idea that at that time we
[30:19] Speaker 11: were going through this paradigm shift in computing and how
[30:22] Speaker 11: companies spent and allocated their money, and there was a
[30:25] Speaker 11: lot of talk by the CEO then that the issues
[30:28] Speaker 11: were temporary, that they were economic, and that they were
[30:30] Speaker 11: right them sales. We know in hindsight that it was
[30:32] Speaker 11: much more structural. Why should we now look at this
[30:36] Speaker 11: major shift going on with AI and the computation involved
[30:40] Speaker 11: in that, and think that this time is different.
[30:44] Speaker 12: Well, we have to look at what the clients are doing.
[30:46] Speaker 12: I always start there. Our opinion is an opinion. What
[30:49] Speaker 12: clients do is what matters. So when I check with
[30:52] Speaker 12: my clients, those who do credit card authorizations, are you
[30:55] Speaker 12: going to maintain the mainframe? And they go to, well,
[30:58] Speaker 12: the resilience, the amount of capacity, the unit cost of
[31:03] Speaker 12: it being five to fifteen times cheaper is important. The
[31:06] Speaker 12: big difference from that time early nineteen nineties to today
[31:10] Speaker 12: Romain is the cost issue. You could not argue that
[31:15] Speaker 12: the mainframe was cheaper on a unit cost basis for
[31:18] Speaker 12: the workloads that were moving off. Then at that time,
[31:21] Speaker 12: the mid range computers I'll call it the whole Unix.
[31:25] Speaker 12: It wasn't really client server, it was Unix taking the
[31:28] Speaker 12: workloads off in the early nineteen nineties. That issue you
[31:32] Speaker 12: have to look at if there is a cheaper alternate
[31:35] Speaker 12: for that workload. I'll sort of look at you and
[31:38] Speaker 12: say that means in five to fifteen years it.
[31:40] Speaker 13: Will move off. But that's not the case right now.
[31:43] Speaker 13: For the workloads that are on.
[31:45] Speaker 12: We can show the clients that it's five to fifteen
[31:47] Speaker 12: times cheaper to keep it on the mainframe.
[31:49] Speaker 13: They're not.
[31:50] Speaker 12: But that's not all workloads. That is workloads that need protection,
[31:54] Speaker 12: that need resilience, that need the burst capacity that comes.
[31:58] Speaker 12: And so for those kinds of workload, the main brame
[32:01] Speaker 12: is the architecturally superior platform. That was not the case
[32:04] Speaker 12: thirty years ago for the workload that did go off.
[32:09] Speaker 2: Live on Bloomberg Television and on Bloomberg Radio. This is
[32:12] Speaker 2: Bloomberg Tech speaking with IBMCO r Vin Krishna. I've been
[32:16] Speaker 2: listening to you a lot recently on long form podcasts,
[32:19] Speaker 2: some clips on the social media about your view of
[32:21] Speaker 2: the world, what it's like to be a CEO in
[32:24] Speaker 2: the domains that IBM operates in. And it's interesting how
[32:27] Speaker 2: quickly it comes back to the macroeconomic backdrop. A lot
[32:31] Speaker 2: of people looked at the guidance for the balance of
[32:33] Speaker 2: this year and would say, and they do say, Orevin
[32:36] Speaker 2: on this show, memory prices are not changing, they continue
[32:40] Speaker 2: to push higher. They look at the outlook for growth.
[32:46] Speaker 2: Could you just explain the data points you rely on
[32:48] Speaker 2: that give you the confidence on the new guidance that
[32:51] Speaker 2: you've given and whether it is actually achievable or it
[32:55] Speaker 2: will be difficult to meet.
[32:56] Speaker 12: Yeah, So ed, I'll come back to For ourselves, we
[33:01] Speaker 12: have to look at our demand pipelines and our yields
[33:04] Speaker 12: and how we get things going. But I think for
[33:07] Speaker 12: your audience, let's look at it this way. Number one
[33:10] Speaker 12: most important is what is GDP growth going to be.
[33:14] Speaker 12: We think that that's between two and three percent for
[33:16] Speaker 12: the year. If I look at the globe and it
[33:19] Speaker 12: is going to be consistent, even in the Middle East,
[33:21] Speaker 12: even in Asia where there is a lot more energy
[33:24] Speaker 12: and disruption, we actually see a lot of growth. Tech
[33:27] Speaker 12: is going to be I think two to three to
[33:29] Speaker 12: four points above that. So that puts tech in terms
[33:33] Speaker 12: of what the market is somewhere in the five six
[33:36] Speaker 12: seven percent.
[33:37] Speaker 13: Then that comes back to.
[33:38] Speaker 12: What parts of our portfolio can play against that demand
[33:42] Speaker 12: and what parts cannot. And so I look at the
[33:44] Speaker 12: parts of the software portfolio. That's what I talked about,
[33:46] Speaker 12: the eight percent growth in the eighty percent of it
[33:49] Speaker 12: that can play right into that. Then I look at
[33:52] Speaker 12: our distributed infrastructure that can play right into that. I
[33:56] Speaker 12: think consulting will be that one two three percent. It
[33:59] Speaker 12: is not going to be in the double digit growers,
[34:02] Speaker 12: but we see the demand, and we see the signings,
[34:05] Speaker 12: and we see the clients leaning in to say they
[34:07] Speaker 12: want transformercial work done. That's how we kind of know,
[34:11] Speaker 12: beginning with the macro and then coming down.
[34:14] Speaker 13: I'll make a prediction for you. I think technology.
[34:18] Speaker 12: Spend is going to become a larger and larger part
[34:21] Speaker 12: of every enterprise's budget. Used to be three percent. Since
[34:26] Speaker 12: Remain raised the thirty years ago, is probably up at
[34:29] Speaker 12: five to six percent on average. I will not be
[34:31] Speaker 12: surprised if by twenty thirty five it's ten percent of
[34:34] Speaker 12: everyone's budget.
[34:36] Speaker 3: Give you a quick micro case study.
[34:38] Speaker 2: Bloomberg reported that Starbucks is replacing some IBM tools on
[34:41] Speaker 2: the software side with in house.
[34:44] Speaker 3: Let's talk to that.
[34:46] Speaker 12: So Starbucks is about a little over two million dollars
[34:49] Speaker 12: a year client for IBM. The portion they're replacing is
[34:52] Speaker 12: a product called tri Rega that does real estate leads management.
[34:57] Speaker 12: The version of it that Starbucks has is almost ten
[35:01] Speaker 12: years old. I'm not surprised at a replacing it because
[35:04] Speaker 12: I've actually been describing publicly. Software which is largely interaction
[35:09] Speaker 12: based and is based on ease of use as opposed
[35:12] Speaker 12: to anything else, can be easily replaced by AI and.
[35:16] Speaker 13: AI agents, and that is what is going on there.
[35:19] Speaker 12: However, if I see other parts of Starbucks and maybe
[35:23] Speaker 12: the ability for our harshy portfolio or security portfolio, I'm
[35:28] Speaker 12: out if afraid it this way, I would not be
[35:30] Speaker 12: surprised if Starbucks is a larger client next year than
[35:33] Speaker 12: it was last year.
[35:34] Speaker 11: With regards to how this world is evolving, Arvin, I
[35:37] Speaker 11: am curious that just internally about your plans to hire,
[35:40] Speaker 11: particularly when it comes to the technological side, have you
[35:46] Speaker 11: been able to keep pace with some of the other
[35:48] Speaker 11: companies out there also trying to do what you do
[35:51] Speaker 11: and pay some of the salaries that.
[35:53] Speaker 13: You have to pay.
[35:55] Speaker 12: Well, we hired three times as many college hires this
[35:59] Speaker 12: year than it did last year, and I think that
[36:03] Speaker 12: given others seem to be backing off college hiring, it
[36:06] Speaker 12: gives us an incredible ability to bring in great talent
[36:09] Speaker 12: and to then grow them inside our company and to
[36:12] Speaker 12: offer them great careers. I think if I looked at
[36:16] Speaker 12: it last we had I think twenty million resumes in
[36:20] Speaker 12: our applicant database, so that gives us a huge field.
[36:24] Speaker 13: To go look at.
[36:25] Speaker 12: So I mean I don't worry about bringing in talent.
[36:29] Speaker 12: I actually worry much more about can we give them
[36:31] Speaker 12: a great career and a great pathway, because not everybody
[36:35] Speaker 12: is cut out to do work that is going to
[36:38] Speaker 12: be demanded, because I don't think there's a lack of employment,
[36:41] Speaker 12: but the nature of the work. If you can't use
[36:43] Speaker 12: AI tools, if you can't use the productivity tools, then
[36:47] Speaker 12: it's going to be really hard for you to be
[36:49] Speaker 12: competitive with your peers.
[36:51] Speaker 11: We always we already know sort of the potential impact
[36:54] Speaker 11: of what AI means for the economy and for your business.
[36:57] Speaker 11: There are a lot of people looking around the corner,
[36:59] Speaker 11: including yourself, to quantum. Is that a viable business on
[37:03] Speaker 11: the horizon or is that just a moonshot that you're
[37:06] Speaker 11: hoping actually.
[37:07] Speaker 12: Sticks well when things are two years away. I wouldn't
[37:10] Speaker 12: call them a moonshot. I think quantum is now in
[37:13] Speaker 12: the engineering realm, as opposed to the science realm for
[37:17] Speaker 12: the next five years, and I believe that by twenty
[37:20] Speaker 12: twenty nine we will deliver a machine that does one
[37:23] Speaker 12: hundred million competitions of large scale, fault tolerant quantum computer
[37:29] Speaker 12: in the next two years. I think that's an incredible opportunity.
[37:33] Speaker 12: I'll quantify it, and this is not just our work,
[37:35] Speaker 12: a lot of third parties. By the end of twenty thirties,
[37:40] Speaker 12: we think it's about a trillion dollar total market opportunity
[37:43] Speaker 12: in terms of value that quantum will create. That is
[37:47] Speaker 12: why we also double down this morning and announced that
[37:50] Speaker 12: we bought HRL from GM and Boeing, and that's going
[37:55] Speaker 12: to help us bring people with a lot of talent
[37:57] Speaker 12: around materials, spin, Dronis, quantum sensing and other sensor technologies
[38:04] Speaker 12: to add to our own effort, so we are even
[38:07] Speaker 12: more well positioned to go win in this market.
[38:11] Speaker 2: Arvin Krishna, IBM CEO, of course, alongside Bloomberg's remain bustic.
[38:15] Speaker 2: Thank you both very much. I would note that IBM
[38:17] Speaker 2: shares now modestly higher three tens to one percent, having
[38:20] Speaker 2: open lower. But of course IBM pre released some of
[38:23] Speaker 2: its financials July fourteenth, from the stockfell twenty two to
[38:26] Speaker 2: twenty five percent. We're going to get back to earnings
[38:30] Speaker 2: and back to Alphabet's big capex number, while Intel's up
[38:33] Speaker 2: next on the chopping block later today.
[38:36] Speaker 3: This is Bloomberg Tech.
[38:46] Speaker 2: Coast Adventures, one of open AI's early investors, is in
[38:49] Speaker 2: talks to raise five point five billion dollars in a
[38:51] Speaker 2: new set of venture investment funds. It would make it
[38:54] Speaker 2: the largest fundraising event in the firm's history. All according
[38:58] Speaker 2: to sources. Bloomberg's and Ascarinus broke the story. No surprise
[39:02] Speaker 2: there find this so interesting. Vinode Coast irregular on the show.
[39:07] Speaker 2: But I think that people may lose sight of how
[39:09] Speaker 2: big a firm it was already, and this raise is
[39:12] Speaker 2: very much in line with what's happening in industry.
[39:15] Speaker 14: I mean, if you were an early Open AI or
[39:18] Speaker 14: an anthropic backer, you probably have one of the best
[39:21] Speaker 14: stories in venture fund raising right now. I mean, last
[39:23] Speaker 14: month we talked about Menlo Adventures raising its largest fund
[39:26] Speaker 14: of all time. Now I'm back talking about Cosla and
[39:29] Speaker 14: early Open AI backer, the first outside investor.
[39:32] Speaker 2: In open five billions split across multiple funds.
[39:35] Speaker 14: We think, yeah, yeah, So in this case, they're going
[39:37] Speaker 14: to keep doing and try and replicate that open AI
[39:39] Speaker 14: early success. So they're going to be putting majority of
[39:42] Speaker 14: the capital into early stage bets, and two point five
[39:44] Speaker 14: billion is reserved for an opportunity fund, which is pretty
[39:47] Speaker 14: much for the later stage investments as round sizes get
[39:50] Speaker 14: bigger and concentration becomes a more in vogue strategy for
[39:53] Speaker 14: venture capitalists.
[39:54] Speaker 2: You know, we do know the firm's namesake, Vinodekosla, but
[39:57] Speaker 2: there are lots of interesting people at that firm investing
[40:00] Speaker 2: at different stages.
[40:02] Speaker 3: Tell us a bit more about the team.
[40:04] Speaker 14: Yeah, I mean this is a team that has basically
[40:07] Speaker 14: chosen to stick with an early stage focus when a
[40:09] Speaker 14: lot of people are getting into a uniquely broad set
[40:13] Speaker 14: of side quests as venture firms. I'm thinking of the
[40:17] Speaker 14: team that is backed Sakana Ai, which is building a
[40:20] Speaker 14: Japanese LM, companies that are looking at robotics and climate.
[40:25] Speaker 14: So yes, they're definitely going after the AI application layer.
[40:29] Speaker 14: But when I think of Coastal Adventures, I actually think
[40:32] Speaker 14: of a much more biotech focus, hard text focus, deep
[40:35] Speaker 14: tech focus.
[40:36] Speaker 2: There are areas of interest for Vino By point out
[40:37] Speaker 2: as well. Yes, absolutely, just real quick I would point
[40:41] Speaker 2: out that a spokesperson at the firm declined to comment
[40:43] Speaker 2: on our reporting. But there's another theme here, which is
[40:46] Speaker 2: coastal race four billion last year. Yeah, other firms have
[40:50] Speaker 2: done that. Big funds last year, quick follow on this year?
[40:53] Speaker 2: What do we need to know?
[40:54] Speaker 14: I mean, these people are going through their funds at
[40:56] Speaker 14: a faster than ever Cadence. I mean, we think that
[40:59] Speaker 14: this is all almost like you're proactively raising ahead of
[41:02] Speaker 14: needing to even touch the capital. So my understanding is
[41:05] Speaker 14: that last year's fund is still being actively deployed. This
[41:08] Speaker 14: fund we broke is being kicked off, and the fundraising
[41:12] Speaker 14: calls and conversations are happening right now. But it does
[41:14] Speaker 14: show you how competitive and how expensive being a venture
[41:17] Speaker 14: capitalist is today.
[41:18] Speaker 2: Bloomberg's and Tashmos are in US with another big story
[41:21] Speaker 2: in the world of bench capital.
[41:22] Speaker 3: Thank you.
[41:23] Speaker 2: Alphabet's increasing capex figures are raising concerns on the cost
[41:27] Speaker 2: of AI. Google's parent raised its capex productions to two
[41:30] Speaker 2: hundred and five billion dollars at the top end for
[41:33] Speaker 2: this year, but the company's cash flow went negative for
[41:36] Speaker 2: the first time in its history as a public company,
[41:38] Speaker 2: which goes back over twenty years. Bloomberg Intelligence senior analyst
[41:42] Speaker 2: Man deeps saying, joins us and Mandy Leeds our entire
[41:44] Speaker 2: team on the tech coverage.
[41:46] Speaker 3: Side at BI. I mean, you were in the.
[41:48] Speaker 2: Camp of people that saw capex going even higher than
[41:50] Speaker 2: two hundred and five billion dollars. But the milestone is
[41:53] Speaker 2: that swing to negative free cash flow. Are you in
[41:55] Speaker 2: the camp of people that are worried about that.
[41:58] Speaker 15: Not for Alphabet? And yes, I am in the camp
[42:02] Speaker 15: that capex is going to go much higher for twenty
[42:05] Speaker 15: twenty seven, and they did say there will be a
[42:08] Speaker 15: significant increase next year. So look, I mean, this is
[42:12] Speaker 15: a full stack company where they've already shown their CAPEX
[42:16] Speaker 15: spend is far more efficient than anyone else out there,
[42:20] Speaker 15: and that's why they're going big in terms of getting
[42:24] Speaker 15: as much capacity because so far, if you have the compute,
[42:28] Speaker 15: you have the power that's translating into cloud revenues and
[42:32] Speaker 15: we saw that in Alphabet's cloud segment growth eighty two percent,
[42:36] Speaker 15: and that's why I think they're going with that TPU
[42:39] Speaker 15: stack as well, where they want to sell that independently
[42:43] Speaker 15: of the cloud, and that could be a big line
[42:45] Speaker 15: of business over the next two years as well.
[42:48] Speaker 3: It's balancing at right.
[42:49] Speaker 2: Google is on the hook for eight hundred and eleven
[42:52] Speaker 2: billion dollars of spending and in your react you kind
[42:55] Speaker 2: of make the point that it's that versus the very
[42:58] Speaker 2: strong cloud gains that they're seeing.
[43:01] Speaker 15: Yeah, I mean, and look, I think they didn't quantify
[43:05] Speaker 15: the margins of those TPU systems. But once you start,
[43:09] Speaker 15: you know, basically selling your designs externally, which so far
[43:14] Speaker 15: TPUs were used mostly for Google Cloud. Now they're talking
[43:18] Speaker 15: about setting up anthropic data centers with their own design
[43:21] Speaker 15: and generating revenue out of that. That's huge, and you know,
[43:25] Speaker 15: it could be a big line of business on its own.
[43:27] Speaker 15: So from that perspective, Google has got it all in
[43:31] Speaker 15: terms of large angred models, TPU systems, and then the
[43:34] Speaker 15: cloud business, and that's why they are really going big
[43:37] Speaker 15: in terms of their capex increase.
[43:39] Speaker 2: So the thing about the TPU business that's called it
[43:42] Speaker 2: is they have these sales packs and I think what
[43:44] Speaker 2: Alphabet CFO said was the company won't even realize revenues
[43:48] Speaker 2: from that until I think they said twenty seven, but
[43:50] Speaker 2: they didn't say which part of twenty seven. But you're
[43:52] Speaker 2: basically saying that's going to be an important business line
[43:54] Speaker 2: for them.
[43:55] Speaker 15: Yes, And I look at what Nvidia has done with
[43:58] Speaker 15: their system sales and how big n Video has gotten
[44:01] Speaker 15: over the past three years. I mean, clearly, this is
[44:05] Speaker 15: a rising tide that's lifting all boats, and we are
[44:08] Speaker 15: in a supply constrained environment. So essentially, you know, for Alphabet,
[44:13] Speaker 15: they are the only ones who have the ability to
[44:16] Speaker 15: do that. Everyone else is still in their earlier versions
[44:19] Speaker 15: of their chip design. I think Amazon is a third
[44:23] Speaker 15: as well. But it's really about how many versions you
[44:26] Speaker 15: have had and how external customers can trust your design
[44:30] Speaker 15: for their workloads, and in this case, Alphabet has the
[44:33] Speaker 15: ability to do that for external workloads.
[44:36] Speaker 2: I would point out that if you just look at
[44:38] Speaker 2: the stock reaction, the stock is down the most since
[44:41] Speaker 2: May of twenty twenty five, and we're showing this astonishing
[44:43] Speaker 2: chart that for the first time in this company's public history,
[44:46] Speaker 2: it is swung to negative free cash flow. That's an
[44:49] Speaker 2: astonishing chart. Man Deep seeing a Bloomberg Intelligence, Thank you
[44:52] Speaker 2: very much. Next up on deck, Intel, the chip maker,
[44:56] Speaker 2: report it second quarter results today after the closing bow,
[44:59] Speaker 2: and investors keeping a close eye is Intel's It could
[45:02] Speaker 2: speak to the strength and breadth of the semiconductor industry
[45:05] Speaker 2: right now. Bloomberzie and King joins us for a preview Intel.
[45:09] Speaker 2: What to make of Intel? What to look for with
[45:11] Speaker 2: Intel this year?
[45:13] Speaker 16: When the stock is up one hundred and seventy two percent,
[45:16] Speaker 16: clearly investors have said, Okay, you're actually back in the game.
[45:20] Speaker 16: But then if you look under the hood at the
[45:23] Speaker 16: actual numbers, it really isn't. It's showing double digit growth,
[45:27] Speaker 16: it's showing some demand for its Zeon products, but compared
[45:31] Speaker 16: to growth compared to what other companies are putting up
[45:34] Speaker 16: and have been putting up for the last couple of years,
[45:36] Speaker 16: not quite there yet. So what Intel has really got
[45:38] Speaker 16: to show is that it's really really a fundamental part
[45:41] Speaker 16: of this AI race.
[45:42] Speaker 2: There's the CPU story, which has been interesting more recently,
[45:46] Speaker 2: and then there's its business as a third party contract
[45:49] Speaker 2: manufacturer and the latest technology process. I always feel like
[45:52] Speaker 2: we want to get answers on that and we don't.
[45:54] Speaker 16: Yeah, I mean that would blow the story wide open
[45:57] Speaker 16: and would make people consider Intel different, that this turnaround
[46:02] Speaker 16: would be vindicated. Haven't seen that yet. The company has said, look,
[46:05] Speaker 16: we can't talk about it. It's up to our customers.
[46:08] Speaker 16: You'll know we're doing it though. You'll know it's real
[46:10] Speaker 16: when we start to spend big money on Kapex to
[46:13] Speaker 16: build those factories out. So we'll be looking for that today.
[46:15] Speaker 2: I know this is a pretty simple question, but what
[46:17] Speaker 2: is like financial metric that you learn the most about
[46:20] Speaker 2: Intel from?
[46:21] Speaker 16: Yeah, obviously growth is very important, but this is a
[46:23] Speaker 16: company that's been losing money and its gross margin is
[46:26] Speaker 16: twenty points south of where it was in the good days, right,
[46:29] Speaker 16: so we need that margin.
[46:31] Speaker 2: Historically above sixty five percent, above sixty percent, above sixty percent,
[46:35] Speaker 2: and yeah, nowhere near that right now. Another busy afternoon
[46:39] Speaker 2: for un for me, Bloomberg Zy and King, thank you
[46:41] Speaker 2: very much. Heavy focus on technology earning. So that does
[46:45] Speaker 2: it for this position of Bloomberg Tech. But go back
[46:47] Speaker 2: and recap across Alphabet, Tesla, IBM, and then in twenty
[46:51] Speaker 2: four hours time we'll do it again with Intel.
[46:53] Speaker 3: Listen on the pod.
[46:54] Speaker 2: I would say that some of the moves directly tied
[46:58] Speaker 2: to earnings are big, right. Alphabet is also by association
[47:01] Speaker 2: dragging down quite a number of the other hyperscalers. Amazon,
[47:04] Speaker 2: for example, has a pretty deep decline for Tesla. This
[47:08] Speaker 2: is a big drop down fourteen percent, and really it's
[47:13] Speaker 2: still a spending.
[47:14] Speaker 3: Story across the board. Stay with us throughout the week,
[47:16] Speaker 3: that says bloombog Tech

Shows are listed from Apple Podcasts' charts and directory, and Podcast Index where it is used. Every show and episode belongs to its publisher and plays from the publisher's own feed; this site hosts none of them.
