Nvidia does what it can to bring cheer. Dark geopolitics still lurk in the background.

Welcome to the Sancto Market Call. Before we get started, it's important we emphasize that the views and opinions expressed in this podcast are those of the host and guests and do not constitute investment advice or recommendations. All information provided is for educational and entertainment purposes only. Hey everyone, it's Thursday, 27th of August, 2026 and market on a holding pattern yesterday ahead of these key tech earnings after the close and video, of course, the big one. And it was generally received positively and it was positive news. I'll get to that in a second. Actually, I'll go straight into the earnings because we can get that cleared away, Rubin. And you can head on about your day. Rubin, we have you in the podcast studio today, Austin to have you back. We had a little bit of discussions this morning around the Nvidia earnings report about what degree it was positive. I mean, clearly the headlines were positive and the reaction, the market was positive up over 5% after the close beating on, of course, all the important things revenue growth. I guess it was that forecast maybe that's part of the enthusiasm here as well. Yeah. Hi, John. Happy to be back. Absolutely. And I think the main reason why the stock was up 5% in after hours was the upbeat guidance above expectations for both next year, what they called the fiscal 2028. But actually super confusing, super confusing in the video. Why? Why? Yeah, it actually refers to covers mostly calendar 2027. So the ends in January, the fiscal 2028 ends at the end of January of 2028, just so we know where we are in the calendar when they say that. Exactly, but so it covers basically 2027. And yeah, the expectation was around 45% growth next year, but the company expects that it will be at least 70%. So it's a science that, you know, the high buildout is not slowing, maybe on the opposite. It continues to accelerate growth and maybe the only issue is supply shortages, especially in the memory space. Yeah, they said something about that, if we could produce the chips, we could sell, I don't know, we could have 100% growth next year, I think it was the indication. Yeah, I think they mentioned something like what is limiting how fast we're growing is actually those shortages. Yeah, yeah. So it's a supply issue. But I guess that, you know, demand is kind of the boring part of the story right now. We know that the spending is high demand is huge. And the interesting question is how quickly, actually, Nvidia can physically supply enough computing power to meet this huge demand. But yeah, just quickly on the numbers, I think they were really staggering. So just worth mentioning that, you know, revenue doubled from last year, data center revenue up to 17% and they got it for 800, 800, 800, 8 billion, 108 billion of revenue next quarter. 108. So I think they're at just under 100 now, 96, some like that. Yeah. So we continue to see sequential growth and this was also a number slightly above expectations. And, you know, no signs of any slowdown yet. But, you know, this is an industry where the product cycle matters a lot as well. And Vera Rubin is also kind of ramping really fast. And that is not affecting at all demand for the black world solutions. That's also suggesting that, you know, there is no this feared air pocket between these product generations. I think that's also good to see, you know, that, you know, even the previous generation is still generating a lot of demand. Yeah. It's kind of interesting too, because maybe I'm front running some of your points, but there's some of these Vera Rubin improvements in performance are pretty staggering. If you look at what they're actually claiming 10 times performance, so that's a combination of just the sort of the, the amount of data it can crunch all at one time. So there's 50 petaflops such as the mind boggling figure versus 10 to 15 depending on whether it's black, well, or black, well, ultra, this massive new expansion in the memory interconnect, which is really critical, critical bottleneck previously. And then the, the interconnect between, between the chip. So you think, if I'm concerned about performance, I have to have Vera Rubin. But yeah, a very interesting point on the black world demand. But I mean, this is all good, very good news, great, great print from Nvidia. I think the earnings call was also quite useful in terms of gauging a little bit what the outlook for the next year is and everything is fine. But there is an important wrinkle to mention, I think, which is the margins. The margins are coming down. They also feel the increased prices in memory and they're not charging a lot more their customers. I think they mentioned something like 15% price increases, which is not huge. It still shows that they have pricing power. But I mean, it was maybe a bit disappointing to see that, you know, the gross margin is going from 75% to roughly 70, 71%, which is still an outstanding, very profitable business. But you know, we don't want to see sequential decline in the margins, for sure. And I also want to say that I would expect a bit more color around the 500 billion in financing from Goldman Sachs and other private equity firms. They didn't mention a lot about that. Maybe it's too early and they don't want to say anything before maybe the deal is completely closed and official. But I would have expected a bit more color and the company, at least the CFO, to elaborate a bit more on that. And also the remains, I think, question around the circularity of finance around these companies. They have a hundred, they have backed actually 105 billion of data center in Ohio and that will be leased to open AI. So I just want to understand a little bit more, I mean, I understand this once in a lifetime opportunity. And this is a very revolutionary technology. And the public requires a lot of capital spending and they want to help the customers. So there could be, you know, a good reason to do it. But I do want to understand that a bit better from management, from the company itself, from Nvidia, which I think they were a bit muted on this topic as well. So those key watch points, I think, remain question marks for the upcoming quarters. Yeah. And then there's also the, maybe related to that, at least one, one sort of independent variable, if you will, that you can track is the CDS prices for Nvidia bonds, which have been, you know, they're still very low levels. So the odds of default super low, of course, but they've been elevated and they've gone up recently. They did come back down a little bit before the earnings report, curious to see how those develop in the wake of, wake of this earnings report. But I mean, just in terms of the demand, there was just a separate news item. I happened to cross that Amazon is committing to buying two million, two million Nvidia chips. And if they mean two million Nvidia chips, meaning via Rubenships, that's, you know, a $10,000, which is probably a lot cheaper than a, than a Rubens Cross. That's, we're talking about 20 billion dollars just for Amazon one, one client. Of course, these top clients are the big business for Nvidia. These top, top hyperscalers. Yeah, I would just add to that that this is the big opportunity, but probably also the big risk. Yeah. I think it's now, it's not good to have such a high customer concentration, but we increasingly see signs of more clients, more customers, more companies buying the Nvidia chips, also outside that the usual suspects in the technology space. So I think that's a good, good early sign that they might diversify a little bit the customer base. But for now, it's very concentrated among the hyperscalers. And they're all, not all of them, but almost all of them are trying to make their own chips to compete directly. So they don't have to pay those fat margins to Nvidia. They can retain some of that themselves. Amazon's training and Google's GPU, there's even this new chip that Broadcom is developing and they're trying to claim that it's just as good as or better than, I can't remember one of the Nvidia chips, jalapeno, I think it's called with the open AI, et cetera. So yeah, it's, well, it's going to be one to watch in the coming, let's call it five to ten years, the degree to which Nvidia can retain its performance and market share crown. All right, let's go on to a couple of quite big and a bit diverse within the AI space. Companies reported also with a very enthusiastic reception to the reports. Sales force is the first one to talk about 14% after the close. I think this was one of these stocks that was painted with the ONO software as a service disruption brush, but it's certainly making a comeback here. What do you make of their earning support in the reaction? Well, the print was good, revenue grew 11%. But that includes a little accounting, I think something to press to be a little bit more precise on the accounting because it raised full year forecast, but that includes also the informatic acquisition and unrealized gains on the anthropic stake as they own a little stake in anthropic as well. So I think the results, the actual operational results were not as strong as maybe they looked at the first glance, but definitely, I think the comments from management were more uplifting and the good news, actually, because they announced an expanded partnership with anthropic. So the idea is that they want to combine the cloud's reasoning capabilities with Salesforce customer data, workflows and business rules. So they want to use their own data, their own customer flows inside cloud. So actually, customers can use cloud to build sales skills and do things such as prepare meetings, review deal health and check the pipeline, the customer pipeline directly in the chatbot. So it's a very interesting use case and the company is also trying to not be so reliant and dependent on one single model, but they want to be model agnostic. So they want to expand those type of partnerships with companies like Google, so Gemini and OpenAI, ChatGbT. So I think that would be a very interesting case and also showing how SAS and AI can coexist and can be increasingly more interconnected. As we shouldn't forget that those software as a service company's own very precious data on their customers, and it's very often confidential data that you're not willing to actually do traditional agents, AI agents cannot really access easily. So I think they can use data as the new kind of currency to collaborate and create these partnerships with AI, with their agents to improve their services. It's pretty interesting also just looking at the course that a stock price, as I was saying, it was one of these ones that was painted with the software as a service brush. What did it peeked out in? Late 2024, $370 something to share, I think it was. Bottomed out at what are we in the Sub-150, and now with this reaction is at 230. So we're talking about up 50% since June. There's been quite an interesting, you know, we've talked about so much in the podcast, but it's just, you know, yet again, we're seeing surprises coming up and markets over playing their hands on some of the skepticism around some of these software companies. Yeah. In general, there has been a revival of those software as a service companies. A little change maybe of sentiment and investors being a bit more selective when it comes to actually gauge this disruption risk. Before, I think they were penalizing company across the board. Now they're becoming a bit more selective and trying to see what the quarters and the management commentaries are before, you know, hammering the stock price. Yeah. Yeah. And we also had, you know, CrowdStrike up 10% hitting on all cylinders. It seems like these big companies with the biggest profiles, biggest market caps that reporting after the close yesterday take us through CrowdStrikes earnings report. CrowdStrike was really strong. I think cyber security may be one of the less or the most overlooked picks and showers, beneficiaries of AI. What I found particularly interesting is actually the file conflux. So basically they have this platform where actually was growing, was revenue doubled from last year's was going 100% which is, I mean, super strong is maybe as good as Nvidia basically. And revenue actually reached 200 more than 2 billion. And it's basically instead of purchasing individual CrowdStrike modules one by one, now a large customer can actually have a pre agreed commitment to the file conflux platform. So they can basically draw on the commitment over time. So adding or swapping models as needed. So it means that instead of buying the whole package, the whole platform that they offer because they have a very all encompassing solution. Now customers can actually buy, for example, end point security. And then later on cannot cloud security, identity protection, data protection, and other modules. So it's one broad agreement that doesn't force the clients to buy immediately the whole package. But they can, it gives a bit more of a modular and flick approach that gives flexibility to customers to not commit a big capital upfront. But over time, they tend to cross sell very effectively. So this solution is growing really fast. And we know that AI creates productivity, of course. But it also creates more software, more agents, more data, more points that need to be protected from hackers and potential attacks. So I do believe cybersecurity, I mean, maybe it's not a revolutionary idea. Many people know it, but it is a peaks and shower for AI. And it's definitely an era that I think we should all follow very closely. Not just peaks and showers, but also a response to a genetic AI. You know, all these agents doing God knows what all over a creation and just needing to have some kind of defense against what they might do, especially all that coverage of Anthropics, most recent model. I'm forgetting what it's called Fable 5 or something like that. That can can turn these agents loose on potentially on your security and go poking around in your systems if they're able to gain access. Ruben, that was awesome to have you with us today. Thanks for the overview. And we'll get on to a couple more news items here in the effects, sorry, effects. The single stock space before moving on to I think a couple other interesting areas in the most pressing being unfortunately on the geopolitical side where there's some things that we need to continue to highlight. The last company that was really enthusiastically responding was Octa, OKTA, after the after hours, up 15%. I'm a bit curious there. I mean, their growth rate is quite steady, quite nice, 10, 11, 12%, but nothing that's been shooting the lights out. But it's just I found the description quite interesting. It kind of shows you what companies are facing in the era of agentic AI. So I was like, what is this Octa company? What do they do? And the answer I got back, the two sentence summation, is identity and access management security platform. So what does that mean? It controls basically who or what and what being potentially an agent can access the company's application data and systems. So it's about the key role being an AI, therefore, is the identity and permission layer for agents. It has to guarantee what their authenticity is and limiting what resources they can access and then governing and auditing their actions. So just basically like, OK, we have an AI agent here. This is what it's allowed to access and to do and monitoring what it's actually doing. Pretty interesting business area for sure. All right. And then finally on single stock news, it was the meta settlement. I have $16.6 billion in front of me as the amount. I think that's correct. 29 states, so they settled with on this AI addiction risk. That's about less than one quarter of net income. But then you wonder, you know, where else could some of these suits be coming from? So there was a bit of a pop in the stock yesterday, but it settled a lot lower from where it opened. And then the overall session yesterday, I didn't even cover it. It was basically flat across the board. There's just no strong headline to break out there. And really curious to see how the price action develops. And now as we have the typical Friday weekly options expiry and the risk of a fall crush. If there is one, which is sort of independent of, you know, the quality of the earnings report and people's enthusiasm for it. And more about options flow and whether the implied volatility is sustained into the Friday options expiry. Those positions that were put on ahead of the event for now and videos holding up quite nicely. But let's see how it goes into the Friday clothes. Alright, a couple of things briefly on the geopolitical front. We still seem to be seeing these, you know, productive headlines for getting Hormuz traffic sorted and flowing through the straight. Ola continues to point out that it's one thing to ship tankers through the straight. If a tanker gets hit, you have an oil leak. Whereas if you have a product ship, product meaning gas diesel, and it gets hit, you have a massive explosion. It's a bomb on the water. So meaning the refined products are not really shipping to the same degree and are far more constrained and are far more important for global markets than the flow of crude at the moment. So it's ongoing question. And then linking to that is Michael Evry pointing out, and I'll put an article, sorry, a link to the article in the podcast episode description. This is a Bloomberg news story, citing sources talking about, quote, "US aims to revive civil war era court to claim Iran oil as prize." And there's something called prize courts and what are called letters of mark. You essentially have to go back to pirating days to understand what this is all about. So if the US doesn't want to fund the wildly expensive US Navy and its ships to run around and confiscate or seize and confiscate tankers and sell their cargos, it can put out so-called letters of mark or whatever the new equivalent of these might be called and allow private operators. You could essentially call them pirates, mercenaries, whatever you want to call them to do the same. They would get some kind of monetary benefit from whatever is in that cargo once it is sold on to the market. So it's basically extending your power through this kind of privateering type of operation. Fascinating stuff, it does the US choose to extend its power reach in this way and this becomes the mechanism by which it is continuing to not just continuing, but extending its chokehold on the Iranian economy, what's it's called the D-Day against Iran's economy. And then how would a power like China responds to something like this if it's a cargo that's heading towards China? All kinds of questions start to multiply if this becomes US policy. It's an area that every was talking about already years ago, this potential use of this letters of mark style approach and interesting to see what the outcome there is. I will also do. just linked to today's FT Alphaville links in general. There seem to be a lot of really good ones in today's list. And I think very much worth pursuing at least half of those. So enjoy those links too. And then just want to, you know, re-emphasize what I said. And the prior podcast concerns about Ukraine and whether things are about to get even worse there. Russia is possibly moving towards this general mobilization that there are these new Shahed based originated drones, Wall Street Journal by the way had an interesting article about the origin of the Shahed drone, which Russia calls Giran. The new ones have jets with a lot faster. It's much harder for Ukraine to shoot these down because they're twice three times the speed and can evade defenses as well as Ukraine missile batteries, the ones that are from the US or basically US missile batteries. I think they're having trouble replenishing those because of the cost of the missiles and the stocks of the missiles, talks of more ballistic missiles coming from Russia is, is this mounting? Is this about to mount into a new crescendo here? I don't know. There's even a terrible Bloomberg mentioned of tactical nuke, Scott forbid. And Poland is moving tanks towards the borders. I'm noticing actually the Polish Latte looks a little bit nervous if I look at a chart. So just I don't like to put out fear and fear and loathing type of vibes, but I am concerned and I just want to mention that on the treasury front, we had a strange couple of days recently. We had that big crush lower and yields made sense. We had a very, very large move in the crude oil on Tuesday. And then yesterday, especially in Europe, there was quite a strong rebound in yields, confusing whether this could continue. I'm surprised they rebounded that much when we have, when we have crude oil still at these, these lower prices, these lower levels, but scratching my head a bit, but just, you know, awaiting for some kind of outcome there. We're all waiting for Jackson Hole, which I've covered a bit at Nazim. So I won't even say any more about it today because I'll still have tomorrow's podcast to talk about it before Mr. Worsh actually speaks concerned that we don't get as much of a signal as the market wants from Mr. Worsh tomorrow. In the meantime, in FX, we are at near record, not record, but we're at near very, very low levels of volatility historically if you're measuring by average true trading range. And in fact, I tracked this on my FX board, which I put out with my FX updates. And euro pairs on average are trading, have a trading ranges in the bottom 5% of observations for the last 1000 trading days just to give you an idea. On the US dollar pairs, we're down to the bottom 6 to 7% of the, of the observations of the last 1000 trading days, so going on 3 and a half plus years. So part of this, I think, is the low volatility process at classes. It's also the suppression of dollar in volatility because the yen has gone back lower. Again, people don't want to maybe push their case there in these yen crosses because of the knowledge that Japanese official dumb and maybe even coordinated official dumb with the US will push back against that move. So we're all waiting and seeing dollar again, meanwhile, back towards the top of that range 15930 plus getting into the 160 area at some point if it goes much higher, which I think were the fewer levels pick up further of intervention. Meanwhile, I've got euro swiss kind of starting to put on back on put back on that carry trade vibe squeezing back towards 94 again, which is near the top of the cycle. Let's see if that continues. I think I would associate that generally with low volatility, continuing across asset classes, where we're not seeing low volatility has been a gold and a Bitcoin of late tried to push higher again. They're not really pushing new high levels gold coming back in offered this morning after pushing to 4640, but look at the grains market. It is on fire very much linked to and I meant to actually link it to the Ukraine story. You have Russia trying to shut down their ability to export and kind of succeeding. Therefore, the the grain markets are a bit nervous. There are all kinds of other supply issues. I'll try to get oil on the podcast tomorrow to talk about all that. We had wheat at limit up at one point yesterday and all three of the grains, major grains wheat corn and soy hitting these new highs for the cycle. So not not good for the inflation story. If food prices are set to be pressuring higher as well in the whole climate situation with El Nino can potentially affect the rice side of the other major global sort of carb input crop, if you will. So a bit of concern on that front as well. Okay. That is a wrap for today. Stay careful out there and we'll be back tomorrow with the next Saxo market call. This has been the Saxo market call podcast. Thanks for joining today's episode. We're always happy for your feedback and questions of all kinds. To reach out, you can drop us an email at [email protected]. That's [email protected]. Saxo. Serious trading worldwide. [Music]