RiskReversal Pod

Dan Ives argues AI boom differs from dot-com bubble, citing enterprise adoption gaps and profitability, while warning of infrastructure debt risks and geopolitical competition.

53 min Listen to the original
Artificial IntelligenceCloud ComputingEnterprise SoftwareCybersecurityGeopoliticsCapital ExpenditureIPO Market

Summary

Dan Ives, founder of Yorkville Ives, characterizes the current AI market phase as an 'air pocket' between capital expenditure and monetization, distinguishing today's profitable tech giants from the unprofitable dot-com era. He highlights that only 3.5% of Fortune 500 companies have fully adopted AI, suggesting significant growth potential for cloud providers like Microsoft Azure. Ives advises long-term investing over short-term trading, noting that volatility creates buying opportunities in quality names. The discussion also covers structural risks, including high debt levels in infrastructure plays like Oracle, potential compute gluts among 'neo-clouds,' and the financial pressure of nearly $800 billion in recent AI sector fundraising. Finally, Ives and the host analyze geopolitical threats, specifically China's open-source strategy and its impact on US tech valuations.

Episode sections

  1. 0:00 – 2:36

    Podcast Intro and Sponsors

    Introduction to the episode, rhetorical questions on AI volatility, and sponsor reads for Strategic Alternatives, iConnections, and the OK Computer Podcast.

  2. 2:36 – 4:22

    Dan Ives' New Firm and ETF Strategy

    Ives discusses his move to Yorkville Ives, describing it as a modern merchant bank, and highlights the launch of the $200 million Ives Ultra Fund for accessing private AI tech.

  3. 4:22 – 6:39

    AI Cycle Comparison: Dot-Com vs. Today

    Ives argues the current AI boom differs from the 1990s dot-com bubble due to the profitability, free cash flow, and reasonable multiples of today's major tech companies.

  4. 6:39 – 9:07

    The 'Air Pocket' and Enterprise AI Adoption

    Ives describes the market phase as an 'air pocket' between capex and monetization, citing that only 3.5% of Fortune 500 companies have fully adopted AI.

  5. 9:07 – 11:59

    Microsoft and Azure Growth Dynamics

    Discussion on Microsoft's enterprise dominance, the synergy between Azure and Office 365 Copilot, and the path of least resistance for enterprises within the Microsoft ecosystem.

  6. 11:59 – 15:53

    Alphabet's Advertising Challenge and Gemini

    Analysis of Alphabet's struggles with AI chatbots cannibalizing ad revenue and the company's pivot to monetize AI via Gemini for Workspace.

  7. 15:53 – 20:01

    Navigating Volatility and Stock Picking

    Ives advises against short-term trading, recommending long-term thesis-based investing and using examples like Intel and ServiceNow to illustrate thesis resilience.

  8. 20:01 – 30:25

    SaaS Disruption and Cybersecurity Opportunities

    Discussion on AI disrupting SaaS narratives and Ives' prediction that IT security budgets will double due to AI-driven threats.

  9. 30:25 – 30:44

    Accounting Mechanics and CapEx Spreading

    Explanation of how Microsoft spreads AI capex over five to six years, masking immediate cash impacts and boosting near-term S&P 500 earnings visibility.

  10. 30:44 – 32:19

    Infrastructure Bottlenecks and Political Risks

    Identification of data center moratoriums, energy access, and political polarization as key bottlenecks and risks to AI infrastructure deployment.

  11. 32:19 – 33:23

    Debt vs. Cash Flow in AI Infrastructure

    Differentiation between AI infrastructure companies based on debt levels, with caution expressed on Oracle due to high leverage.

  12. 33:23 – 34:23

    Neo-Clouds and Compute Glut Risks

    Discussion of Elon Musk's prediction that only 7-10 neo-clouds will survive, implying potential compute gluts and consolidation.

  13. 34:23 – 35:52

    Pricing Models and Token Economics

    Analysis of the industry transition from 'token maxing' to sustainable pricing models and the commoditization of AI models.

  14. 35:52 – 52:56

    Anthropic IPO and Market Appetite

    Speculation on Anthropic's IPO timing and the risk that poor Q3 data could negatively impact broader AI market sentiment.

Assets discussed

AssetKindStanceContext
NVIDIANVDAequitymixedIf you think about the hyperscalers were some of the early beneficiaries, obviously NVIDIA, a handful of other chip companies, but then we saw the trade broaden out a bit.
MicrosoftMSFTequitybullishLet's just talk about Microsoft and Meta, for instance, and without getting into their fundamentals, but the stocks have had each one of them three. Microsoft was down 35% from its 52-week highs as it's going back a few months ago.
AlphabetGOOGLequityneutralSo when we view like where Google is and where Amazon, I could argue like all three hyperscalers could have significant growth over the next two years. Oracle is different because they're trying to build themselves into a hyperscaler.
AmazonAMZNequitymixedSo when we view like where Google is and where Amazon, I could argue like all three hyperscalers could have significant growth over the next two years.
Meta PlatformsMETAequityneutralThis is gonna be something with Muse that Meta is gonna have to figure out over the next few months or so. So, how do you respond to that when someone like Elon or when you see the other Frontier labs really with the frontier models, not Grok, how do you see them going from token maxing to the way they're doing now in consumption?
AppleAAPLequitybullishBut it's no different than like if I think about Apple, it was when the stocks sold off to 220. You're like, you even if they don't even get AI right, you own okay.

“Stance” is how the speakers talked about the asset, not a recommendation.

Claims worth verifying

The NASDAQ was down 3.5% on the Friday prior to the recording, with some stocks down 6.5%.

Unidentified speaker ·1:11

How to verify: Check historical market data for the NASDAQ Composite index and individual stock performance for the trading day immediately preceding the podcast release date (November 3, 2026).

The S&P 500 averaged over 20% year-over-year gains from 1995 to 1999.

Unidentified speaker ·4:33

How to verify: Review historical S&P 500 annual return data from Yahoo Finance or Bloomberg for the years 1995 through 1999.

In March 2000, the average tech company traded at 32 times revenues, whereas today's major tech companies trade at 20 times earnings.

Unidentified speaker ·5:59

How to verify: Calculate the average Price-to-Sales ratio for a basket of tech companies in March 2000 and the average Price-to-Earnings ratio for major tech companies at the time of the podcast.

Only about 3.5% of Fortune 500 companies have fully adopted AI.

Unidentified speaker ·6:54

How to verify: Verify this statistic against recent enterprise AI adoption surveys from firms like Gartner, IDC, or McKinsey.

Microsoft Azure cloud growth was up 50% in the referenced quarter.

Unidentified speaker ·9:41

How to verify: Check Microsoft's quarterly earnings report for the period discussed to verify the reported growth rate of the Intelligent Cloud/Azure segment.

Google Cloud Platform (GCP) grew 100% year-over-year in the last quarter.

Unidentified speaker ·10:38

How to verify: Check Alphabet's quarterly earnings report for the reported growth rate of the Google Cloud segment.

Amazon Web Services (AWS) growth accelerated to 39% year-over-year.

Unidentified speaker ·10:45

How to verify: Check Amazon's quarterly earnings report for the reported growth rate of the AWS segment.

Intel stock sold off 30% (or 50% in one instance) following a 'gut punch' from management guidance.

Unidentified speaker ·17:00

How to verify: Review Intel's stock price movement and press releases/earnings guidance around the time of the podcast recording.

Fortinet stock was down approximately 25% at the time of the discussion.

Unidentified speaker ·24:32

How to verify: Check Fortinet's stock price performance relative to its recent highs around the podcast recording date.

The Ives Ultra Fund is a $200 million IPO vehicle for investing in private AI tech companies.

Unidentified speaker ·3:53

How to verify: Search for the prospectus or press release for the 'Ives Ultra Fund' IPO launched by Yorkville Ives.

Microsoft spreads its AI-related CapEx over five or six years for accounting purposes.

Unidentified speaker ·

How to verify: Review Microsoft’s (MSFT) quarterly earnings call transcripts or 10-K filings regarding capitalization of software development costs and data center infrastructure.

Microsoft has thousands to hundreds of thousands of AI chips not yet deployed due to bottlenecks.

Unidentified speaker ·

How to verify: Check Microsoft’s investor presentations or earnings calls for commentary on Azure capacity, GPU inventory, and deployment timelines.

Oracle is rated 'cautious to negative' by the speaker due to high debt levels.

Unidentified speaker ·

How to verify: Verify Dan Ives’ current rating on Oracle (ORCL) via his firm’s (Goldman Sachs) research notes or public statements.

Elon Musk stated there might be only seven to ten 'neo-clouds' left in the market.

Unidentified speaker ·

How to verify: Search for Elon Musk’s recent tweets or public statements regarding the 'neo-cloud' market size and consolidation.

OpenAI raised $122 billion in equity this year.

Unidentified speaker ·

How to verify: Review financial news reports or SEC filings (if applicable) regarding OpenAI’s funding rounds in 2026.

Anthropic has raised $95 billion in equity.

Unidentified speaker ·

How to verify: Review financial news reports or SEC filings regarding Anthropic’s funding rounds in 2026.

SpaceX raised $75 billion when it went public.

Unidentified speaker ·

How to verify: Check SpaceX’s IPO prospectus or financial news archives for its 2026 public offering details.

Alphabet raised $85 billion in equity.

Unidentified speaker ·

How to verify: Review Alphabet’s (GOOGL) equity issuance announcements or SEC filings for 2026.

77% of global token usage comes from China, but they only spend 22% of the total spend.

Unidentified speaker ·

How to verify: Locate the cited Goldman Sachs report or similar industry analysis on global AI token consumption and spending by region.

NVIDIA trades at 11-12 times earnings, or roughly 9 times 'real numbers'.

Unidentified speaker ·

How to verify: Check NVIDIA’s (NVDA) current P/E ratio and forward earnings estimates from financial data providers like Bloomberg or Yahoo Finance.

Do your own research

  • Verify Microsoft's (MSFT) actual capital expenditure schedules and accounting policies regarding the amortization of AI infrastructure costs in their latest 10-K or 10-Q filings.
  • Check the latest quarterly earnings reports for Alphabet (GOOGL) to assess trends in digital advertising revenue and the adoption metrics of Gemini for Workspace.
  • Review Oracle's (ORCL) balance sheet and debt covenants to evaluate the 'canary in the coal mine' risk regarding high leverage in AI infrastructure plays.
  • Research the latest data on enterprise AI adoption rates from independent market research firms (e.g., Gartner, IDC) to verify the '3.5% Fortune 500 adoption' claim.
  • Analyze the financial disclosures and fundraising rounds for Anthropic and OpenAI to understand the valuation implications of the cited $95 billion and $122 billion raises.
  • Investigate China's AI policy and open-source model distribution to assess the competitive threat to US-based frontier model providers.
  • Review the prospectus and financial metrics for any pending AI-related IPOs, such as Anthropic, to evaluate revenue growth deceleration risks.
  • Compare current P/E ratios and free cash flow yields of major AI tech stocks against historical dot-com era multiples to validate the 'reasonable multiples' assertion.
Full transcriptexpand
0:00
Unidentified speaker: How do you position capital when geopolitics is permanently repricing risk?
0:04
Unidentified speaker: Is AI structural or cyclical?
0:06
Unidentified speaker: Or does it even matter?
0:07
Unidentified speaker: And with MA accelerating, where's the value actually being created?
0:11
Unidentified speaker: Strategic Alternatives, the RBC Capital Markets Podcast, gives investors and deal makers the insight to find the answers and stay ahead of what's reshaping business and markets.
0:20
Unidentified speaker: Listen and subscribe to Strategic Alternatives on Apple, Spotify, or wherever you get your podcasts.
0:25
Unidentified speaker: iConnections is the largest membership-only platform for the alternate investment industry, bringing together thousands of fund managers and allocators in one powerful hub.
0:33
Unidentified speaker: iConnections reimagines how the industry connects through its digital offerings and premier in-person events, empowering allocators and managers to meet, build relationships, and do business anytime, anywhere.
0:42
Unidentified speaker: To explore more about iConnections, its events, and gain access to the members-only platform, visit iConnections.io.
0:50
Unidentified speaker: Welcome to the OK Computer Podcast.
0:52
Unidentified speaker: I'm Dan Nathan.
0:53
Unidentified speaker: This is my very good friend Dan Ives partner at Yorkville Ives, a new firm.
0:58
Unidentified speaker: Dan, you and I have a ton to talk about.
1:00
Unidentified speaker: I definitely want to hit your firm a second, but we're recording this Monday morning right after the opening, and we're seeing a little bit of a follow-through from Friday's action.
1:10
Unidentified speaker: Friday's action, man.
1:11
Unidentified speaker: You had the NASDAQ down what 3.5%?
1:14
Unidentified speaker: You had the stocks down 6.5%, something like that.
1:17
Unidentified speaker: A lot of the Mag 7s were getting killed.
1:20
Unidentified speaker: Follow through from what we saw to Meta and Google, the kind of weakness there.
1:25
Unidentified speaker: Want to talk about who's winning, who's losing in the AI sort of trade.
1:29
Unidentified speaker: We want to talk about the funding, the cap raising, anthropic.
1:32
Unidentified speaker: Is it going to come?
1:33
Unidentified speaker: What should investors be doing?
1:35
Unidentified speaker: I keep getting asked that question.
1:36
Unidentified speaker: You're getting asked that question.
1:38
Unidentified speaker: Mine is more from retail investors.
1:40
Unidentified speaker: Yours is obviously interacting with a lot of hedge funds, mutual funds, institutions in general.
1:45
Unidentified speaker: And I'd love to get the vibe there and a whole host of other things.
1:49
Unidentified speaker: Let's talk about is the SaaSpockLix back, that sort of thing.
1:52
Unidentified speaker: So thank you for being here again.
1:54
Unidentified speaker: I think this is probably the 10th time you've been on the pod over the last few years, and we really appreciate your love it.
1:59
Unidentified speaker: Because you also were able to sort of glean out and kind of dig deeper into key topics that I think so many people want to understand.
2:07
Unidentified speaker: I think our listeners and our viewers have gotten very used to the fact that you do not mind being challenged in your very bullish view.
2:15
Unidentified speaker: And really a lot of it comes down to time horizons.
2:17
Unidentified speaker: We're going to give you the opportunity to kind of talk about with what's going on right now relative to how you see it playing out over two, three, five years.
2:25
Unidentified speaker: You are taking the long view.
2:26
Unidentified speaker: You have said consistently, whether it's on CNBC or on podcasts, that sort of thing, that we are in the first inning, right?
2:33
Unidentified speaker: Of this whole thing.
2:33
Unidentified speaker: So first things first, let's talk about your firm, my man.
2:36
Unidentified speaker: Yeah.
2:36
Unidentified speaker: So it's been great.
2:38
Unidentified speaker: What are you doing though?
2:38
Unidentified speaker: Because you were at Wedbush.
2:39
Unidentified speaker: You had an ETF there, the Wedbush Ives ETF.
2:42
Unidentified speaker: We want to know if that's still going, but what's going on at the new firm?
2:45
Unidentified speaker: Yeah, and that Ives ETF, we have a relationship and we put out press release with Wedbush.
2:50
Unidentified speaker: So I'm still 100% associated with it.
2:53
Unidentified speaker: It's really nothing changes.
2:54
Unidentified speaker: The only difference is the research comes from Yorkville Ives instead of come from Wedbush.
3:00
Unidentified speaker: Yes, we've hired about 25 people now at Yorkville Ives.
3:03
Unidentified speaker: And we call it modern merchant bank, but basically it's a new investment bank with a hedge fund that's long standing in terms of Yorkville.
3:12
Unidentified speaker: And we're focused on tech, energy, infrastructure.
3:16
Unidentified speaker: That the whole reason we built it is my view of wanting to be like a white glove service, not just for companies, but for investors institutionally to do more things around ETFs and other products.
3:31
Unidentified speaker: Because I think now is that time.
3:33
Unidentified speaker: Obviously, in the disruption that we're seeing, I think opportunities.
3:37
Unidentified speaker: We've had 25 people we've hired, super excited about how we're building out an equity research platform, which is what you've done.
3:44
Unidentified speaker: So the investing is both public and private markets.
3:47
Unidentified speaker: Yeah, and then on the private, we launched uh called three, four weeks ago, an IPO, like the I the Ivy AI.
3:53
Unidentified speaker: That's the Ives Ultra Fund, which is really it's the first tech, really the first public company to invest in private AI tech companies.
4:03
Unidentified speaker: We launched that, and that's super exciting because it's $200 million IPO, but it speaks, in my view, of a democratization that's happening where I want to give institutional or retail investors the opportunity to invest in private AI tech companies to a public vehicle.
4:21
Unidentified speaker: Let's actually start there.
4:22
Unidentified speaker: That's a great way to do it because I go back to the 90s and I know that you were had a front row seat for this kind of dot-com.
4:28
Unidentified speaker: It ended up being a bubble the whole way in the 90s.
4:31
Unidentified speaker: Great returns, not too different, actually, if you think about it.
4:33
Unidentified speaker: I think from 95 to 99, the SP averaged like 20 plus percent gains like year over year.
4:40
Unidentified speaker: We've been in that mode since 2022 where we had this kind of bear market.
4:45
Unidentified speaker: It wasn't kind of it was a bear market, but 23, 24, 25, here we are in 26, and maybe it might be get tough to get above those 20%.
4:53
Unidentified speaker: So we saw that.
4:54
Unidentified speaker: You know all the comparisons to dot com, but a lot of those companies were like public.
4:59
Unidentified speaker: Yahoo is public after a year and a half.
5:01
Unidentified speaker: So right now we're seeing these companies private for longer, we're seeing these mega IPOs.
5:07
Unidentified speaker: You're I assume investing in some of the stuff that you see in the application layer.
5:12
Unidentified speaker: So give me a sense for that because that is very different.
5:15
Unidentified speaker: VC was much smaller in the 90s than it is right now.
5:17
Unidentified speaker: No doubt.
5:18
Unidentified speaker: But also, it's like, look, I learned just like you did, learn being an analyst during the bubble and burst.
5:24
Unidentified speaker: Okay, going back in hindsight, like what were the signs comparing relative to today?
5:30
Unidentified speaker: Are there things that remind you of it?
5:32
Unidentified speaker: And like I continue to say, like, this is the first time, like where I view it like the dot-com sort of bubble moment was almost the beginning of I'll call like a 30, 40 year cycle, whether it's laying the fiber, laying the foundation to actually where we are today.
5:49
Unidentified speaker: Because the the use cases for AI are real.
5:53
Unidentified speaker: It's like this is being funded by big tech, trillion dollars in the balance sheet, three to four hundred billion a year free cash flow.
5:59
Unidentified speaker: Across all the major and I just truly think it called 20 times earnings.
6:05
Unidentified speaker: If you go back to March 2000, the average tech company was 32 times revenues, profitability wasn't there.
6:11
Unidentified speaker: So, of course, there's like always gonna be like parallels and people would try to draw conclusions.
6:16
Unidentified speaker: But I view it just apples to oranges, just given where I believe this is truly year three of an eight to 10 year, potentially damn 12-year build out of the AI revolution.
6:30
Unidentified speaker: That's that and some like myself has spent so much time, like we were in Korea a few weeks ago, whether it's on the chip side, on memories, understanding where their demand is.
6:39
Unidentified speaker: Last week I was with 80 customers, these are all Fortune 500 companies in terms of who are going down the AI path when it comes to the city.
6:54
Unidentified speaker: We think it's about 3.5%.
6:56
Unidentified speaker: Really?
6:56
Unidentified speaker: Okay, and so to you that's bullish, right?
6:58
Unidentified speaker: So to me, I'm not saying that we're not gonna go through just like we are in a white knuckle periods, almost between the capex and the build-out to the monetization, as an investor, and this is a big issue for institutional, specifically hedge funds, in that air pocket period, just like you saw this week with the ServiceNow or AlphaBet, whatever, you can't just be like longer term, three to five years.
7:25
Unidentified speaker: No, because you could lose your job if you're not positioned right.
7:29
Unidentified speaker: So I think it's what makes this such a complex environment to navigate and not to necessarily get scared, but to understand who are the winners.
7:40
Unidentified speaker: What are the names you buy on dips?
7:43
Unidentified speaker: What are the names that you stay away?
7:45
Unidentified speaker: So I think that's what we've been trying to do with investors.
7:48
Unidentified speaker: Okay, we've seen change of fortunes, right?
7:50
Unidentified speaker: If you think about the hyperscalers were some of the early beneficiaries, obviously NVIDIA, a handful of other chip companies, but then we saw the trade broaden out a bit.
7:59
Unidentified speaker: And it's interesting that you guys are covering energy because I think your view is that is one of the huge bottlenecks and have the continued growth there.
8:06
Unidentified speaker: And I think that makes sense to figure out other ways to express views, that sort of thing.
8:10
Unidentified speaker: I know that you've been picking and choosing within the software space because there will be some clear winners.
8:15
Unidentified speaker: We know the SAS pocalyp was baby with the bathwater.
8:18
Unidentified speaker: How do you figure out?
8:19
Unidentified speaker: Let's just talk about Microsoft and Meta, for instance, and without getting into their fundamentals, but the stocks have had each one of them three.
8:26
Unidentified speaker: Microsoft was down 35% from its 52-week highs as it's going back a few months ago.
8:30
Unidentified speaker: And then it rallied, I don't know, 40-50%.
8:33
Unidentified speaker: And now it's come in just a very little bit.
8:35
Unidentified speaker: It shows great relative strength.
8:36
Unidentified speaker: Meta is a great example, is that it was down 30 some percent.
8:39
Unidentified speaker: This is going back really in late summer.
8:42
Unidentified speaker: They introduced Muse, so it's getting punished for the capex, right?
8:46
Unidentified speaker: And then you found a way in which the folks thought that it were gonna monetize it, but then they announced their spending and now it comes back, it gives up most of the move that it's had.
8:54
Unidentified speaker: How do you navigate that?
8:56
Unidentified speaker: Because investors are getting turned around.
8:58
Unidentified speaker: And is it is it it seems harder than it looks to pick the ones that are gonna be successful long term?
9:04
Unidentified speaker: Because we will see changing of fortunes.
9:07
Unidentified speaker: Look, Microsoft's a good example.
9:08
Unidentified speaker: See, for Microsoft, all they do is enterprise.
9:13
Unidentified speaker: So when you think it would be no different than you can say Palantir government as well, but like when it comes to AI, it's not like there's another piece of the business that's getting cannibalized because of AI.
9:27
Unidentified speaker: So for Microsoft on the enterprise, my view is just a very simple one.
9:31
Unidentified speaker: It's like we talk about 3.5, 4%.
9:33
Unidentified speaker: As more enterprises move to AI with Azure, cloud growth is significantly underestimated.
9:41
Unidentified speaker: It was up 50%.
9:42
Unidentified speaker: Yeah, and I think if I look at street numbers, probably underestimated 15, 20%, next 18, 20, like 24 months, that that will ultimately continue to come through and the stocks gonna go higher because under some of the parts, we talk about six, six fifty, seven hundred hours.
10:01
Unidentified speaker: You could rationalize that pretty easily as these numbers come through.
10:05
Unidentified speaker: And sticking with Microsoft for a second, so you have one component which is Azure, the other component is the uptake of Office 365 co-pilot on that.
10:14
Unidentified speaker: Are they dependent on each other?
10:16
Unidentified speaker: Because obviously Azure is they want to capture more and more customers, you know, leveraging off their compute.
10:23
Unidentified speaker: This is one of the reasons why they have these contracts with NeoClouds because they have their own, you know, pace in which they can build out their own compute, and then they go to a core weave or something like that.
10:34
Unidentified speaker: Is there a scenario where Azure starts to lose a little bit of market share?
10:38
Unidentified speaker: GCP grew 100% year over year in that last quarter.
10:41
Unidentified speaker: AWS is massively accelerated.
10:43
Unidentified speaker: I think it bottomed out in low teens.
10:45
Unidentified speaker: Now it got back up to 39% year over year growth.
10:48
Unidentified speaker: How important or how connected is Azure growth to 365 co-pilot growth?
10:54
Unidentified speaker: It's like hearts and it's interconnected.
10:57
Unidentified speaker: So that's why if you go back to like when you had the significant sell-off in Microsoft, a lot of it was like co-pilot, a lot of the frustration of customers.
11:06
Unidentified speaker: Could this be the albatross to some extent?
11:11
Unidentified speaker: They've been able to smooth a lot of that out.
11:13
Unidentified speaker: And at the end of the day, if you're a core Microsoft customer on the enterprise, this you I wouldn't say you don't have a choice, but the path of least resistance is to go with Microsoft.
11:27
Unidentified speaker: So when we view like where Google is and where Amazon, I could argue like all three hyperscalers could have significant growth over the next two years.
11:39
Unidentified speaker: Oracle is different because they're trying to build themselves into a hyperscaler.
11:43
Unidentified speaker: Just for one customer, though.
11:45
Unidentified speaker: For basically for 300 billion of the 600 is because of open AI.
11:50
Unidentified speaker: SpaceX is essentially building itself into hyperscaler.
11:55
Unidentified speaker: The ones that are best positioned are these three.
11:59
Unidentified speaker: Now, alphabet is is this is a separate issue because of the advertising piece.
12:05
Unidentified speaker: And this was obviously the first quarter where it's like, whoa, like two quarters in a row on search.
12:10
Unidentified speaker: It's true, but this quarter in particular, it's like, okay, is AI cannibalizing the future of the advertising, the golden goose?
12:19
Unidentified speaker: And Anthropic has said they will not put advertising on Claude, but OpenAI did start an advertising business.
12:26
Unidentified speaker: And when you think about it, my good friend Joe Marcese, who you know, he wrote a book.
12:30
Unidentified speaker: It's coming out next week, and he's gonna be on the podcast.
12:32
Unidentified speaker: We're gonna give a bunch of those away.
12:33
Unidentified speaker: You got to write a book, by the way.
12:35
Unidentified speaker: And he's got a couple really great in taglines in the description of the book.
12:40
Unidentified speaker: I just started reading it.
12:40
Unidentified speaker: And he's talking about basically the entire economy that we live in, the digital economy, has been built on digital ads and it's subsidizing everything else.
12:48
Unidentified speaker: And if you start to see these chatbots really do a number on digital ads, some of these business models that were built on them, it's not just Google, might have a difficult time over the next few years as this whole model is being challenged, and then investors are digesting this.
13:04
Unidentified speaker: No, look, but it speaks to for Google, as well positioned as they are in GCP, the advertising piece, it's like a race against the clock.
13:15
Unidentified speaker: Gemini, market share, they need to ultimately accelerate market share, but then also they have to almost try to figure out the way that they're gonna monetize on the advertising side relative to what we're seeing on a search query perspective.
13:33
Unidentified speaker: I think the street is modeling, like if you think about the decline that we've seen relative, that this is a sustainable thing.
13:41
Unidentified speaker: I disagree just relative to so many CMUs that we talked to, so many that rely on this.
13:47
Unidentified speaker: The view is that they're gonna, similar as Microsoft figured out Copilot, that they're gonna be able to figure out the monetization engine to ultimately see this thing.
13:59
Unidentified speaker: Well, as bad as that quarter was, and Sundar did talk about Gemini for Argon and Argon, Argon, and how weird names for all these things, man.
14:08
Unidentified speaker: It's like really hard.
14:09
Unidentified speaker: And the Chinese ones are weird too.
14:10
Unidentified speaker: But he's talking about this as really their first agentic product for the enterprise.
14:14
Unidentified speaker: And so the idea that this might be, if you think about what's going on with search, this might be the sort of next leg, despite, let's say, the cannibalization to some degree of using AI mode and that sort of thing.
14:26
Unidentified speaker: And I'm just sure if you were trying to look for silver linings in the Google quarter, was that it?
14:31
Unidentified speaker: And what else other than GCP growth could you look at that makes you feel less bad about the sell-off that we saw?
14:39
Unidentified speaker: If you look at the queries and you look at maybe like some of the me from a conversion data perspective, like I could argue over the coming quarters, this will be viewed as more of like an air pocket interim period rather than the start of some secular trend on the advertising business.
14:59
Unidentified speaker: Look, there a lot of times, like my view, and even on like the private side, is like you invest in very smart people that they figure it out.
15:10
Unidentified speaker: When it comes to like what we're seeing in Alphabet, I think now it's about them.
15:16
Unidentified speaker: They have the install base, they have the engine.
15:20
Unidentified speaker: Now it's them figuring out the monetization piece.
15:22
Unidentified speaker: I think that but if you look what's in the stock right now, investors are probably putting a 30 to 40 percent discount on the future advertising business of Google.
15:35
Unidentified speaker: My argument would be like, I'm not saying it doesn't get dinged, but if you think it's anywhere better than this worst case that's being factored in, this is a stock that has significant upside here.
15:47
Unidentified speaker: All right.
15:48
Unidentified speaker: So how do you deal?
15:49
Unidentified speaker: And I'm just gonna ping pong around a little bit, and so hopefully listeners will get some of the vibe here.
15:53
Unidentified speaker: So going back to Microsoft and being very differently positioned than Google.
15:58
Unidentified speaker: And Microsoft was there.
15:59
Unidentified speaker: Microsoft was one not to call you up, but Microsoft was one where it was like that from a thesis perspective was like a table pounder relative to like you could be wrong on a quarter, but the checks were such that when the stock was like 360, 370, you them is new outside of like some like you know, but how do investors deal with that volatility?
16:22
Unidentified speaker: So let's talk about obviously to some degree, institutions can manage risk, right?
16:27
Unidentified speaker: In a portfolio fashion.
16:28
Unidentified speaker: There's a lot of obviously they can use derivatives and that sort of thing.
16:32
Unidentified speaker: But as retail investors, self-directed investors, it's really hard.
16:36
Unidentified speaker: These stocks they go up on these bounce of spikes and then they get flushed out.
16:40
Unidentified speaker: And it just doesn't seem like we're at this period anymore where it's a nice like bottom left, upper right sort of thing.
16:47
Unidentified speaker: And it really comes down to stock picking.
16:49
Unidentified speaker: But like every quarter, it seems like it could go back and forth and that that chopped.
16:54
Unidentified speaker: And we're not just talking about 10%, we're talking about 25, 30 percent decline.
17:00
Unidentified speaker: And look at this quarter with the Intel and the 30% sell-off, and ServiceNow and the sell-off.
17:06
Unidentified speaker: So when you're selling a stock like Intel down 30%, the government invested at 20 bucks, NVIDIA invested at 25.
17:14
Unidentified speaker: This is going back to the summer of 25.
17:16
Unidentified speaker: These stocks were up hundreds of percent.
17:19
Unidentified speaker: And then they come down like this.
17:21
Unidentified speaker: It's that's a hard way to invest because a lot of folks are just buying it up because they're taking a long-term view.
17:27
Unidentified speaker: They're like, and then they're down 30%.
17:29
Unidentified speaker: That's a huge hit.
17:30
Unidentified speaker: So, how do you manage that risk if you are just out there, you believe in, let's say, some very bullish folks like on podcasts or CNBC or yourself, that sort of thing, and they don't have the ability to manage risk that way.
17:43
Unidentified speaker: First of all, I think it's the like Intel is like not a name that like I'm a huge supporter of, but the point is for me, it's it's trying to have the 20 to 30 stop.
17:52
Unidentified speaker: Let's call it the 20 to which is what your ETF is.
17:55
Unidentified speaker: Yeah, but across tack.
17:56
Unidentified speaker: Yeah.
17:57
Unidentified speaker: And be like, and some of those, there could be thesis changers that then you have to move, you gotta basically just move m move forward with.
18:07
Unidentified speaker: But a lot of them, it's ones where like when you navigate periods, whether it's Palantir, whether it's Salesforce, whether it's ServiceNow, whether it's Microsoft, whether it's NVIDIA, whether it's AMD.
18:21
Unidentified speaker: Again, to me, it's more like that's all the work that we do with why we travel so much.
18:27
Unidentified speaker: It's trying to have a thesis based on chip demand, what CIOs are telling us, what we're seeing in terms of sales and just demand relative to hyperscalers, what we're seeing on use cases, what's happened on cybersecurity.
18:43
Unidentified speaker: You're there's times in terms of navigating from a retail investor.
18:46
Unidentified speaker: It's just like but you ha if you just have a thesis over 60 days, 90 days in this market, I think it's it's a very dangerous It's a dangerous way to navigate because I think you have to view it as what your risk profile is, what are the names you love, and ultimately when they get just crushed to these levels, are they at points where could they be a downside?
19:15
Unidentified speaker: Could there be another 10-15% downside?
19:18
Unidentified speaker: Of course, risk reward is one that you own them.
19:21
Unidentified speaker: Yeah, and that's how we even in ServiceNow is a good example, just like we've done with Salesforce.
19:26
Unidentified speaker: It's like, do you think Service Now is broken?
19:31
Unidentified speaker: If you do, then you get out of it.
19:33
Unidentified speaker: We believe ServiceNow has a phenomenal customer base, salespeople and partners that believe in the strategy that that ultimately they're selling.
19:46
Unidentified speaker: And I just think they're violent air pockets, but I believe like when a stock like that sells off the way it does, especially given my view of McDermott, the leadership there, and the valuation, you own it.
19:59
Unidentified speaker: What other modes is the one question I would say.
20:01
Unidentified speaker: Is yeah, you keep hearing about switching costs, that sort of thing.
20:04
Unidentified speaker: But when I think about a service now, this narrative has been really hard because if you look at all these narratives, Salesforce, too.
20:11
Unidentified speaker: Yeah, Salesforce, all these narratives, they were playing really well for these SaaS names.
20:16
Unidentified speaker: This is in 2023, because you'd look at those companies and you'd say to yourself, you weren't even thinking about competition at that point.
20:23
Unidentified speaker: You were thinking about this technology should just really be a jet pack for what they're doing.
20:28
Unidentified speaker: But if anything, it started, they were slow to invest, it seemed like, right?
20:33
Unidentified speaker: And then all of a sudden, the big Frontier Labs, you could see as they were putting out product that this is going to be really competitive in many ways.
20:41
Unidentified speaker: And then when I think about that Salesforce quarter a month or two ago, if when they went to on TV with Benioff, if Dario was not sitting next to him, I think that stock would have been down a lot.
20:53
Unidentified speaker: But investors were like, oh, look at this partnership.
20:56
Unidentified speaker: Those are just commercial partnerships that go on all the time, right?
21:00
Unidentified speaker: So I just think that we're getting to a point where there's a lot of funny business.
21:03
Unidentified speaker: And if I'm Dario, I need to do enterprise deals, right?
21:07
Unidentified speaker: I need to like cover my apps.
21:10
Unidentified speaker: But at the end of the day, SaaS is tough.
21:12
Unidentified speaker: Okay.
21:13
Unidentified speaker: And it's one I'm curious as you think about when you're investing as a firm or the research that you're doing.
21:20
Unidentified speaker: Are you much more focused now on hyperscalers, component suppliers, the adjacency as it relates to software?
21:28
Unidentified speaker: Where are you most convicted at right now?
21:30
Unidentified speaker: Because we're seeing big corrections across a lot of these different kinds of verticals I just mentioned.
21:35
Unidentified speaker: I think it's in ships and hyperscalers where you that's been our sort of foundation keep our thesis from NVIDIA to Microsoft to Alphabet to AMD and so many others.
21:48
Unidentified speaker: But I actually think the biggest opportunities are now as the use cases get built.
21:55
Unidentified speaker: Who are the software companies, even some of the IT server like Accenture and others that you feel cybersecurity as you follow the sort of yellow brick road?
22:09
Unidentified speaker: Who are the ones that are gonna benefit that the stocks are not being priced like that?
22:13
Unidentified speaker: Now there, but there are moments like if I go back to like March and April, that was where I mean you talked about like cybersecurity stocks were just dude, I don't care, like oil, this 10-year FET, you just don't like it.
22:27
Unidentified speaker: And evaluations don't matter.
22:28
Unidentified speaker: They're trading at huge premiums to let's say the hyperscalers.
22:31
Unidentified speaker: But I'm saying back then, like it's oh no, no, yeah, and they were trading very poorly at the beginning of the state.
22:35
Unidentified speaker: We're gonna have I actually think like three, four times a year, we're gonna have these moments where you're like they get disc certain things get disconnected relative to fundamentally where I think they're gonna go.
22:51
Unidentified speaker: Cybersecurity be a perfect example, like budgets are going to potentially double in cybersecurity for IT over the next two, three years, just given surface area, agents, where everything's heading.
23:05
Unidentified speaker: So that's a good example.
23:06
Unidentified speaker: Like can I ask you a question?
23:07
Unidentified speaker: Sorry to cut you off there.
23:08
Unidentified speaker: So let's just assume you see a hugging face, you see some other incident.
23:12
Unidentified speaker: I saw some stat the other day that like five percent of let's say all of the hacks that have been reported over the last 10 years happened like last month or something like that.
23:24
Unidentified speaker: So you're seeing an acceleration by folks, bad actors using these tools.
23:28
Unidentified speaker: What if we keep seeing a big hugging face, like once a month, that sort of thing?
23:32
Unidentified speaker: Might investors be like, these are not the right companies to do this sort of thing.
23:36
Unidentified speaker: I'm just curious if that from a sentiment standpoint, because these stocks have acted very well, they traded huge multiples.
23:42
Unidentified speaker: And I'm just curious, like, how does the cyber story come undone, at least in the near term, because they overshot the way semis did it and the way, let's say, the some of the hyperscalers did it?
23:52
Unidentified speaker: Yeah.
23:52
Unidentified speaker: But I think if you look, you took three and a half percent like penetrations in terms of the cybersecurity software piece, it's less than one percent.
24:04
Unidentified speaker: So in terms of like what would change sentiment, I actually view like the opposite.
24:08
Unidentified speaker: I think when you have these moments of hacking, it increases the opportunity and the demand for the crowd strikes, for the palaltos, for the Z scours.
24:21
Unidentified speaker: I think part of what happened too is like let's say if you saw like Fortinet checkpoints are good examples that had like specifically Fortinet, like very softer quarters.
24:32
Unidentified speaker: Fortinet sits on more firewall side.
24:35
Unidentified speaker: So Fortinet last week was down like 25% or something.
24:39
Unidentified speaker: Yeah, and again, it's all relative to like the stocks had like a huge run.
24:43
Unidentified speaker: But I don't put Fortinet in the same category as CrowdStrike or Pow Auto, but it's a good example, like what's essentially happening, and this is like an overall you know, covering this space for whatever, almost 30 years, like software salespeople, partners, customers, I don't care like where in the food chain, the level of disruption that's happening across the IT spending cycle is unlike anything we've ever seen.
25:17
Unidentified speaker: What I mean is you could be software company A, and you look at your pipeline, there's typical conversion of 70% based on pipeline, and you need certain coverage ratios, and all of a sudden deals are getting taken out of the pipeline that you thought were gonna convert because they're not AI the or they're disrupted by other money that's going toward AI-driven projects, and maybe you had a document recovery product that's not associated with it.
25:50
Unidentified speaker: So I think that also if you see with Lip and Intel, or you see some of these conference calls, some of these CEOs too, they almost seem lost to some extent in terms of trying to understand because I do think this is it's a transformation that's happened across the industry because of what's happened with AI.
26:16
Unidentified speaker: And I think it's it complicates a lot of things that we're seeing during earnings season.
26:22
Unidentified speaker: Yeah, it's a tough one because I can't look at too many names that have reported, let's just call it calendar Q3 in the last couple weeks where the stocks traded well afterwards.
26:33
Unidentified speaker: I literally can't think of one.
26:34
Unidentified speaker: And when I think of like an Intel was down 50%, I think of the stocks in general is down 30%.
26:40
Unidentified speaker: But also to the Intel as an example, like if you're an institutional investor in the last three months, everyone knew his memory prices, what's happened on PC, CPU, and then all of a sudden, like management or everyone feels like they're they're very comfortable managing the cycles through this, they've given guidance accordingly.
27:02
Unidentified speaker: The intel was a huge gut punch to I think just to so many investors.
27:08
Unidentified speaker: And I think what you've seen during this earnings season, going back to what complicates it, is how do you own the stocks during these periods of volatility where the level of disruption is one where you feel like a lot of times like the Madgen teams are almost like the last to know about it.
27:33
Unidentified speaker: Markets are moving faster and decisions are getting harder.
27:36
Unidentified speaker: But the companies that win aren't waiting for clarity.
27:38
Unidentified speaker: They're having strategic conversations early.
27:40
Unidentified speaker: RBC Capital Markets brings the insight, advice, and execution to help clients find paths to act with confidence when it matters most.
27:47
Unidentified speaker: From capital raises and IPOs to transformative MA, RBC's global investment banking team advises the most ambitious companies through their most defining decisions across industries, capital structures, and geographies.
27:59
Unidentified speaker: They know where capital is heading next because they're helping put it to work.
28:03
Unidentified speaker: Your capital needs a strategy.
28:04
Unidentified speaker: RBC Capital Markets Insights deliver results.
28:08
Unidentified speaker: iConnections is the largest membership-only platform for the alternate investment industry, bringing together thousands of fund managers and allocators in one powerful hub.
28:15
Unidentified speaker: iConnections reimagines how the industry connects through its digital offerings and premier in-person events, empowering allocators and managers to meet, build relationships, and do business anytime, anywhere.
28:25
Unidentified speaker: To explore more about iConnections, its events, and gain access to the members-only platform, visit iConnections.io.
28:48
Unidentified speaker: I think it might have been Ed Zitron, who's also a noted bear.
28:51
Unidentified speaker: And listen, anybody who's listening to this, discount that you could say they're perma bears, and maybe they don't get the technology.
28:56
Unidentified speaker: They're more fun, like Jim is more of a finance guy, you know, that sort of thing.
29:00
Unidentified speaker: And maybe Ed was never a technologist.
29:03
Unidentified speaker: Gary Marcus definitely is.
29:05
Unidentified speaker: And like here you are on the other side of all of that.
29:08
Unidentified speaker: Bears are called ten of the last two downturns.
29:10
Unidentified speaker: And that's fine.
29:11
Unidentified speaker: But some of this stuff, because it's so technical and because it's meant to be so widespread, I think it's worth to try to, if you are a massive bull, but you need to understand, like Chino's all those guys, you need to understand the bear thesis.
29:26
Unidentified speaker: And also, so many of my good friends or people I've known are are super bearish.
29:30
Unidentified speaker: Yeah.
29:31
Unidentified speaker: And I actually think to some extent, what makes you a better analyst, or you have to understand those things.
29:40
Unidentified speaker: Because if you just go like this and don't listen, then and but now, but I do think, especially in social media and everything that we see now, a lot of times narratives could easily form that create sometimes an overhang in a stock that at least I view relative to the work that we do in the field is maybe fictional.
30:03
Unidentified speaker: But that creates opportunities, and that's why this bull's embarrassed.
30:06
Unidentified speaker: All right, so let's talk about this really quickly.
30:08
Unidentified speaker: So is this whole idea a bit of underappreciated GPUs?
30:11
Unidentified speaker: No one seems to be able to put their finger on how to do that.
30:14
Unidentified speaker: But Jim's bringing up the point where you have Microsoft, let's just say, buys GPUs from NVIDIA's.
30:19
Unidentified speaker: NVIDIA books the profit, but Microsoft spreads the cost over, let's call it six years, right?
30:25
Unidentified speaker: So it's from an earnings standpoint, great for NVIDIA, right?
30:29
Unidentified speaker: And Microsoft gets to make it look like they made this purchase as it relates to CapEx, but they are basically spreading that amount over, let's say, five or six years.
30:39
Unidentified speaker: So it's a double benefit to S P 500 earnings if you think about it.
30:44
Unidentified speaker: And so then there's also this idea that Microsoft's got thousands, maybe hundreds of thousands, these chips that have not been deployed because of the bottlenecks that we're talking about, because of the moratorium on some data centers being built, the access to energy, memory prices being so high.
31:01
Unidentified speaker: I could continue to go on and on.
31:03
Unidentified speaker: How do you does the rubber have to hit the road at some point for kind of reconciling the difference in which these revenues are being accounted for?
31:12
Unidentified speaker: No doubt.
31:13
Unidentified speaker: And I think that's Amy Hood, Microsoft CFO.
31:15
Unidentified speaker: I think she's done a great job of maybe explaining some of that and navigating it.
31:21
Unidentified speaker: Book is as we get into the next six to nine months, the rubber does meet the road.
31:26
Unidentified speaker: I like when you say to me the biggest risk to the AI revolution, a lot of it, there's the political, right?
31:34
Unidentified speaker: Relative to midterms, you have a blue wave.
31:37
Unidentified speaker: What that does relative to does it slow it down, put sand in the gears to energy shortages because you need to you need the data centers, you need to deploy the GPUs, but you need energy.
31:51
Unidentified speaker: And the third thing is like the gap between catbacks and build outs to monetization and the market screaming it needs to be shortened.
32:05
Unidentified speaker: And that's why until it's shortened, credit spreads will blow out depending on when the bond market is.
32:19
Unidentified speaker: Okay.
32:19
Unidentified speaker: But really, the two that people have been tracking are Oracle and Core Weed.
32:23
Unidentified speaker: And it's freaking people out.
32:26
Unidentified speaker: And those stocks really have not been able to catch a bid.
32:29
Unidentified speaker: They'll have little counter trend short rallies, that sort of thing, but they come back pretty quickly.
32:34
Unidentified speaker: And it seems like investors are really starting to differentiate between those that are so dependent on debt to build out, and those that basically can use their their cash flow, go negative, right?
32:46
Unidentified speaker: But they can pull back and at some point get that earnings machine covered.
32:49
Unidentified speaker: And Oracle, we've been cautious to negative ish on, which is rare for because of that.
32:56
Unidentified speaker: See, I separate or distinguish an Oracle from a Microsoft or Met because the level of debt that they're taking on.
33:06
Unidentified speaker: And also I think the other issue is relative to cutting into the muscle or the bones of Oracle.
33:13
Unidentified speaker: I think that that's been a that's been a big issue.
33:16
Unidentified speaker: It's really fascinating with the Oracle.
33:18
Unidentified speaker: At some point, let's just say we do have a sustained sell-off, let's say we do have a bear market.
33:22
Unidentified speaker: Oracle will be so obvious to be the canary and the coal mine.
33:26
Unidentified speaker: And then if you think of some of these neo clouds, and getting back to what you said with the musk.
33:31
Unidentified speaker: Yeah, all right, let's go there for a second.
33:32
Unidentified speaker: So, SpaceX, you just said that becoming a Neo Cloud, they did these two huge deals.
33:36
Unidentified speaker: 40B.
33:37
Unidentified speaker: Yeah, and one was with Anthropic, one was with Google.
33:40
Unidentified speaker: We know we have this frenemy situation, that sort of thing.
33:42
Unidentified speaker: They're competing on the models, they're competing on compute and all that sort of stuff.
33:46
Unidentified speaker: Elon has said two things over the last week.
33:49
Unidentified speaker: He said basically, what did he say?
33:51
Unidentified speaker: Something about guppies, the neo cloud, whatever it was, cloud remote.
33:54
Unidentified speaker: And he's basically saying there might be seven to ten left, but there's tons out there.
33:58
Unidentified speaker: Okay.
33:59
Unidentified speaker: So that would be there's a lot of ways to think about that.
34:02
Unidentified speaker: It might be great for the hyperscalers, but if those go away, it means that their compute that they've built out is not necessary.
34:09
Unidentified speaker: And then you have a huge glut in that part of the market, which will not be, it will weigh on valuations of the hyperscalers.
34:17
Unidentified speaker: It might put them in a better position longer term to stop up that excess compute once theirs is built out.
34:22
Unidentified speaker: Okay, that's one.
34:23
Unidentified speaker: The other thing that Elon said, this is over the last couple of days, is said it in a tweet.
34:28
Unidentified speaker: Someone tweeted at him about their user of Grok, who's talking about now it's going to, and this is the end of token maxing.
34:34
Unidentified speaker: We know that.
34:35
Unidentified speaker: And that's been, and he was basically saying the cost to run Grok basically was unsustainable, right?
34:41
Unidentified speaker: And so now they're going into a pricing model.
34:43
Unidentified speaker: This is gonna be something with Muse that Meta is gonna have to figure out over the next few months or so.
34:49
Unidentified speaker: So, how do you respond to that when someone like Elon or when you see the other Frontier labs really with the frontier models, not Grok, how do you see them going from token maxing to the way they're doing now in consumption?
35:01
Unidentified speaker: Yeah, and they're looking it's a natural transition, like what they're gonna have to do.
35:04
Unidentified speaker: The one thing I'd say with Musk, remember, Neo clouds are competitive to some extent to what SpaceX is doing.
35:11
Unidentified speaker: And then it comes down to NeoClouds, okay, could you you'll have Core Weave, Nebias, other there will be a lot of consolidation in that market.
35:18
Unidentified speaker: But look, I in terms of like the price of tokens will continue to come down.
35:24
Unidentified speaker: It's our view, models will be become more and more commoditized.
35:29
Unidentified speaker: But you will the rubber doesn't hit the road, meet the road, probably for another few years relative to like some of those theses that Musk is talking about.
35:40
Unidentified speaker: But it's a real issue that I think the market's facing relative to okay, the compute, the power, the capacity, not everyone's gonna be a winner.
35:52
Unidentified speaker: I think everyone's trying to figure out in the game of musical chairs who's left out.
35:58
Unidentified speaker: Yeah, and so when I think about an anthropic though, again, they're trying to come out, IPO, supposedly in the next couple weeks.
36:04
Unidentified speaker: I'd be shocked in this environment if they can bring an IPO, that might be honestly.
36:08
Unidentified speaker: See, I would be that's one I'd be shocked if it doesn't happen.
36:14
Unidentified speaker: Right the the one thing I'd say is that because so many of these companies have been private and there's such good price discovery in the private markets.
36:25
Unidentified speaker: I I do think like the appetite for anthropic for an IPO, I still think unless there's something anomalous happens, I'd be shocked if it didn't happen.
36:35
Unidentified speaker: I think that Q3 that we haven't seen yet, as far as their revenue growth, which is decelerating, right?
36:40
Unidentified speaker: They've talked about building out their own compute, like their own data centers.
36:45
Unidentified speaker: To me, that just seems like one of the ones where it's just wait, you're just waiting for that.
36:50
Unidentified speaker: Is if they start that commitment, it's one thing to dull out big contracts to, let's say, XAI or Oracle.
36:57
Unidentified speaker: It's another thing when you're actually competing with them, especially if your contracts are not long term.
37:02
Unidentified speaker: And the ones with XAI are not long term.
37:05
Unidentified speaker: So a lot of these companies on the NeoCloud side are trying to take price.
37:09
Unidentified speaker: Yeah, to me, it just has the ability, in my opinion, to be a big mess.
37:14
Unidentified speaker: And I think the best thing in this environment, if there's not data coming out of the anthropic updated S1 with the Q3, and it doesn't look particularly good on a sequential basis, I think it could be a disaster for the markets.
37:28
Unidentified speaker: I really do.
37:28
Unidentified speaker: And so then I think about this, and I just want to read these numbers because I I put them all down.
37:33
Unidentified speaker: Okay.
37:34
Unidentified speaker: So this year, in equity fundraises, and again, we're just we have two months left in the year.
37:39
Unidentified speaker: OpenAI has raised 122 billion, anthropic has raised already 95, and they're gonna be obviously doing a bunch more.
37:48
Unidentified speaker: SpaceX, 75 billion when it went public, Alphabet 85 billion.
37:53
Unidentified speaker: This is just equity.
37:54
Unidentified speaker: Oracle 25 billion, XAI 20 billion.
37:57
Unidentified speaker: That was prior to, let's say, the IPO, and then Meta has this at the money, and they've already raised a lot.
38:03
Unidentified speaker: And so now I want to go.
38:05
Unidentified speaker: So this is private labs was 237, IPOs were 75 billion, Alphabet has been out there.
38:12
Unidentified speaker: They raised 85 billion, didn't they, on that deal?
38:15
Unidentified speaker: Yeah, and then Oracle 20 billion.
38:17
Unidentified speaker: Okay, so now on the debt side, Amazon 54 billion, that's continues to raise.
38:22
Unidentified speaker: Alphabet, they're in the market again.
38:24
Unidentified speaker: They have to go overseas to raise this debt.
38:26
Unidentified speaker: They've already done like about 30, that was in Q3.
38:29
Unidentified speaker: Oracle, 25 billion debt, Meta, 25 billion in debt.
38:33
Unidentified speaker: Nvidia, SpaceX, and Amazon, 75 billion combined.
38:37
Unidentified speaker: Okay, so when I think about this, okay, bond issuance about 400 billion.
38:41
Unidentified speaker: So we're talking between equity and debt.
38:43
Unidentified speaker: We're talking about almost a little more than 800 billion.
38:45
Unidentified speaker: So anthropic's coming, OpenAI is meant to come.
38:48
Unidentified speaker: If everyone in their mother is selling all of these companies, equity and debt, this goes back to a question: why should I buy anthropic?
38:56
Unidentified speaker: Why should I buy these stocks that have negative free cash flow?
39:00
Unidentified speaker: They're spending obviously a shit ton, and they don't know what the demand's gonna be.
39:05
Unidentified speaker: Anybody can look at this technology.
39:07
Unidentified speaker: I love the technology, both personally and professionally.
39:11
Unidentified speaker: Okay, but if I'm an investor in these stocks, I have no idea when I get the payoff.
39:16
Unidentified speaker: Look, you have to, and it's great.
39:19
Unidentified speaker: Look, I think it's a great sort of to lay it out there, and I think it's the big debate right now.
39:25
Unidentified speaker: I think you have to believe three things in order to buy these stocks.
39:28
Unidentified speaker: One, you have to believe that the demand opportunity when you look at the next three, four years, that eventually the free cash flow in 20, 29, 30.
39:40
Unidentified speaker: So you have to believe that.
39:42
Unidentified speaker: Two, you have to believe that these companies' ability to raise that money from an investor perspective, once they puts a floor, but makes you realize that okay, they're gonna now have the capacity to do what they need to do in this AI arms race.
39:59
Unidentified speaker: They're gonna have the funds to build it out.
40:01
Unidentified speaker: Because, like I've said, those that don't have the funds, they won't be able to be what I view as a tier A.
40:08
Unidentified speaker: What about OpenAI though?
40:09
Unidentified speaker: They were trying to raise 30 billion at 1.4 trillion.
40:13
Unidentified speaker: This is going back over the last few weeks.
40:15
Unidentified speaker: The fact that it's taken over a month and that comes out over the weekend, that they're only going to be able to raise 20 billion at 1.2 billion.
40:23
Unidentified speaker: So in that environment, and again, if you tell me that anthropics numbers look so much better, how do you, how does open AI come out if people, you know what I mean?
40:32
Unidentified speaker: There's been insatiable demand.
40:34
Unidentified speaker: And if that's cooling a little bit, it seems like that is not a great thing for the anthropic IPO because they're targeting what, one and a half trillion to maybe 175 or something like that.
40:45
Unidentified speaker: And I think one, it's more one five.
40:47
Unidentified speaker: You think so?
40:47
Unidentified speaker: Okay.
40:48
Unidentified speaker: So when I think about that, again, I have folks, I have friends asking me, should I buy anthropic here?
40:55
Unidentified speaker: They were asking, should I buy SpaceX?
40:58
Unidentified speaker: If you bought SpaceX, you were paying up, let's say 20% over the IPO price.
41:03
Unidentified speaker: It went up a lot more over a week, but now it's back at its IPO price.
41:08
Unidentified speaker: So again, you could say, well, that's just been three and a half months.
41:10
Unidentified speaker: So I'm I'm just trying to get my arms around this because my answer is it's not a recommendation, but I'm not buying them here.
41:17
Unidentified speaker: You know what I mean?
41:17
Unidentified speaker: Like that sort of thing.
41:18
Unidentified speaker: And again, I'd rather buy a Google down 25%.
41:22
Unidentified speaker: No, but that's how but that goes to the city.
41:24
Unidentified speaker: Back to what you started out by saying.
41:25
Unidentified speaker: Yeah.
41:25
Unidentified speaker: So to me, you don't you you don't just own them all, but it it's more when the stocks sell off to levels, those are the ones where I call them like whatever tail pounds, you have to own them.
41:38
Unidentified speaker: But it's no different than like if I think about Apple, it was when the stocks sold off to 220.
41:45
Unidentified speaker: You're like, you even if they don't even get AI right, you own okay.
41:49
Unidentified speaker: Net stocks 340.
41:51
Unidentified speaker: The point is, I think there's opportunities going back to Microsoft 360, 370, going back to Palantir when it almost went sub 100.
42:00
Unidentified speaker: The point is you can't just be like, own it.
42:03
Unidentified speaker: But it's like what if you have if your thesis is intact, you use those sell-offs.
42:09
Unidentified speaker: Service now is a good example where it don't.
42:13
Unidentified speaker: And I'm not saying stock doubles, you could take some profits, that's fine.
42:17
Unidentified speaker: But I'm saying to me, that's the way.
42:19
Unidentified speaker: All right, so that's a great point.
42:20
Unidentified speaker: So in your ETF and the way you think about the way you write your research, I think anybody who reads you, listens to you, they know that you have a long-term view.
42:28
Unidentified speaker: But but it's picking, like anyone that that follows my stuff knows like SAS Apocalypse came up.
42:36
Unidentified speaker: You have like this is just I'm not saying there's not gonna be headwinds, Adobe into whatever, but this is like a generational near-term opportunity.
42:45
Unidentified speaker: But stock picking.
42:46
Unidentified speaker: Stock picking stock picking I think has become and will so important in this market because the two names could be drastically different position.
43:05
Unidentified speaker: And if they are, like you have to own the one that's right.
43:08
Unidentified speaker: Let's talk about commoditization.
43:10
Unidentified speaker: I think on the model front, this is a big narrative and it's talked about that.
43:14
Unidentified speaker: Yeah, this is one thing that I think if you're somewhat cautious about evaluation of an open AI or an anthropic here, these these two companies are literally at the epicenter of that sort of narrative that could be most adversely affected.
43:30
Unidentified speaker: So if you think about this, and I know you spend a lot of time thinking about Asia, China, and I read this stat, oh, it's from a Goldman Sachs report that 77%, 77% of the global token usage is coming from China, but they only spend 22% of the spend.
43:52
Unidentified speaker: Okay.
43:52
Unidentified speaker: So when you think about that and all the open source models, and a lot of our, whether it's co-pilot or a lot of the, they are offering Chinese open source models.
44:03
Unidentified speaker: And so does that have the ability to kill a bunch of the investment thesis here in the US?
44:10
Unidentified speaker: And when you think about China and Louis Vincent Gobb, who's been on the podcast and he's over in Hong Kong, spends all he lives there or you know, moves around a lot.
44:18
Unidentified speaker: He's been saying this on our podcast for the last year.
44:20
Unidentified speaker: When you think about China, he says when China walks into the room, profits go out.
44:25
Unidentified speaker: They have the cheapest cost of capital, cheapest cost of labor, and then the cheapest cost of energy.
44:31
Unidentified speaker: And that was all that thesis was all right till AI.
44:36
Unidentified speaker: Because there's one chip in the world fueling AI, it's NVIDIA.
44:40
Unidentified speaker: The best frontier, the models, when you think about anthropic open AI, they're years ahead of what China has.
44:48
Unidentified speaker: The hyperscalers are all if you look relative to what's happening, big tech in China, and then going on Palantir on the software side and cybersecurity, CrowdStrike, and others.
44:59
Unidentified speaker: China is a phenomenal market.
45:03
Unidentified speaker: Robotics, energy, way ahead.
45:06
Unidentified speaker: But I just think it's part of why China's market's doing what it's doing, is that the US, for the first time in 30 years, is way ahead.
45:15
Unidentified speaker: Except physical AI, like our good friend Gene Munster.
45:18
Unidentified speaker: That's what he's excited.
45:19
Unidentified speaker: I know you're excited about that, right?
45:21
Unidentified speaker: And China's just killing it as far as robotics and stuff like that.
45:25
Unidentified speaker: No doubt.
45:25
Unidentified speaker: Physical AI, they're way and also regulatory-wise, it's much more receptive.
45:32
Unidentified speaker: They know from a national security standpoint, they're a national interest.
45:35
Unidentified speaker: And the thing that I worry about, or I would worry about as an investor in the valuations that we're seeing in some of these IPOs that are coming, and more so in the private markets on the way to the public markets, is that China has a huge interest to push their open source models around the world almost like a digital belt and road, if you will.
45:55
Unidentified speaker: You know what I mean?
45:56
Unidentified speaker: And then it makes we've always known that they have their firewall, but as they push those out, the purpose that they're open source is to get them around the world.
46:04
Unidentified speaker: But then NVIDIA hugging open.
46:06
Unidentified speaker: The point is, oh Jensen's a huge believer in open, and so am I.
46:10
Unidentified speaker: Like open source.
46:12
Unidentified speaker: He listened he'd love to have all his chips in China.
46:14
Unidentified speaker: He'd love to be able to sell it open source because then there's more users and the more token costs.
46:19
Unidentified speaker: And it comes at a time where he's never had more competition.
46:23
Unidentified speaker: If you think about all this custom silicon, you think about Anthropic and the deals that they're doing with Blackstone and Apollo and Broadcom backstop so they can buy TPUs from Google, which is great for Google, but it just seems so complicated at the moment.
46:38
Unidentified speaker: And all of these companies, the Frontier Labs and the Hyperscalers, and maybe not the NeoClouds, they're all developing their own specialized chips as we get past training, we get to inference in agentic AI.
46:50
Unidentified speaker: So I'm curious how you think of that as it relates to NVIDIA in particular, because it seems so complicated.
46:56
Unidentified speaker: And then NVIDIA is investing anywhere they can to get the ecosystem to buy their chips.
47:02
Unidentified speaker: I'm not officially, but it's you talk to any executive at Alibaba, the within a millisecond, they'd want NVIDIA chips versus domestic.
47:12
Unidentified speaker: Look, but that's also why Jensen's a believer that they have to sell into China because by not selling to China, it makes China that much more competitive.
47:21
Unidentified speaker: But look where Nvidia's trade, 11, 12 times earnings, probably nine times what real numbers.
47:28
Unidentified speaker: What makes you the most nervous about your AI revolution call?
47:35
Unidentified speaker: I think listen, I think most folks who are looking at the space from an investment standpoint probably agree with most of your view looking out, let's call it five, 10 years, right?
47:46
Unidentified speaker: But in between that, and we've just discussed this, sure, there's going to be lots of fits and starts.
47:50
Unidentified speaker: And maybe this is one of them right here.
47:53
Unidentified speaker: But it didn't take a lot for the 2020 and 2021 rally that we saw, this raging bull market to basically correct in 2022.
48:05
Unidentified speaker: Without Chat GPT in late 22, and again, that's stupid for me to say that because we can't prove you know what I mean.
48:12
Unidentified speaker: What we probably wouldn't have the sort of multi-year run that we've had, right?
48:17
Unidentified speaker: But of course, AI, and no doubt.
48:19
Unidentified speaker: If you go back to 22, where AI was the catalog.
48:23
Unidentified speaker: So based on everything that we've just talked about, and my pushback to some degree, and I hope it doesn't come across that I'm like ultimately so negative.
48:31
Unidentified speaker: We talked about this when we started the thing, fleshing it out.
48:33
Unidentified speaker: What makes it technology?
48:35
Unidentified speaker: So what no, you do I mean what makes you nervous in the near term that we could have a bear market that lasts six, nine, twelve months or something like that?
48:45
Unidentified speaker: Because it's not unheard of.
48:46
Unidentified speaker: And we're already seeing here's a great example.
48:48
Unidentified speaker: Reddit was down 30% on Friday because they're getting absolutely slayed by all these models coming in, taking their data, right?
48:57
Unidentified speaker: Now they're going to a closed web.
48:59
Unidentified speaker: They're like enough of the open web.
49:01
Unidentified speaker: And so there's so much disruption, whether it's software, whether it's these internet names, we've seen the way that, you know, a whole in this agentic AI world, the way travel companies have been hit, and the list goes on and on.
49:14
Unidentified speaker: That's just in the last few weeks.
49:15
Unidentified speaker: What makes you nervous about the potential, let's say, of a 30 to 40% drawdown in the Nasdaq 100, especially at a time where we've seen the sox correct materially, and the sox was leading the way out of the gate and a really outperformer for years in this sort of trade, if you will.
49:33
Unidentified speaker: Look, the thing that makes me the most nervous, none that to that level, but I'm saying just in overall in the thesis, look, it's really it's the politicalization.
49:42
Unidentified speaker: It's the the polarization of the Bears versus the Bulls, you think?
49:46
Unidentified speaker: It's the politicalization from DC.
49:48
Unidentified speaker: It's the worry about like in terms of data center, the data center build out.
49:53
Unidentified speaker: And if you have a blue wave, do you have like more terms like Hokel in New York?
49:58
Unidentified speaker: But that to me, as some of for the first time in 30 years, the US is ahead of China when it comes to tech.
50:04
Unidentified speaker: That to me is the biggest risk.
50:07
Unidentified speaker: Okay.
50:07
Unidentified speaker: So we're gonna know tomorrow, Tuesday is the midterms, and maybe we'll have to do that.
50:11
Unidentified speaker: As well as relative to grandstanding, seeing what they're gonna do, but then ultimately what Okay.
50:16
Unidentified speaker: So let me ask you this.
50:17
Unidentified speaker: Let's just play this out for one minute.
50:19
Unidentified speaker: So let's say tomorrow, I think it's pretty well established that the Dems are gonna win the House.
50:25
Unidentified speaker: The Senate, it's never been this close.
50:27
Unidentified speaker: So let's just say they win the Senate and they have, let's say, a two, you know what I mean, three seat majority, there's some Texas, which is the biggest data center state.
50:37
Unidentified speaker: That Abbott, the current governor, put a moratorium for political reasons, you know what I mean, on new data centers.
50:44
Unidentified speaker: And that race looks really tight.
50:46
Unidentified speaker: So if you have some governors, some statehouse flip, that sort of thing, what do you think happens to, let's say, the major, you know, like the whole trade, if you will?
50:55
Unidentified speaker: I think it's gonna look that there's nervousness.
50:59
Unidentified speaker: I think part of what's happened in the market now going, it's been, I think in the last few weeks, I think that's played into it in terms of the nervousness.
51:07
Unidentified speaker: But as we've seen many times, a lot of times what p politicians talk about are much different because the reality is data centers will go from one state to the other, and the engineering jobs and what accompanies them will also go.
51:20
Unidentified speaker: But I think, look, this is we are going through just a very nervous period where these tech companies, these stalwarts, need to show monetization, and you need to really show that the money they're raising is leading to what is gonna be like a whole new growth frontier.
51:43
Unidentified speaker: There is a scenario where they raise the money, they start achieving some of the monetization that they promised, they take their foot off the pedal of the capex bit, and then they end up having great balance sheets, and then they can they already had great balance sheets, and then they get back to positive free cash flow.
51:59
Unidentified speaker: That's my view it is 27 is the call the low water mark or whatever, and then that's where capex and monetization flip.
52:10
Unidentified speaker: But there's a lot of time between now and then, and I think we're going into one of these periods.
52:15
Unidentified speaker: If this is managed right, I think there's a tech bull market, like I said, next three years.
52:21
Unidentified speaker: Yeah.
52:22
Unidentified speaker: Dan Ives, you've always been so generous with your time.
52:24
Unidentified speaker: I love fleshing this out, my man.
52:26
Unidentified speaker: Let's do it again after earnings season when we have a good sense of what the midterms look like, that sort of thing.
52:32
Unidentified speaker: Because my good friend Mike Wilson over at Morgan Stanley, he looks at this, the midterms and getting through it, and maybe some of the stuff that's going AI and the trepidation about what it means for GDP, the CapEx as almost a clearing event.
52:44
Unidentified speaker: And he's basically saying 8,000 on the SP 500.
52:48
Unidentified speaker: It can't get there without these major hyperscalers and some of the tech ecosystem.
52:52
Unidentified speaker: So let's do it again.
52:53
Unidentified speaker: I really appreciate your time, my man.
This page was produced automatically: the audio was transcribed with nvidia/parakeet-tdt-0.6b-v3 and interpreted by Qwen3.6-35B-A3B-GGUF:UD-Q4_K_XL. Transcription and interpretation can contain errors; always confirm against the original episode. Nothing here is investment advice.

More from RiskReversal Pod

RiskReversal Pod

There’s a New Kid in Town: Nvidia, Anthropic & the AI Unwind

Episode discussion: Hosts discuss Nvidia valuation, Anthropic IPO, and broad AI trade unwind amid leadership changes.

Speakers’ claims: Guy Adami and Dan Nathan analyze Nvidia's recent price action and valuation, noting the stock traded near support levels despite high growth expectations. They highlight the leadership change with Jensen Huang's departure and cite a Wall Street Journal report on internal concerns and circular financing involving neo-clouds. The hosts connect the AI sector's weakness to rising Treasury yields and Bitcoin's drop, while discussing Anthropic's massive IPO. They draw parallels to Cisco's 2007 collapse and note a broad unwind in semiconductors, leaving only large-cap tech stocks as safe havens.

NvidiaAI TradeAnthropic IPO
36 min6 assets20 claims to verify
RiskReversal Pod

The Jensen Huang Saga Is Far From Over + The Future of Search, Ads & AI with Joe Marchese

Episode discussion: Jensen Huang's resignation and AI agent disruption reshape tech valuations and ad models

Speakers’ claims: Dan Nathan and Guy Adami discuss Jensen Huang's resignation from Nvidia, citing espionage concerns, margin pressure, and customer concentration risks. Joe Marchese joins to argue that AI agents will disrupt traditional search and advertising by prioritizing fiduciary duty over ad revenue. The speakers contrast the resilience of physical experience-based businesses with the vulnerability of digital 'toll takers.' They also analyze the competitive positioning of major tech giants like Microsoft, Google, and Meta in the emerging AI landscape.

NvidiaJensen HuangAI Agents
1h 10m7 assets20 claims to verify
All insights