NVDA是逢低买入的机会；尽管存在竞争和利润率担忧，但其0.7 PEG的估值显得便宜。

Uh, I don't think it's going to I personally am not the biggest uh believer that uh we still won't have some nervousness over the next couple days because people get nervous about Nvidia earnings coming up, but I don't think uh any of this will be uh too terribly extreme.

So, um we'll get through it. Uh we were talking about that a little bit in game planning this morning with Broadcom and uh Marll, AMD, Nvidia, and Apple as well.

Okay, good. So, uh all all eyes really on Nvidia right now. Nvidia, if we take a peek at it for earnings tomorrow, this is trading for about a.7 peg right now. The valuation for Nvidia is so low uh very very ridiculously low. I think there's a lot of fear that uh competition is coming to Nvidia and uh their reign of terror, Jensen's terror is over. I don't know. The markets tend to get overly enthusiastic or overly pessimistic pretty regularly. But we'll see if this stuff holds.

We've got a nice little pop on Marll. Uh they we've got a little bit of a drop on Nvidia coming in pre earnings. This can be a wait can eventually end up being a wait on the cues if Nvidia tanks too much. Of course, it's just the first minute candle, so who cares?

Nice recovery here on the uh Philly Semi index as Nvidia is green for now.

However, we still have Nvidia earnings coming tomorrow. And then we still have people getting nervous about Kevin Wars rug pulling us on Friday. I don't actually think he will, you know, I think he's gonna be too much of a Trump puppet and he's gonna do nothing, but we'll see.

Uh, Nvidia up 28 right now.

So here for example these are the GPU hours right now. Some of these they don't actually quote like the GB300 they're not quoting and the GB200 they're not quoting. These are just the newer black wells. So they're quoting the older stuff. See, like, look at this. The Nvidia RTX Pro 6000. Okay, why don't I just buy that myself? Oh, wait. We did. But why am I going to pay you a buck 80 per GPU hour? You know, I could I could buy this thing on Amazon. I think it's like 15 grand or whatever. Uh, let me see. We bought ours for like seven grand. Nvidia 6000 RTX. So, right now traded for Oh, that's hilarious. That's actually hilarious. You can see my last order. We bought multiple of these. Uh you can see my last order at the top there. It says uh you know, last purchase August 27th of 2025. That's funny. Uh but anyway, so I mean you could buy this now when we bought it for whatever it was uh7,000 or whatever and uh 144 is what you could buy it for. So you know how many hours in in a month? Well, I don't know. Let's let's just assume for a giggle that I'm running it 24 hours a day. That's 720. What's my payback period? 720 in this case. you need to pay I need I'd still need a year and a half you know at that price you know cuz once you know and then you got to install it into a machine too you know so it's probably still going to cost you like 17 grand to run one of those or you put it into a server rack so you got to add a little bit for for it. So, you're still probably at if you're running it 24/7, 17,000 divided by 720. Yeah, that's still a 2-year payback period, but then you own it and it's your data. So, I think that's that's becoming more and more of the competition. Uh, and that's the ondemand. I guess I I guess in fairness, if you're using it all the time, they do give you pricing discounts here. So, you know, you contract it up. But that makes sense. But at this price, at this price, it's a four-year break even for them, unless they're getting, you know, bulk pricing, which maybe they are. They probably are getting some kind of discount from Nvidia because that's retail pricing on Amazon. So, you could see the business model. You know, let's say they're all in on these RTXs. Maybe they got them a year ago like we did and they were, you know, seven grand or whatever. They got Nvidia prices and they're getting them for seven grand. Well, $7,000 still takes me uh well, basically 7,000 hours plus the miscellaneous other hardware takes me call it 9,000 hours to pay for itself. 9,000 hours divided by 365. It takes a year. That's still a year. I mean, it's good for Nebius. Yeah, they they could pull off, but then then again, like if the payback periods are so low, why don't you buy it yourself? Not everybody's obviously going to buy themselves, but that's going to become more competition. Certainly not these these larger racks, the more Frontier racks, but that's that's interesting seeing the pricing for this. Then you own your own hardware. I mean, for those that we bought last year, we're we're probably already at the break even price because we got them so much cheaper. Now, it's like a two-year payback period. It was a one-year payback period because the prices have gone up. But anyway, uh so this is cool. Oracles got getting gotten a little bit of a pump in here. This is I feel like your OG Nebius uh credit default swaps are wild on this puppy. But the debt is also insane over here and that's one of the the challenges that I find of the uh these guys.

OpenAI Nvidia Yappen. So, this is a Bloomberg piece that came in. Open AAI says its new Jalapeno chips performed better than Nvidia's current lineup during testing. So, we don't really know exactly what they're referring to because you've got these new chips coming out. semi. Oh, it was measured against the GB300. No way. Wow. Wow. That's impressive. The processor led in two categories. The amount of work it could handle per unit and its speed at returning responses, right? But it's not so much the issue of can you have a chip that's better. It's can you scale the chip? That's the question is can this actually get scaled up? Can they get the allocation from TSM or whomever or where they're creating them with the you know where they're creating their processors. So the maker uh well so OpenAI plans to use the new chips to support AI models later this year. Part of a push to build its own AI joining a stampede of companies trying to compete. This is a partnership with Broadcom. The Broadcom AS6. Oh, we were just talking about Broadco Broadcom in the um in the course member live stream. Broadcom's flat on the news. Wow.

So, we don't really know exactly what they're referring to because you've got these new chips coming out. semi. Oh, it was measured against the GB300. No way. Wow. Wow. That's impressive. The processor led in two categories. The amount of work it could handle per unit and its speed at returning responses, right?

But it's not so much the issue of can you have a chip that's better. It's can you scale the chip? That's the question is can this actually get scaled up? Can they get the allocation from TSM or whomever or where they're creating them with the you know where they're creating their processors.

Okay, so that's this of the three items. My take my take this uh optimal is 45% human 45% open weight 10% frontier because the models like Quen uh which uh Reuters is now using from Alibaba are really good uh seen as you know maybe 3 to four months behind big L uh you know Frontier at fraction of cost and they keep their data right like why why would you give all your stuff to the Frontier Labs

Now Russia also has a lot of drone equipment which Nvidia chips have been implicated implicated in being inside of those Russian drones.

And we looked at Broadcom. We've got this OpenAI. This was the OpenAI TPU news. Halleluena wasn't tested against the new generation Vera Rubin which just began shipping. It's also not designed for training. Right. Well, it's it's that's the thing. It's Nvidia Nvidia's highest margins are on those training chips.

It's Nvidia Nvidia's highest margins are on those training chips. Is intended for the inference phase. Those are what the A6 are great for. Speed benefits such as OpenAI can achieve results only previously possible with chips that use a different kind of memory uh design.

For example, OpenAI currently relies on cerebra systems for some of its models, but that kind of chip is best suited for smaller models, right? Like voice inference and that.

Nvidia just announced the Jetson Orin Nano2 module and kit expected first half of 2027. Jalapeno can handle larger ones. But it won't replace providers like Cerebrus anytime soon.

Well, yeah, because it's all about scaling the actual tech. So much need for compute. This is why we signed up with so many different providers. Going to continue for a while. I think it's just a basket that does well.

the inference chip basket. Okay, so you've got obviously Nvidia, which would could be at the 5090 level, it could be at the 6000 level, uh it could be at the Groke level, etc. Right? You've got Broadcom for A6. You've got and and the Google TPU, which is also in partnership with um Broadcom. You've got Marll A6 and uh you obviously you've got AMD competing TSM as uh the upstream sort of manufacturer manufacturer. You've got uh ASML as the upstream manufacturer for the manufacturer, right? But um who am I missing on on like I mean there are obviously a ton of other individual you know chip component manufacturers chip components uh like LSCC's in there uh you know plus many others uh Vertive etc right but I mean this is more of like your your traditional like let's this is I'll just call this more the traditional um you know all chip stocks we think and this this would be more inference only so I'm going to say not not the big boys like Vera Rubin.

you've got obviously Nvidia, which would could be at the 5090 level, it could be at the 6000 level, uh it could be at the Groke level, etc.

So you just donated $10 or did this pop up twi? No, it only popped up once. donated $5 to say uh your uh your your courses aren't showing up. Why? Why? I can't help you live. But I can say you could just email the very support email. Yeah, you could have saved the $5. Like I give you your $5 back. Well, technically it's probably like $3.40, but just just email us. We take care of we clear out all of our emails every single day. Just email uh staff me.com. We'll get you taken care of. It's uh it, you know, no big deal at all. We'll gladly help you out. We help everybody out. So, okay. So, the inference chip basket right here,

Nvidia, the smaller chips, these are going to be lower margin though, right? So, these are lower margin side potentially uh for inference at uh Nvidia.

Uh, oh, I should put the G gro Gro ones. Well, Grock for high inference, Grock for high ultra ultra fast tokens.

This Nvidia is probably the cheapest out of all of these. Uh, pricing the cheapest out of all. The problem with Nvidia is you have uh uh probably some fear the what's it called? Um uh frontier margins will decay right in favor of inference. So that's some fear here.

Nvidia is probably the cheapest out of all of these. The problem with Nvidia is you have uh uh probably some fear the what's it called? Um uh frontier margins will decay right in favor of inference.

Q's are only up 40 bips right now. So, you're kind of seeing that decay. Uh, mostly probably because you've got Nvidia earnings coming up. You know, it's like that creates nervousness.

mostly probably because you've got Nvidia earnings coming up. You know, it's like that creates nervousness. Okay. Let's see what other news I can find here. especially on these on Nvidia. Let me go see if there's any juice on Nvidia. Nvidia. So, I've got yeah, the robotics computer, the open AI news, and then we had the Musk news about the space data centers, which we've heard for a while.

especially on these on Nvidia. Let me go see if there's any juice on Nvidia. Nvidia. So, I've got yeah, the robotics computer, the open AI news, and then we had the Musk news about the space data centers, which we've heard for a while.

If I search just Nvidia, Nvidia Q3. All right. So, earnings expectations. Let's go through a little bit of Nvidia expectations.

Nvidia Q3. All right. So, earnings expectations. Let's go through a little bit of Nvidia expectations. Okay. So, let's write some of this down. Okay. Nvidia expectations and uh the open AI news because that's obviously the big thing for tomorrow. Okay. earnings tomorrow before the bell.

Nvidia expectations and uh the open AI news because that's obviously the big thing for tomorrow. earnings tomorrow before the bell.

let's focus on Nvidia right now. So, Nvidia earnings expectations, they're expected to be I've got EPS adjusted expected to be at $2 uh 8.

We're looking at uh that's a beat seven out of eight times. Uh, historically, I've got EPS gap at $2.15. I'll just round this beats 8 out of eight.

I've got revenue expected to be 92.29 billion and that tends to beat eight out of eight times. But it's it's always a matter of like how much does it beat by, right? That's that's the big question.

let's focus on Nvidia here. So net is expected to be 50.98 billion. That beats seven out of eight times. And then let's see here notes we've got here. Likely to beat again networking and early Reuben strength. Oh yeah, that's we're going to see some uh Bloomberg suggests early Reuben strength is coming. Uh margin durability will be focused due to memory. That's fair as well as let's see uh top commentary on China. Not really. Nvidia needs a bigger upside to move the stock. Beaten raise remains intact.

notes we've got here. Likely to beat again networking and early Reuben strength. Oh yeah, that's we're going to see some uh Bloomberg suggests early Reuben strength is coming.

Uh margin durability will be focused due to memory. That's fair as well as let's see uh top commentary on China. Not really. Nvidia needs a bigger upside to move the stock. Beaten raise remains intact.

Open AAI says their TPU with Broadcom or I should say ASIC. I don't know if that's a TPU. ASIC with Broadcom Beats uh Nvidia's GB300 not yet tested against Vera Rubin for inference.

So that's you know sort of where my opinion remains. I uh think most inference applications don't need the biggestations don't need the largest uh chips and that's where you know these other inference plays that we were looking at I think have real upside in enterprise inference.

So, you know, we've done this we've um been doing this with reinvest AI. So, that's our, you know, um the homes AI and stock AI. it. And we, you know, we run these on on, uh, 5090s, 6000 Blackwells, uh, CPUs, smaller, uh, models.

And we're starting to see other companies do this as well. A lot of companies are doing this actually. Uh, companies are in-housing uh, the compute this way. You know, payback period are one to two years.

And then in addition to that, you've got, you know, it's your data. It's your data, not sending it anywhere. This is what uh Mona is doing. uh Reuters uh just announced uh they're doing uh they're running their own uh you know models without sending uh their proprietary uh data to the uh Frontier Labs.

The the issue with that for Nvidia is what are the margin differences which they don't break out but we expect their margins are much higher on the frontier chips.

So we expect I'll say that we expect margins on frontier chips are much higher hence focus on frontier spending and enthropic this talk about the open AI as beating this is not a surprise

Okay, what else do we have? Data center capital spending room to increase blah blah blah capex financing. Nvidia is increasingly using its balance sheet to expand the pool of infrastructure, right? And so this is this is where you've got big financing still coming.

So this is the bullish portion for Nvidia. What's bullish for Nvidia? $500 billion of financing 500 billion of financing for data centers.

You've got the SEC uh assetbacked security uh rule loosening no 5% skin in game and um details on what's in ABS's. So it's it's really, you know, that that a little bit feels like it's reminiscent of like 2008, right? It's like, oh yeah, let's let's remove some of the regulation. Hold on a sec. Turn that AC on a little bit more.

But, you know, in the short term, like long term, that's bubbly. Longterm, uh, that's bubbly. short term bullish, right? Uh what what else is bullish? Well, uh SpaceX spending, right? That's, you know, 80 billion raised. You've got uh Nebius has 8 billion of cash. But just to show you the difference, you know, if SpaceX has $80 billion raised, Google has uh $86 billion of uh money raised, and then you've got anthropic uh IPO coming, right? So, those are those are all sort of bullish Nvidia catalysts still. Well, uh SpaceX spending, right? That's, you know, 80 billion raised.

You've got uh Nebius has 8 billion of cash. But just to show you the difference, you know, if SpaceX has $80 billion raised, Google has uh $86 billion of uh money raised, and then you've got anthropic uh IPO coming, right? So, those are those are all sort of bullish Nvidia catalysts still.

Okay, let's uh let's send that through here and then keep keep yapping about this. I I think this is really understated. The enterprise in-house enterprise AI in-house enterprise AI is really understated. I think this is really understated. The enterprise in-house enterprise AI in-house enterprise AI is really understated. Um that's where we could say that the Gavin model that was my thesis is that we're probably 454510. Let's mention that. Okay, here we go. So that's this of the three items. My take my take this uh optimal is 45% human 45% open weight 10% frontier because the models like Quen uh which uh Reuters is now using from Alibaba are really good uh seen as you know maybe 3 to four months behind big L uh you know Frontier at fraction of cost and they keep their data right like why why would you give all your stuff to the Frontier Labs so that's interesting

All right, boring. I thought there'd be something a little bit maybe more Nvidia related since we're yapping about Nvidia right now. Okay, let's go back to this Nvidia talk.

Okay, let's go back to this Nvidia talk. So this was the open AI chip. See what else there is. Uh so many different providers. Let's see other startups are working on similar ideas. Uh low voltage chip over here from etched. Everybody's trying to get into the chip is OpenAI publicly tested it new chips using smaller open source models.

What? OpenAI tested it with Moonshot? No way. During internal test, it performed well running some of OpenAI's large advanced models that have been yet to be released. So, they Oh, they tested on on both of them. That's very interesting.

That's very interesting. Okay. Open said it used its own AI models to more rapidly develop the chip. Second version is already far along in the process. Company expects to tape out the chip. Final step in design in the coming months. designing the concepts for third generation where cost level power level infrastructure costs down. Yeah, I mean this is the race is to try to steal margins from Nvidia.

designing the concepts for third generation where cost level power level infrastructure costs down. Yeah, I mean this is the race is to try to steal margins from Nvidia.

Everybody is racing to take some of those Nvidia margins. Uh bearish for Nvidia. Everyone wants a slice or a piece piece of the cookie jar of margins from chips.

Okay, so let me see. Do I have my last Nvidia earnings piece? Nvidia earnings. Nvidia earnings. It's weird. That was basically three months ago.

It's weird. That was basically three months ago. Nvidia earnings. Come on, Apple search function. Come on, Apple search function. Okay, here we go. So, that would be three months ago would be like May, May, June maybe because Nvidia earnings 10Q. Oh, wow. Here we go.

Okay, pop this open. So, this is Nvidia's last earning sheet. 75% gross profit, 71.4% net. It's just so good, right? So, that's the income statement.

So, this is Nvidia's last earning sheet. 75% gross profit, 71.4% net. It's just so good, right? So, that's the income statement. And then let's see what the current uh let's see if these growth forecasts have been changed because a lot of them are turning negative by 2030 20 31. Let me see what these are updated. So I'll write down updated growth forecast 825.

Let me see. So here I have $9.3 of EPS projected. Before we had $8.99, so it hasn't really changed. And then the growth of 1309,627 3059. You actually get a little bit of boost here at the end. It's interesting on EPS.

So that's just over that's like 21%ish. So what's it trading for? 210 divided by 9003 equals it's trading for 23 times divided by you know call it 21 equals it's trading for you know one 1.1 peg right around there. Trading ain't cheap.

That is a I feel like that's a reduced estimate from what we've seen though because I think the the last estimates had higher growth. Nvidia. Yeah. So, I've got the growth at 32 was the prior estimate. So, why did the growth Oh, oh, oh, oh, oh. I screwed it up. I know what I did wrong. I made a little oopsies. All right. All right. So, I got a little oopsies on this Nvidia valuation here. So, right here, this I wrote down the actual EPS numbers, not the growth rate. My bad. That's funny that it actually comes out so close. So, these are the EPS growth rates. I don't How do they expect 87 over here and then negative 22.61? I mean, what a mess. Okay, there we go. 134 68. Oh, wow. Those have really shot up 33%. So now 2325 divided by 33.67 equals No, it's trading for.7 peg. Wow. Yeah, it's it's very cheap.

So a lot I feel like of the pain already feels discounted in. Uh but it's gone down each of the last four earnings that have come out. It could just be, you know, a continuation of competition fears. Uh, or it's just such a big company, right? What is it now? 5.7 trillion or a little less than that. Five five something. Uh, Nvidia Nvidia sits at 5.1 trillion as a cap.

See what else we have. 9.033. Let's update that. Okay, perfect. Okay, so that was the Gavin Baker item that we threw in for Nvidia.

Let's see. Let's see. Let's see. Let's see. Talked about that. Talked about that.

Morgan Stanley TPU sales for Google. CDS's uh for Broadcom or rocketing.

We could throw these up on screen. That's interesting. Nvidia is not on here. We could always add Nvidia, but this just I think I think the CDS's are really just hedging by uh hedge funds.

Let me grab this because think about like practically if you're long these stocks and then you get a credit default swap and they're rising in value, your hedge is actually going up while your potential exposure is going up. Uh ironic CDS um impact. So, as uh your AI stocks go up, your hedges have also been going up. So, you're winning both ways.

Insurance policies for default are going up and the underlying uh uh company values going up. you know, obviously, you know, the last couple months being a little uh a little funky, but but that is that is interesting.

So, let's consider that for a moment. Uh so, okay, good. So, that's sent. Good. Okay. So, we have this uh let's think of what else. So, we've got cash flow at Nvidia. Another thing people are going to be looking for is that cash flow statement. I think there's going to be a lot of focus on how much they're throwing out at investments.

So, we have this uh let's think of what else. So, we've got cash flow at Nvidia. Another thing people are going to be looking for is that cash flow statement. I think there's going to be a lot of focus on how much they're throwing out at investments.

So, where you're going to see that is going to be right purchases of non-marketable securities right here and marketable debt and equity securities. That's gonna be the next place you see a lot of this, I think, spawn in, if you will. Okay, so let's go back here. Ultraast tokens. this double hedging. I mean, you're hedging your upside and your downside.

Let's see here. Okay. All right. So, let's go delete this. Let's go back to Okay, so you're buying you win both ways here. That's fine. Just trying to think of the implications for this for Nvidia. I mean, so really you could be long Uh this way you can be long AI and be protected by default and right now both are going up which you know doesn't really seem sustainable but it kind of also makes sense because you know and and the more debt the more debt spending uh the more CDS's rise but the more debt spending, the the more the underlying AI names get more revenue. So, you got this like crazy cycle right now where you could literally be hedged and be winning on your hedges and your upsides right now. That's wild.

I mean, so really you could be long Uh this way you can be long AI and be protected by default and right now both are going up which you know doesn't really seem sustainable but it kind of also makes sense because you know and and the more debt the more debt spending uh the more CDS's rise but the more debt spending, the the more the underlying AI names get more revenue. So, you got this like crazy cycle right now where you could literally be hedged and be winning on your hedges and your upsides right now. That's wild.

That's pretty wild. Okay, I think that's cool. All right. Omani and Iranian foreign ministers discuss phase framework that provide practical and implemental basis implementable basis to resume safe navigation through the straight. Oh, that's cool. Well, that's going to I wonder if that's going to piss off Trump again. I don't know. We got We got to pay for the We got to pay for the uh what's it called? Um uh the API access for Truth Social so we could get it milliseconds faster.

Um, okay. All right. Very well. Now what? Okay, what a boot. What was the other news for Invidia? Video, we got the expectations. We know margin gaming has been a struggle. We know that.

So, this would be Oh, there's another piece that is really good. I Let me see if that's in my photos. I took pictures of the screenshots.

This could be really useful though in looking at Nvidia as well. Yeah, here it is. Okay. Yeah, this is a lot. Okay. This this this this this. Okay, let me airdrop this over. These are my annotated notes on what's going on with servers.

All right, they've arrived and now plug them in. So that is plug in get these images in. All right, here we go. downloads. Oh, why are they HIC format? Of course, I can't upload HICC's. Oh, that's so annoying. You take a screenshot with an iPad and they all turn into those high efficiency photos that Apple does. It's so dumb.

Oh, and I can't turn them all into a JPEG at once. Oh my lord. Technology is something else. It's all right. No problem. No problem. We'll do it. We'll do it one at a time. Oh, there we go. Okay. I can do it in two batches cuz I really want to show you these pictures. Oh, three batches. Stupid. I blame I blame Apple. I blame Apple for making it complicated. It's their fault. It's not our fault. They're delaying us. I want to get the Edgewater research up. It's really good. It's got some really cool insights.

We want it. We want it all. When do we want it? We want it now. Got it. All right. It might be a little out of order, but I think that's okay. Cool. So, here we go. This was the Edgewater piece.

So uh practical takeaways uh from Edgewater. So we expect GB 200 to 300 price increases up 20 to 30% in Q3. We did hear a 15% increase this weekend. uh a bit lower than Edgewater expected.

This is server CPUs expected to jump 50% year-over-year in 2027. That's kind of bullish. Vera and AMD at Nvidia and uh AMD. You've got Intel CPU sold out in 2027.

You've got this package shrinkage talk about memory coming down. I don't think that's that big of a deal. I think that lowers the price. Package shrinkage probably just to lower price. And do we really need all that memory? Right? That's the other aspect to think of because the memory companies are taking so much of that price.

RTX 6000 up. See RTX uh 6000 getting price increase uh reiterates my enterprise AI thesis at Nvidia. Gaming demand continues to weaken. Uh let's see here. Channel demand servers.

Channel demand for servers is starting to decline. So, you know, is is channel demand for servers declining an early warning or anything? Pro maybe I don't know. Channel demand for servers declining. H we'll just put the we'll put the hmm behind that. There we go. Okay.

We've got demand concentrating to the richest is what I mentioned. Yes, I agree with that. Um, okay, let's see. Yeah, I think you know the highest income for Nvidia is coming from like the labs like OpenAI, Enthropic and and that cost increases mostly memory. We know that memory prices are up.

Nvidia. Oh. Oh, they mentioned that uh the the stepping down of specs from Nvidia could be due to ASIC competition. Uh Edgewater believes spec down on memory uh for Nvidia GPUs could be due to ASIC competition.

Competition. Get that right. Okay, then potential for cutting HBM talking come out of view as potentially taking a step back with steps open weight Chinese models. Uh PC demand declining. They don't care if PCs go off because the frontiers are selling. Gaming, they don't care about the market.

Orders for PCs are slowing. Price forecasts for DRAM continue through at least the first half. See, here's the the cooling in memory DRAM prices uh for growth. you really see that cooling in 2027, but we kind of I've already expected that.

Okay. Cooling of cooling of price increases for DRAM expected in 2027 per edgewater. So, I thought this was this was actually really insightful. There was a lot of this and and you know, I just read off my highlights here. There is so much in these documents. Uh it's pretty remarkable.

All right, let's go see what we have for Oh, it is too cold now. Let me go see what the info is saying.

Okay. Uh that one. Let me lower this a little bit. Go. All right. Uh, Click House, OpenAI agent use, AI bots, Meta plans to launch Hatch AI agent platform. Apple CEO farewell. Fintex muscle in on AI spending. Let me see what this is.

This has to do with payments. Many sellers of AI tools have moved away from subscription pricing to charging by usage or tokens, right? Companies can also buy AI through application programming interface. I mean, this is when when there are tokens to burn. Price of AMI models themselves change a lot. See your dashboard needs to match what your bill says. different variables, API usage, whatever. That's boring. That's not that useful.

Okay, how the stick's doing right now. There's Nvidia.

Marll's doing great. Um, you know, and it's it's come down a little bit since the Leo fold. last I mean mostly today it's it's popping back what you've seen here same thing with AMD even since the 17th. Let's go back out. Yeah, here's AMD stock. Here's Marll.

This was peaked out right around the suckening. Oh, Netflix is up 2% today. That's cool. They kind of they they deserve it. I think they they make so much money. It's incredible how much how punished they've gotten in the market. They make a crapload of money. Tesla 352.

Yeah, Bitcoin still under 80. Corave 89 bucks. Applovin back over 300. Boy, that one's had some struggles to hold on to 300.

What else? What else? Disney Disney's gone up like 15% in the last week or so. Anyways, my cruise, it was right when I Look at that. My cruise was right here at the bottom. Well, that makes sense. I was on vacation the 24th to the 1st right here. Disney was trading while I was on the Disney cruise. You were trading for $93 to $95. You're now at 111 divided by 93. You have 19% the cruise. That's funny.

Yep. Yep. That's pretty cool. I like it. Okay. DRAM ETF is trying to This is the 2x ETF for memory. And then if I go to DRAM, I think that's just the onex. Yeah. So, you've got a little bit of a retracement here. Let's actually map it.

Uh, let's go the other way. Get the extension. Here we go. Let's go gray and make sure we got our numbers right. 34. Yeah, cool. Block that. Yeah, getting rejected by the second retracement here. Things have been a little stuck in the mud. Waiting on Nvidia.

Things have been a little stuck in the mud. Waiting on Nvidia. Waiting on Nvidia.

be interesting to see if uh you know how we end up closing today and tomorrow going into uh into the close. So this Axios no additional information on the purpose of the Radcliffe trip was provided in the Axios report. Okay. So then what is Axio reporting?

Let me see. Oxious. Uh, Radcliffe visiting Moscow for talks with the Russians. Highest level visit. We saw this. We talked about this. A C7 CIA didn't respond. There's no no news on that. So, nothing. Okay. No wonder they gave you a private tour. I will say touring the bridge was awesome. Get a picture with the captain and tour the bridge. It was kind of cool. I mean, and and you know, they like I was talking to them about aviation, too, and they they were really into commanding machinery. So, they they were fun to chat with.

Okay, what else? Let's go. Let's listen to the doomers. And we got to do a little breakdown on what's going on with Nvidia, too. Uh, we'll break down all of our notes in just a moment into some coordinated concept. Oman and Iran proposed temporary shipping lane through Straight of Horse. Yeah. Well, that's what Trump has been fussing about for a while already.

And we got to do a little breakdown on what's going on with Nvidia, too. Uh, we'll break down all of our notes in just a moment into some coordinated concept.

And then I got to hit the restroom, too. And then we will do some Invidior stuff. Good old Nvidia. Plane believed to be carrying Radcliffe departs Moscow. Okay, so now it's already leaving Moscow.

All right. Very well. All right. Let's listen into the doomers and then we break down Nvidia before earnings and then we break down Nvidia before earnings

>> the solutions for it. So without getting too much into the details and being all bit over simplification, >> we're helping them train the models in order for them to be safer. We're figuring out what could happen and then we're providing the solutions to make sure that it will never happen. And we're working with the Frontier Labs and also with enterprises that are either using their products or open weights models in order for them to be as safe as the best models in the in the industry because effectively we're seeing the market demands democratization of uh usage of AI and right now there's democratization of of harm as well when we want to make sure there's also democratization of protection and safety and security and that's all what we're about.

Nome Schwarz, founder CEO of Alice. It's great to have you on Bloomberg Tech. >> Thank you very much. >> Thank you very much indeed.

let's go do a little Nvidia breakdown. Did we do a price forecast on Nvidia? I don't think so yet. Okay. So this is the stock AI tab for Nvidia. I've got them at 886.

Uh let me see some extra notes we have here. GB 300 deployments are accelerating. Uh velocity, fastest product ramp in history, fastest product.

So, the pricing has been pretty stable. Here's your claw pop. Uh, right here. So, let's draw from there on. If I go first, let's do a rectangle. That's your claw pop, which was huge moment in agentic coding basically. applaud a gente coding pop. And then from here we can actually do kind of a you know you could see this trend line before. Let's do it like this. So this is your trend line of growth beforehand, right? And then if I draw in a second line to get growth afterwards, we could see the slope of these rental prices is actually sort of inlected up quite a bit. Right? That right there, that's pretty nice. You know, that's that's bullish. So this is a um bullish inflection up post um agentic coding uh codeex claude etc. And so far it's sustaining and that's going into August here. So that's decent. Got to give him credit for that.

Oh, let's see. They're talking about Nvidia here, too. Let's listen. >> Results that we got a couple of weeks ago. They really underlined how these companies remain pretty committed to pretty aggressive spending on AI. However, anything the company says about its outlook, what kind of adoption it's seeing, its pricing, uh, following the big price hike that was reported over the weekend, that's all going to be very much in focus and going to really help determine, uh, at least near-term sentiment for AI.

I'm more worried about this leaving Nvidia relying on concentrated hyperscaler customer base for growth. That is a risk because of this transition from hypers scale uh training to enterprise.

Nvidia dominates inference but SRAMM based accelerators this would be like your cerebrus rather for ultraast that's where gro uh you know gro is coming in gro whatever you want to call it

vera rubin architecture integrates seven excuse me uh purpose-built chips across five accelerated racks many flawless advanced I'm I'm not so worried about their packaging or production delays I'm not really worried about Um

uh valuation heavily heavily correlated into hyperscalers. Yes, that is a very fair risk. Very fair risk. US export controls on China. That's not a big deal to me. I will mark that as green. I don't care about China. We've already kind of written off China as

It's actually kind of like a crazy little intro to the Nvidia story because if that's true, then then Nvidia looks really cheap.

See, there you hear they're talking about Nvidia again. >> I think that's in front of us. I think that's obvious. Um, I I just want to add I really think that this call, this earnings call on Wednesday is going to be critical >> for Jensen Wong to be the voice that pushes against the data center push back that we are seeing around the country. I think it is going to intensify as you move into the fall. It's an interesting statistic that was brought to my desk this morning. data center construction is now 2.5%. That is bigger than infrastructure construction in the entire country. Think about that. We are now building out data centers more than we're building out infrastructure. So I think this has to be the beginning of that conversation where he engages because to Josh's point, if he's the central figure in the AI construction, he needs to address that. He needs to address that element.

if the growth is slowing at Nvidia, if that's the indication that we get from the company, that's the reason why that they're doing more in terms of shareholder returns. Number one, I think the fundamentals are going to be great. I think they're going to see more compute backlog higher uh because we're short compute, we're short memory and hyperscalers are spending all their money capex right now on memory. So I think you're going to see good backlogs, free cash flow. Getting back to your point, 200 billion this year, 300 billion next year. They already committed to returning 50% of their free cash flow to shareholders.

My biggest con my biggest question is if they do it is it is it because the growth is slowing and I don't think that is the case. no no no no speaking on behalf of long-term Nvidia shareholders We do not want increased dividends. There the the tax treatment of dividends is absurd relative to um the the tax treatment on buybacks. The tax treatment on buybacks is 1% the company pays. Dividends are double taxation. Meaning Nvidia is paying the tax on earning the profit and then distributing it to us and we pay another tax. Nobody wants that. Nvidia is not going to yield 4% to the point where the dividend is going to attract a new shareholder base. So I of all of the things that they can do to me for me and I think I'm speaking on behalf of many Nvidia shareholders institutions and individuals please reinvest into robots automation get involved in space anything but paying me a.5% dividend. There the the tax treatment of dividends is absurd relative to um the the tax treatment on buybacks. The tax treatment on buybacks is 1% the company pays. Dividends are double taxation. Meaning Nvidia is paying the tax on earning the profit and then distributing it to us and we pay another tax. Nobody wants that.

Nvidia is not going to yield 4% to the point where the dividend is going to attract a new shareholder base. So I of all of the things that they can do to me for me and I think I'm speaking on behalf of many Nvidia shareholders institutions and individuals please reinvest into robots automation get involved in space anything but paying me a.5% dividend. They literally are doing all of that. That's exactly what people complain about that their balance sheet is uh is is basically getting utilized to reinvest in all of those things.

I think robotic AI which is heavily vision based is super accretive to Nvidia uh in my opinion. What did we have Nvidia announce today? They announced here it is. Uh they announced Jetson Orin Jetson Orin Nano2 robotics computer announced today.

We have Musk space optimized Vera Rubin. Musk cease space optimized Vera Ruben NVLink 72 for launch in Q4 2027 uh scale in 2028. I think that's a little bit more of um a little more nonsensical at this point. I think it's way too early for that. I'm not saying it's not going to happen in the future. I just think it's I think that's too early. Way too early.

I think that's a little bit more of um a little more nonsensical at this point. I think it's way too early for that. I'm not saying it's not going to happen in the future. I just think it's I think that's too early.

according to Wall Street Journal, Nvidia will license pools technology and bring more than 100 employees into their Neotron project. The goal is to challenge Chinese open models,

6 billion into poolside for open weight

Nvidia Nvidia is investing six billion into poolside to compete in open weight. I mean that's the direction it goes. That's the enterprise direction.

Jensen announced the entrylevel edge AI robotics computer. Groot Foundation model, we know that.

Oh, they're talking about Marll now. That's cool.

Well, it looks like Nvidia is now expected. Oh, wait. It's not Nvidia. Take two. Take two. You screwed it up, Kevin. You said Nvidia. You meant to say anthropic.

That is like a total redefinition of total addressable market since that's basically 100% of the entire GDP of the United States in 2025 which sounds bullish maybe for Nvidia.

And that's what we got to talk about in this segment. Nvidia expectations for earnings, the new Jalapeno chip from Open AI. We got to talk Nvidia robotics, Nvidia openweight, enterprise AI, GPU rental rates.

So, first thing that we need to do is uh recognize that Nvidia earnings are tomorrow. Uh they are as usual after the bell. We do expect Nvidia to beat. Historically they always end up beating.

Adjusted EPS expectations $29. Nvidia beats seven out of eight of the last times. Uh EPS is expected on the GAP basis at $2.15. Nvidia beats eight out of eight times on this. Revenue is expected at 92.29 billion with a beat eight out of eight of the last times.

Bank of America actually thinks they'll be closer to 94 billion. And that ends up being a really big deal for Nvidia because even though Nvidia can beat on revenue eight out of eight times, Nvidia in the last four earnings releases has gone down afterwards, which is an insult to this business that makes so much money and is so freaking cheap right now. It's crazy.

$94 billion is the Bank of America expectation. you really have to beat the expectations uh that are on the higher end. So current expectation on average consensus is 92.29. Usually for the stock to go up, you have to beat the higher of the expectations. It's such a weird like psychology game.

Anyway, net $50.98 billion expected, which if you do the math on that, the net divided by the expected revenue means they're taking about 55.2% of the bottom line. every dollar that they make, 55.2% is after tax profit. It's so good.

Anyway, um the big focus on this earnings report is all going to have to do with margin, especially since we just saw a 15% price increase, which was actually lower than the 20 to 30% price increase that Edgewater Research was projecting for the third quarter. We had a 15% price increase to offset some of those memory prices.

And we've seen some respecting and some rejiggering of some of these leading edge uh Nvidia chips where we're seeing them come in with lower headroom for uh VRAMm. So basically lower uh exposure to that memory issue which then makes you wonder like why why are we making like creating shrinkflation in chips? Why are we reducing memory there?

But it could also be that maybe we just don't need that much memory in these chips.

Bloomberg released this article that says OpenAI claims its new chips can outperform Nvidia processors in tests. This chip is called the Jalapeno chip uh and it beats the GB300 in their benchmarking. Obviously this is biased and I'm bringing this up because it is something that people see as a risk factor.

If we can get them from SRAMM based chips like the Nvidia Gro that was just announced for ultraast token generation, super low latency, imperceptible latency for like voice AI customer service. That's an Nvidia product. the Gro they they acquired essentially all the IP from from Grock or Gro G R O Q

you can see there's a lot of benefit from the now Frontier Lab saying, "Hey, you know, we're we're not just focused on the greatest leading edge frontier chips from Nvidia anymore, like the Vera Rubin chip. We're also really focused on how can we fill out our inference stack and Nvidia is realizing this and that's going to be one of the risk factors for Nvidia. And it's probably why Nvidia is selling for such a dirt cheap valuation.

Nvidia's PEG ratio right now sits at about 69, which is an insult to Nvidia how cheap this is. This is why our stock AI indicates this has like a 4x upside at our price target is $886 for Nvidia uh at at a fair value just because their margins are so high. Why then are they selling for so much lower? Like what's going on?

Okay, so first of all, uh you've got Nvidia. It unfortunately has this large company deficit. You know, it's $5.7 trillion. It takes a lot for a $5.7 trillion stock to move. So, that makes it hard. That's one thing it has going against it.

The other thing is it is not the sole competitor in uh or competitor in ultra fast token generation because you have Cerebrris who's actually ahead of them. This just entered production with Nvidia which it's an Nvidia product now so it should do well but Cerebrris is ahead of them.

But the issue here is you have almost two equal I would almost say Cerebrus is ahead competitors whereas Nvidia has this big moat in the bleeding edge right you think about like the Vera Rubin uh space compute which in fairness you've got Elon Musk talking about how Vera Rubin space optimized chips are going to get launched into space in Q427 and then scale in 2028. That's great, but that's a bit of a long shot. That's sort of your uh Vera Rubin for space. I would call it uh hope right now. It's it's a little early for that. And so that's getting discounted.

Groke is getting discounted. Uh its size is getting discounted. Like can this really 2x from here? Right? That's the problem. You know, Forex is going to require this being like a 20 trillion dollar company. Based on how much money they make, you could justify it, but can you actually get to it is the other question, right?

Then robotics are still very early. Nvidia is uh just announced uh yet another uh sort of robotics release, if you will. It's the Jetson Orin Nano2 robotics computer for sort of lower power edge computing. That is the future. But robotics again, it's a future hope that could be a decade out before we really scale GPU heavy vision-based training for robotics.

You're more lab-based edge edgebased right now. I should be careful with the word lab since the frontier labs use a crapload of GPUs, right? Like anthropic and open AAI. That should not be confused with lab-based robotics, which is more, you know, very very early stage mostly because you still have issues with actuators and the actual mechanics of keeping these things running for more than, you know, a 30-minute demo.

So anyway, so you've got some of these issues, large size. This is maybe Cerebrus might be beating them here. here. I mean, they already are to market. They've got Cerebrus has customer concentration risk with OpenAI. Some of their uh IPO issues have d-risked, especially now that the stock has come down as much as it has. So, I like it better now than I did then.

So, I like it better now than I did then. Vera Rubin uh space is more of a future optionality. Robots are future optionality. So, a lot right now comes down to margins. And this is going to be the focus from folks is if you're raising prices 15% but memory prices have gone up substantially more than that, how stable are your margins going to be going into earnings? And so there's some nervousness that maybe we'll see some softening of earnings, especially as some of the focus from uh enterprise companies goes towards inference chips rather than these bleeding edge frontier chips.

And this is going to be the focus from folks is if you're raising prices 15% but memory prices have gone up substantially more than that, how stable are your margins going to be going into earnings? And so there's some nervousness that maybe we'll see some softening of earnings, especially as some of the focus from uh enterprise companies goes towards inference chips rather than these bleeding edge frontier chips.

Think about inference chips for a moment. This sounds crazy, but the gaming PC uh that operates or any gaming PC that operates on a 5090 GPU is pretty much good enough to run a lot of enterprise artificial intelligence. You don't even have to buy what we did, which is a Blackwell 6000. We've bought, you know, a few of these Blackwell 6000 chips. We bought these back when they were $7,000. We 2xed our money on the damn chip. Like I don't even know. I don't even think the stock 2xed in the time that the chip 2xed, right?

You don't even have to buy what we did, which is a Blackwell 6000. We've bought, you know, a few of these Blackwell 6000 chips. We bought these back when they were $7,000. We 2xed our money on the damn chip. Like I don't even know. I don't even think the stock 2xed in the time that the chip 2xed, right? Uh so it's kind of crazy, but a lot of enterprise AI does not need this memory level.

Uh so it's kind of crazy, but a lot of enterprise AI does not need this memory level. You don't need 96 GB of DDR7 headroom. you can handle a lot of enterprise uh inference with a 5090 which operates at a third of the size of this on a 32 gigabyte uh uh you know VRAM set.

You don't need 96 GB of DDR7 headroom. you can handle a lot of enterprise uh inference with a 5090 which operates at a third of the size of this on a 32 gigabyte uh uh you know VRAM set.

So what does that mean? Well, what that means is you might end up seeing companies preferring A6 from Broadcom, Google TPUs, which are expecting to scale next year, uh, or Marll A6, uh, or Nvidia 5090s or 6000s over the bleeding edge Vera Rubin chips. That, in my opinion, is why Nvidia is starting to

That, in my opinion, is why Nvidia is starting to say, "Hey, we're actually going to downsize how much memory we're putting on these massive frontier uh uh chips because the enterprise applications don't actually need that much memory on those big chip stacks."

So, there's been a lot of rumoring in social media over, oh, they're reducing how much memory they're putting in the chips. This is shrinkflation, you know, they're they're cutting corners. I think it's so they could keep the cost down and actually make it useful for enterprises because the chips are getting too big.

We don't need these massive stacks because you could do so much work with these little chips. You could literally take a $4,900 I think is the retail price. That's not even what you could get them at bulk pricing, right? You could take an Nvidia 5090 RTX chip right now. Oh, they've gone up another $200. Are you kidding me? These were literally $4,900. one left in stock at least on this store. Uh but anyway, you know, this was $4,900 like last week. What? And the delivery, you can't even get it until September. That's stupid.

Uh so anyway, yeah, I mean, these are selling out. One of the reasons these are selling out is because you can take a company like Moderna who's doing fantastic work with using artificial intelligence to maximize their speed to therapies for their clients. You could, in my opinion, do what they're doing with artificial intelligence on a 5090 GPU and you don't need these crazy uh Vera Rubin CPUs. I'm not GPUs GPUs.

What I mean by that is, and I want you to visualize this because I think it's really useful. Back in the day, it used to take Mona, like back in 2022, uh 2022, let's call it, it would take about 60 days to sequence a cancer patient's tumor and then select which mutations they wanted their therapy to target. And so then in 60 days, they would target maybe one to five mutations in these cancer cancer cells. That was not really ideal. You want to target a lot more mutations so you have a more durable solution.

Now in AI in 2026 or with AI, you could actually do this in hours and you could get to 32 m mutations that you can target in these therapies custom made for a cancer patient. and you could literally run it on a 5090 gaming GPU. The problem is these gaming GPUs are lower margin for Nvidia than the big $4 million racks.

Now, one of the reasons this is so relatable to me, this this Mona example, and I I don't like to sort of sound like a broken record, but that's what we do with our Reinvest AI. It's our, you know, our homes AI where you go into a a county and you're like, I want the best deal in this area and it sorts deals by here's how much money you can make. That's proprietary AI. Uh, or stock AI, proprietary AI. Uh, we're not making chat bots. We don't care to compete in that space. We're trying to compete in a in a practical AI place. Just like Mona, they're not trying to make you a chatbot. They're trying to give you practical artificial intelligence. and they're practically AI could work on 5090s and you don't even need to stack them together. You know, there's this idea that, oh, you could stack four 5090s together and operate them together. You don't even need to do that because you could just do four patients with four 5090 GPUs rather than stacking four together to solve one problem. You can solve four problems. The point of that is enterprise AI is going towards smaller chips, not bigger chips.

My take. We'll see. We're seeing a lot of companies do that and the payback periods are honestly like one to two years on a lot of those chips. We were calculating some of this in uh in our course member live stream. Remember, you can always get our trade alerts over there. See the long-term portfolio, what how we're investing for the next decade, how we're allocating. But, um, join that over at meet.com before Friday. We got a big expiration there. But, uh, you know, we were calculating some of the break even periods and we're like, yeah, you know, our chips have basically already paid for themselves. So, there is some reasonleness to the idea of short payback periods. The question is, how long are those pay payback periods going to last like that? To be determined, right? Those are the spot prices we're monitoring. How do the spot prices turn into long-term contracts? We'll end up seeing.

Mona gets to in-house their own data. Reinvest gets to in-house its own data. We don't have to send it all to the labs. The problem is that's bad for margins for Nvidia potentially in the future if they're now selling fewer of those Frontier chips and more of the enterprise chips. You get compression there.

Now, those are some of the issues that people are worried about with Nvidia. I'm not really worried about that just for this next earning set because I think Nvidia is just so cheap and I think the stock has already discounted a lot of that.

Again, if you're trading for a 69 peg, it's dirt cheap. Let's actually go look at the financials so you could see it. So, these are the Nvidia financials right here. And these are the numbers we have for Nvidia. The current growth forecast for August 25th are that uh at the end of the year they will generate $9 of earnings per share and their growth rate is expected to be 44 24 87 and then negative 22%. So you do have sort of a a curving out on growth at the end of the decade. That works out to about an average growth rate of about 33%. which uh means they're trading for about a 69 peg right now which is very cheap.

Part of that could be because there is negative growth forecast in about four years. That could be as we see Frontier trip uh chips become less critical and the inference chips much like the jalapeno from OpenAI created in partnership with Rodcom taking over. That's bullish for a company like Broadcom.

Now, what's like what are some things that are really bullish for Nvidia right now? Well, $500 billion of financing for data centers and the SEC changing the assetbacked uh securization rules. That's freaking crazy. Okay, that's like 2008 kind of stuff where the SEC says, "Hey, you no longer need 5% skin in the game if you're going to sell asset back securities for data centers because we think this time is different." Oh, we're also going to loosen the rules and say you don't actually have to disclose what the components of these assetbacked securities are or what the utilization rates are inside these data centers. To me, that sounds like 2008 all over again. I don't like that. I personally still maintain that, you know, you don't really see 2008 style tops until you get uh like Anthropic rollover. You know, we led this segment with this idea that Enthropic is projecting $30 trillion of potential revenue. I mean, that's like the greatest TAM sales pitch you could potentially ever come up with. Uh but whatever. Those rules are short-term bullish, more debt, but also more spending.

And that's where one of the weird ironies comes in. A lot of people are talking about, "But Kevin, credit default swaps are skyrocketing on these plays. Like, isn't that isn't that a risk?" Sure it is. But there's an irony here. Hedge funds who were exposed to Meta, Oracle, Google, Nvidia, because Nvidia credit default swaps are also rising. hedge funds who are buying Nvidia or Meta and Google and then they're hedging with credit default swaps are actually winning on both sides right now.

Think about how weird that is. Usually, usually a hedge would mean if Nvidia stock goes up, a put contract would go down. But that's not what's happening here as a hedge. In this case, as Nvidia goes up, credit default swap hedges are actually going up. So, if you own Nvidia and you own credit default swaps, Nvidia goes up because there's more debt going into the system creating more spending. Nvidia stock goes up and that extra debt then makes people nervous. So, credit default swaps go up and then your hedge actually goes up, which would be your insurance plan in case Nvidia defaulted, right?

So, weirdly, we're at a place right now where hedge funds can literally hedge and see their hedges and their stocks both go positive together at the same time. Uh, if Nvidia or Oracle or whatever go down, those CDS's, ironically, might still go up. So, it's sort of like the best hedge. The stock goes down, your hedge goes up. The uh stock goes up because of more debt financing, your hedge goes up. It almost makes you just want to buy credit default swaps. I mean, maybe that's not a bad idea. Obviously, that's derivative based, trading based. It's a little different.

But think about the other things that are bullish for Nvidia. SpaceX spending, $80 billion of spending or $80 billion raised on IPO. They're probably going to raise another three to $400 billion. Uh, Nebius has $8 billion of cash on hand. Google had just raised $86 billion for uh for Capex. You've got the Anthropic IPO coming. Yes, everybody wants their hand in the cookie jar to get a piece of these chips. Yes, we're going to see more movement towards inference chips and enterprise chips versus these Frontier chips in my opinion. But so far, Nvidia looks pretty freaking discounted for all of that uh you know in mind.

So, let's look at some more more of these takeaways here. Uh and let's look at a little bit research from a little bit of research from Edgewater. So this is a uh a bunch of server based research uh for Nvidia and uh CPUs. Uh GB200s to 300s had were expected to increase by 20 to 30% in Q3. Those actually ended up getting increased by about 15%. So a bit lower than expected.

Server CPU deliveries are expected to rise 50% in uh 2027 which is really fantastic. Uh that's right here. uh Nvidia or Intel CPU capacity is sold out through 2027 or into 2027.

uh Nvidia or Intel CPU capacity is sold out through 2027 or into 2027. It says price increases are still expected on Blackwell products, but they haven't happened yet. They just happened. This was published like days before the price increases. The cost increases are mostly memory as GPU prices are unchanged and Nvidia has secured enough supply. So that does indicate we expect all that attention on margin. Did you guys raise prices enough to offset the memory cost increases? That's where there's going to be a lot of focus on Nvidia earnings. We'll cover those tomorrow after the market close.

The cost increases are mostly memory as GPU prices are unchanged and Nvidia has secured enough supply.

So that does indicate we expect all that attention on margin. Did you guys raise prices enough to offset the memory cost increases? That's where there's going to be a lot of focus on Nvidia earnings.

We'll cover those tomorrow after the market close. This uh this idea about package shrinkage. A lot of people talking about how this is just inflation from memory. I personally also think it's just staying competitive with prices for that enterprise AI. We touched on that. We've got um the RTX 6000 right here, which we I just showed on Amazon, has gotten an increase on its MSRP from $8,000 to $11,000. So, you are increasing the prices 37% on enterprise chips because that is where the demand is going, which is pretty impressive. I agree with that.

We've got um the RTX 6000 right here, which we I just showed on Amazon, has gotten an increase on its MSRP from $8,000 to $11,000. So, you are increasing the prices 37% on enterprise chips because that is where the demand is going, which is pretty impressive.

Channel demand for servers is declining and gaming demand is weakening. gaming and PC deliveries. Not a surprise that we're seeing weakness there. I do think it's interesting that Edgewater literally says we don't know how customers continue to get more of a budget for AI servers, but standard compute servers are in decline. So I think when they say here channel demand for servers is starting to decline, I think they're talking about traditional servers because AI servers so far are still booming. uh as they say here standard compute our customers are only upgrading if they have to. So that's your classic cloud compute rather than AI compute.

I do think it's interesting that Edgewater literally says we don't know how customers continue to get more of a budget for AI servers, but standard compute servers are in decline.

So I think when they say here channel demand for servers is starting to decline, I think they're talking about traditional servers because AI servers so far are still booming.

Uh the other notes that they make is that they expect DRAM prices to cool in 27. This is also not surprising. We could see price increases. So that second derivative, the growth rate gets sharply negative into 2027 with DRAM prices only expected to grow 0 to 5% in 2027 depending on which stack of DRAM you're in. Whereas at the beginning of the first half of 2026, you know, these were all up 60 to 120%. A lot of that is expected to cool.

Edgewater also argues that Nvidia might be starting to feel some pressure on uh openweight models, which is why they're getting in. You know, they're spending $6 billion to get in with poolside to launch their own openweight models. But they're also possibly trying to compete

You know, they're spending $6 billion to get in with poolside to launch their own openweight models. But they're also possibly trying to compete against the Deep Seeks and the Moonshots as they mentioned here, openweight Chinese LLMs.

And I mention that I think what's happening is there's an economizing of spend going on. So, hey, why are we going to overload these chips with all this memory if people don't need it? Let's cut the memory, cut the entry cost for these chips, get these GPUs into the hands of people, uh, spend less on memory as a proportion of that, and then let them go apply their enterprise uh, AI.

Again, though, the question is ultimately, what is the margin for a Vera Rubin versus an RTX 6000? And that's probably why we just saw the RTX 6000 go up 37 and a half% in MSRP because that's where enterprises are moving to.

that's that's kind of the core set of info for for Nvidia. Uh again, they're expected to beat across the board as they historically have, but that's not what's going to move the stock. What's going to move the stock is what we hear about margins, frontier chip deployment, competition against open weight, lower margin enterprise. Those are the red flags we're going to look for.

We really going to care right now about space or Groot for the robotics AI or the robotics chips? Not really. Do we think Gro is really going to definitely be better than Cerebrris for an SRAMM inference chip? We don't know. We'll see. Remains to be seen.

All those things are really a sideeshow. Gaming is expected to continue to decline. That's also a sideeshow.

Another thing we'll be looking at is Nvidia's investments into other companies. You'll be able to see that not as capex. It doesn't show up as capex because you know capital expenditures will show up here in PPE. Where you'll actually see it is right above that in purchases of non-marketable securities and marketable debt and equity securities.

And you can see this about $26 billion of spend right here in investments so that those companies they're investing in can go buy more Nvidia product. So that's where some of the circular allegations come from. So we'll be watching for that as well.

Broadcom's still flat. AMD is still sitting at about 775. It's been pretty stable there. A couple percent over here on Nvidia been pretty stable on the day.

Nvidia Ruben Ultra will ship with one uh 92 GB of HBM 4. Huge downgrade from the preview of one terabyte of HBM. Even lower than the regular Reuben, which has 288. How did it get cut to 1/5?

Nvidia has not confirmed this. I don't think it really matters. But I don't think it's a big deal that the cutting memory. I actually think it's good.

Elon Musk and Nvidia Jensen uh just essentially talked about how they think they'll have a space optimized Veraro Ruben chip by the end of 2027 ready for launch by the end of 27 and probably scaling into 2028.

Nvidia. If you're having trouble keeping all this straight, so am I. But don't worry, Kate Rooney will make sense of it next. Next. I want to hear it now. We'll make sense of it right now. We already did. We made our We made our Nvidia video. Nvidia video.

traders await Nvidia's numbers after the bell tomorrow.

Largest bottleneck the brain quality training data that's what Nvidia is trying to solve.

Historically, we wouldn't even have this debate, but I think that the earnings report tomorrow from Nvidia supersede everything else this everything else that happens this week. But what he says about the report, the margins, the Reuben, and really the whole ecosystem, I think to me that's more important because this market is trading on the AI trade, Scott, more so than it's trading on interest rates at the moment.

We're not suggesting that, you know, what he says won't have a direct impact on Nvidia, of course it will. But the way that that stock tends to trade around its earnings, you know, in the immediate aftermath and the way that other parts of the market have traded, I just feel like it's the hundreds of other things that are potentially at stake, which carries maybe more weight in terms of what he might say and how those other things trade and then how people view the trajectory of the market from here.

So I totally agree with you and I think this quarter more than ever but one of the reasons that Nvidia tends to not have this enthusiasm post report uh I think is in they're choosing to report so late that we get a lot of the runup a lot of the anticipation from the other companies so that you're you're sort of thinking about what Nvidia is going to do because we've seen so much from it before.

but one of the reasons that Nvidia tends to not have this enthusiasm post report uh I think is in they're choosing to report so late that we get a lot of the runup a lot of the anticipation from the other companies so that you're you're sort of thinking about what Nvidia is going to do because we've seen so much from it before.

Yeah. This thing's running down into the print right after this really nice rebound. This been down seven straight days heading into today. Okay, it's green now, but it hasn't looked good lately.

Well, from a price perspective, I think you can buy it before the print, but who knows? We know the implied move tomorrow is going to probably be five or 6% one way or the other.

But that's that concentration risk is exactly the concern as to why Nvidia is trading for the discount it's trading for because it's like oh it's so heavily reliant on anthropic and open AI. Do we really need those?

What's on the line do you think for for Nvidia in terms of the AI trade? It may not be just about Nvidia as we've been talking about all hour. It's tens hundreds of stocks that lie in the balance maybe.

I think people look to Nvidia as sort of a B. I've got Nvidia's holding on to 2%. That's good.

We could probably do a video on it for Nvidia.

It's probably what you're going to get out of Nvidia, too.