Limitless Podcast - THIS WEEK IN AI: Microsoft Build, NVIDIA at Computex, New Glenn Explosion

Episode Date: June 5, 2026

Today, we're covering Microsoft Build, including its new AI models and concept hardware, as well as NVIDIA’s latest announcements at Computex. We also discuss Blue Origin’s New Glenn test failure, Cloudflare’s bot traffic data, and updates from OpenAI, Grok, and Google.------🌌 LIMITLESS HQ ⬇️NEWSLETTER:    https://limitlessft.substack.com/FOLLOW ON X:   https://x.com/LimitlessFTSPOTIFY:             https://open.spotify.com/show/5oV29YUL8AzzwXkxEXlRMQAPPLE:                 https://podcasts.apple.com/us/podcast/limitless-podcast/id1813210890RSS FEED:           https://limitlessft.substack.com/------TIMESTAMPS0:00 Microsoft’s AI Hardware7:07 Nvidia’s Hardware Blitz11:53 Jensen's Next Pick14:11 Blue Origin New Glenn Failure17:15 Bots Outnumber Humans20:39 OpenAI’s Compute Crunch27:11 Grok’s New Coding Layer29:45 Google Raises More Cash------RESOURCESJosh: https://x.com/JoshKaleEjaaz: https://x.com/cryptopunk7213------Not financial or tax advice. See our investment disclosures here:https://www.bankless.com/disclosures⁠

We had a pretty big and exciting week in the world of AI and frontier technology. Lots of things happening. New AI hardware devices unveiled, rocket chips exploding. I want to start with the Microsoft event. Microsoft held their build event. Microsoft notoriously has been in the AI race only via companionship to Open AI. They are the largest investor, largest owner of Open AI. And that has been the connection.

This week, they severed even more ties with Open AI as they decided to build their own new hardware. Now, they released seven new AI models. they released two new AI devices. And to hear that Microsoft is building an AI hardware device was a little shocking. So I went on this kind of like race to try to figure out more information to understand how it worked. And what I found is these two devices that we're seeing on screen. We have a desktop device that looks kind of like an Amazon Echo Home Hub. And apparently it's very smart.

You can converse with it. It works with their new AI models. And the second device is this small looking key card that's something you wear around your neck. It's very smart. It has a display built into it. and it's used for a lot of kind of practical in the job type use cases. What I found out that disappointed me is that all of this was a concept. They marketed this as this amazing product that everyone was going to be able to use

showcasing all of the use cases, but the reality is that they're never actually going to ship these. In fact, this is just a concept for other companies to go off and get inspired to then build their own versions of. So here I was thinking, oh my God, wait, Microsoft's making a comeback. They have seven new models. They're releasing new hardware. And I think the reality is that Microsoft still kind of feels like they're in the same place they were prior to this build event.

Yeah, I've watched this video twice now, and I still don't know what these devices do. I feel like I've been transported back a decade ago, and Amazon's just announced the Echo again. Also, why are we wearing landyards as like futuristic AI tech? Now, I got hopeful when I saw this announcement, Josh, but I almost immediately killed that hope for one specific reason. I don't know anyone that is conceivably using Microsoft's AI products to a larger extent, right? So co-pilot is like their flagship AI assistant, and they've admitted that their co-product product is way behind the likes of ChatGBTGBT and Claude itself. So if I already know that people aren't really using the Microsoft AI ecosystem at the software level, why on earth would they use it at the hardware level? Now, I dug in to a little bit about what this thing is supposedly meant to do,

and it's supposedly meant to be like an AI agent, kind of like physical surface area. So what I mean by that is it's meant to listen to the things that you do. It's meant to see the things that you write or speak about, and then it's meant to kind of like ingest this data and improve your own personal agent, your own personal model. There's one issue, though. Microsoft doesn't even have a functioning V1 of an AI agent that can do all of this for you. They kind of branded as co-opter, but it's really,

chat GPD and it's an older model of chat GPD and it's less effective than chat GPD. So why wouldn't I just use chat GPD? Now, understandably, Microsoft recently has been trying to move away from their partnership with Open AI. There was a bit of tension between Sam Altman and Satney and Adela. And they're trying to kind of like sever that partnership and start becoming a more multimodal operation. They've started using Claude for a lot of their products and other types of models.

This is a step into the device side of things. This wasn't exciting to me and it maddens me so much that this is just a concept that apparently they don't plan to build at all. What did excite me was something that we covered earlier on in an earlier episode this week, which was their new Microsoft Surface AI engineered laptop, which is meant to run like one of Nvidia's latest chips, which allows you to kind of like run models locally. But this doesn't really inspire me. Well, something that might inspire some people if the hardware isn't doing it is the software

because Microsoft isn't just sitting there being dormant. They are working on their own AI models. They're working on large language models that I have to assume are somewhat derived from chat GPT models because they have that proprietary license. They must be leaning on that to some extent. And what we have here is seven new models that they're working on in parallel. And initially, I was a little frustrated by this too. I'm like, why the hell are they working on seven models at once?

But in the context of this image that we're seeing on screen, it makes a lot of sense. The models are supposed to be working on specific modalities altogether. So we have an image model. we have a large one and a flash one for smaller. We have a transcription model. We have a reasoning model. We have two voice models and a coding model. So they're building all of these in parallel.

I have to imagine they're going to be building the orchestrator, or at least using co-pilot as the kind of orchestration harness engine for these models, knowing what to serve them. And I think it's at least a start. Microsoft has done pretty much nothing in the world of AI. I'm not sure anybody uses co-pilot who isn't forced to buy their business. and I'm hopeful that these modalities, these models that they're working on

will actually yield some sort of impressive results from Microsoft. Otherwise, I'm getting close to putting Microsoft in the Apple category of just like, did not finish. Like they just like, were not able to actually execute, able to deliver anything valuable in this AI model race. Maybe this changes things.

Who knows? But after watching this, I kind of didn't feel that way. It was like, okay, the hardware that you designed that was really cool looking. It's just a concept. The software will, It's great on paper. These haven't been built.

These haven't been deployed. It would have been awesome if we could actually play with them today. That is not the case. So there's still a lot of work that needs to happen for Microsoft. But that was the build event. Yeah, I mean, the best thing that Microsoft has done in the entire AI race was invest in their biggest competitor, Open AI. When it comes to actually building models, it's not exactly their specialty.

But I applaud them for giving it a go. Now, one might think that they're using OpenAI's IP. They have OpenAI's IP for chat GPT because they have one of the largest. investors in that until I think it's 2030 or 2032 so they can technically use this IP to create their own models. But for one of these models specifically, which is the one I'm circling right now, MAI Thinking One, that is their main reasoning model. It's their main orchestrator model that orchestrates all these other transcription image, video and audio models. And it hasn't been distilled or using open AIP at all. It's been built from the ground.

out up, which is pretty ambitious for Microsoft. Now, you might be wondering, okay, what are some of the advantages that it's unlocked? Well, collectively across all of these different models, it uses 60% fewer tokens than your frontier models. But the models that it can kind of compare with them that it's on par with are like three generations behind the frontier models that we're seeing from Anthropic and Open Air right now. So it's a good attempt, but it's got me thinking, like, I don't know why Microsoft is focused

on this specifically when they should be kind of, doubling down on like the AI agent harness and trying to build a better AI experience across the humongous distribution that they have. They have like hundreds of millions of co-pilot users across the world. Like they should be able to kind of like monetize that somehow. And I think they're kind of wasting their time here. But there was a bunch of Microsoft announcements and one of them was announced in partnership with Nvidia.

And it was with their Surface laptop. And Nvidia's been kind of like on a streak this week. So there's this conference that's taking place in Taiwan right now called Computex. And I'm kind of calling it Nvidia GTC, which is their main conference, 2.0. It's basically Jensen's show. And he's taken the stage. Everyone's kind of like crowded out the entire rooms. And he's made a bunch of big announcements for Nvidia.

One of the main ones is Vera Rubin, NVL-72, which is kind of like it's the next hottest generation that has actually been built for GPU. is finally coming online, and there's a few partners that are bringing in online. You've got Dell that we mentioned earlier this week. We put out an episode, definitely go check that out because Dell's actually a sleeper hit for AI investing. Corweave and a number of other providers have helped stand up the very first Vera Rubin stack. It looks incredibly complex and quite impressive from an engineering feat.

I don't know if you've seen a picture of this, Josh, but like there are pipes and wires teeming out of this thing, and it's actually bigger than me. if I'm around like 6-3. It's like a humongous type thing. Incredibly impressive feat of networking architecture. That is one of like the major announcements. Jensen's kind of like leaning into a lot of new hardware stacks. Now, if you've watched an episode that we put out about a month ago,

we described why Jensen's kind of going all in on chips, if you didn't think you already was doing this, by recreating 15 of the core components that is required to build the best GPU. Now, typically, GPU providers aren't recommended to do that, because if something goes wrong, you end up needing to delay the launch of your next model. Not only did Jensen go for it, but he pulled it off. And so now we see these new GPUs come to life. He's also entering a new market, which is the CPU market.

Typically, Jensen likes to play in his lane. He kind of like sits at his own level of the stack. He owns the picks and shovels, and he doesn't want to get involved in anything else. But he's peaked over at Intel and AMD, and he's like, damn, these guys are making a bunch of money from CPUs. And in a future world where AI agents are everywhere, we're going to need CPUs to orchestrate these guys. And so he announced the launch of Nvidia CPUs, which currently is on track for a $20 billion revenue run rate this year. It's pretty insane. This is the first time I've ever heard of Computex as an event.

I didn't really even know that this existed. But it's basically the CES version in Taiwan. Like the Consumer Electronics Show, it happens every January in the United States. It's where all of the cool new technology and hardware gets announced. With Computex, it felt like the gaming version almost of this. And it's weird to say that because, sure, a lot of this is AI focused. But a lot of the offerings, a lot of the hardware was based on, it's funny, there's a lot of crossover, it turns out, between AI local inference running and gaming.

And that crossover is the GPU. And famously, Nvidia stopped producing the higher end GPUs that everyone was using in their computers to run local inference, but also to run video games at very high fidelity. What's happening here is through the release of the RTX Spark, which is pretty much the flagship announcement from this event, that's the new supercomputer, computer, laptop, MacBook competitor. It's now bringing the power back to the consumer, not only for gaming, but also for that Edge AI compute,

and we have pricing for it, which I thought was pretty interesting and competitive. The N1 systems featuring this RTS Spark, they are going to cost $1,800, and that is for a very high-end GPU that you can run locally on your, machine to run edge inference if you want to try small models or to actually run very high fidelity video games. The higher end of the system starts at $2,900. That includes the N1X system. And that can run like pretty beefy language models locally on your machine. It has 128 gigabytes of memory. It has, I think, 6,144 kuda cores, which for those who are unfamiliar with kudakores, it's a lot. You can just do a lot

of AI stuff on them. And it's really impressive. The other announcement that I really loved is this this thing called a NUC, an NUK. It's like they create these small computers. If you've ever ran an Ethereum validator or a cryptovalidator, odds are it was run locally on a little device called a NUC. They've built these single purpose machines now with these new graphical cards built into it. So now you could have a desktop machine that isn't really a laptop per se, but it's meant specifically for one purpose. And that purpose is running local edge inference on your device, running video games the highest fidelity. These are really powerful machines.

And I have to imagine this is beginning to, or I guess this is a continuation of the trend that we've been seeing of companies pushing these consumers to edge devices for this types of compute. It was very cool to see. Yeah, and it wouldn't be a Jensen Huang conference if he didn't bless one of the next companies that are going to be potentially a trillion dollar in valuation.

And he said it very explicitly on stage. The next trillion dollar company is Marvel. Now, I got to pull up the ticker. Yeah, yeah. Please, M-R-V-L, and please limit it to the last five days of price action. Because after he said those words, the next day the stock boomed up, I think it was like 100 to 150%. Something obscene like that. And if you're wondering why Jensen brought the CEO and founder of Marvel, and if you're

wondering what the hell is Marvel doing, I must say it has got nothing to do with Ironman or any of the cinematography around that. It is a much more boring kind of company, which is focused on next. networking architecture for all these GPUs and data centers. So a problem that a lot of AI labs have is when they're scaling their data centers, they get all these brand new expensive GPUs, but you need to be able to connect these GPUs together and get them to communicate with them at lightning speed as quick as they can. Sometimes these GPUs overheat, so you need to make sure you manage the electricity

and make sure that the data is transferred at the right amounts of times and the right quantities overnight versus day. There's a ton of complex networking architecture that is required. There's wiring pipelines, all this kind of stuff. Marvel as a company as one of the leaders that specialize in this. And Jets and Huang placed a $2 billion check via Nvidia into them. I think it was about in March, potentially. And then he made this announcement basically saying like,

you know, Marvel is the ones that are going to basically allow Nvidia to kind of like build out its moat as a data center kind of like expert. and their stock went up a massive amount of time. And he's pointing at one very important thing, which I want to mention, the next bottleneck constraint. So we've spoken about GPUs, compute. We've spoken about memory and architecture for memory a lot here, and those stocks have gone up.

He's saying the next bottleneck is going to be on power networking specifically, and that's what Marvel specializes. And they specialize in optical fiber networks and a ton of other stuff there. So that's going to be an interesting one to view. But moving on, there's been a massive explosion. I saw this all over my feet, Josh, over the weekend. And it's actually pretty intimidating to an extent. But this is Blue Origins New Glen Rocket,

which was being tested at Cape Canarvel and unfortunately exploded massively. Like, what is going on here? Why did this happen? Yeah, it's pretty intense. I actually had a post about this. This is one of my largest ever. Five million people saw. And it was because the comparison.

Harrison to this explosion is that it is roughly 20% of the energy of the Hiroshima atomic bomb that was released on the launch pad of Blue Origin. It is a horribly tragic happening for the space program as a whole, but also Blue Origin as a company. The pad, unfortunately, is the only pad that is capable of launching the Newland rockets to Earth, and now it's gone. It costs them over a billion dollars to make, and SpaceX needed well over seven months to rebuild theirs. after they had a similar issue. So this is a pretty large setback, and it comes back against the problem that Amazon needs, I think, the number of 1,18 satellites up in orbit by July 30th to maintain its license with the FCC. It currently has about 300. So the rocket that was supposed to really help with this contract is now gone, and the pad that was supposed to be launching it from

is gone with it. Now, they did have a response that I saw that seemed somewhat reassuring. It was from someone within the company. I think he's the, uh, oh, actually the CEO of Blue Origin. It's funny. I think Jeff Bezos is the CEO. Turns out it's this other guy named Dave Limp. And Dave Limb said that now that we've had access to the pad and integration facility, we could share a good bit of news. The propellant farm, oxygen, liquid hydrogen, and LNG tanks, liquid natural gas, are all in good shape. This is good because those are very long lead times. So it seems like a lot of the core components that were going to be causing a longer delay. aren't actually affected too badly.

There was a support tower that if you watch the video, that thing got destroyed. That's the strong arm that holds the rocket up. Totally got wiped out. He said, well, good news. We were planning to get rid of that anyway. So they're very much making light of the situation. They're trying to work their hardest. I just wanted to share this because it was one unbelievable explosion,

great content to watch if you are seeing the video version of this episode. But also that it may not be the worst setback in the world. And in terms of getting these blue origin satellites into space, thankfully they still do have SpaceX with the Falcon 9, Falcon Heavy, and soon to be Starship. So a bit of a setback in the space race, it's unfortunate. Like, SpaceX really could benefit from some competition. We don't want them going into their IPO as a complete and total monopoly. But here we are.

They're the only ones that can figure it out. And it's funny, we took space for granted because SpaceX has just been landing boosters back on barges in the middle of the ocean for like five plus years now. And it turns out space and rocket science really are like pretty difficult categories. So wishing the team all the best. A bummer, but alas, we will carry on. It was, it's probably like the best timing for Elon Musk. He's probably sort of, and he was like, you know, this is upsetting for the industry as a whole, but perfect for my IPO that's happening in probably next week.

Moving on, there was this interesting stat that we pulled from Cloudfare, which basically said that agentic traffic is growing so fast that it now has surpassed human traffic. It is over 50% of internet traffic online. Now, I can't say that this surprises me as much. I think the adoption and growth of people using AI LLMs and setting up their own agents, vibe coding different kinds of things that kind of like run autonomously 24-7 has increased exponentially at this point. And I think this is only going to get more aggressive. We were having a discussion prior to starting this recording.

And I was like pondering about a world where like, you know, future kids or generations of kids are largely going to interact with entities or beings online that don't have a physical body or if they do it might be robotic and we got into like a debate around like what that might potentially mean like from an educational stance right if you had a tutor that can kind of like personally tailor its lessons to you for this particular child that might be end up being useful but for someone who is like engaging in entertainment or leisure with a type of AI personality it gets a little blurry as to whether this is good or not. And I guess some versions of this already do exist. Like I watch a lot of YouTube. I kind of follow my favorite creators online. I don't really know them. I just know

them through the screen and through the content that I consume. So what if it means that it's AI generated completely? Anyway, I think we're going to see a lot of this going forwards. I think we're going to interact with a lot of businesses that are completely autonomously run by agents, but we would never know because we interact with them via tech, or they generate their own physical, sorry, digital AI avatars, but just an interesting stat and one that has gone under the radar, I think. Yeah, well, there's been this concept of the dead internet theory for a very long time, which basically says that a lot of the internet has ceased to be human-centric and is now mostly composed of bot activity. And we've seen this before. We've seen this through mostly

an election cycle. So you remember back in 2016, a lot of the propaganda that we were seeing on social media aggregator sites that were being published to the people, a lot of lot of it was bot activity. So we've had at least the appearance of bot activity, particularly on curated algorithmic feeds for a very long time. Well, Cloudflare sharing is basically the idea that not only is it greater than 50% of the content that you see, but genuinely now, 57% as of recording that's based on their charts, is bot activity, provably bot activity. So we've officially crossed the chasm. It's no longer a dead internet theory. It is a dead internet. And we are living in a bot driven world. The downstream implication.

of this are going to be interesting to think about, to discuss. I'm sure this is a very long conversation we're going to be having for many, many years going forward, but it's official. The flippinging has happened, as confirmed by Cloudflare, one of the largest internet providers, I think the largest internet provider in the world. So it's funny, too, they were guiding, they say in the post for the end of 2027, and then they bumped their timeline up to early 2027, and now we're in mid-20206, and that flip has happened. So things are clearly accelerating faster than I think anyone expect it. And what that means for the internet as a whole, for the people, the human users of it is to be determined. Now, one thing that we've spoken a lot about on this show

is the idea of scaling laws. So that follows the rule where if you want to create a better AI model, if you want to train a better AI model, if you want to inference a better AI model, you need a lot more compute. That comes in the form of GPUs. It comes in the form of energy or electricity to power these GPUs and to train these different models. Sarah Fryer, the CFO of OpenAI, is one of the few people probably in the industry that knows to the bones how true this statement is or how true this scaling law actually is upheld. And she had a very clear statement on her appearance on the all in conference where she said, compute through 2027 is completely sold out. And that is the clearest statement ever that the scaling laws are still kind of holding. We're still in the early,

early parts of 2026. We've heard similar trends for AI memory being sold out through the end of 2027, but not really on compute. And we also know that compute data centers specifically is the current focus of every major AI CEO in terms of scaling. They're trying to like put boots on the ground and build these things, hire all the technical employees that are required to set these things up. It is the most aggressive expenditure. Google just raised $85 billion. All these IPOs that was spoken about, Anthropic, Open AI and SpaceX is all to get more. money to invest in AI CAPEX. It's the number one focus. It's Krishna Rao, CFO of Anthropic's number one focus. So for Sarah Fry to also confirm this and say that she feels like, you know,

computer's going to be short for a long period of time through 2030 and 2032 is the clearest statement ever that Open AI, Anthropic and the likes of other top major AI labs aren't going to go away when it comes to AI CAPEX. Now, one thing that this forces me to think is all like challenge myself is it probably means then that we are not limited by capital. We're not limited by kind of this AI bubble that is being inflated. We're being limited by physical atoms itself. Like we don't have enough, no matter how much money you have, we don't have enough human capital to actually build these things. And so we're constrained by that, the regulations determining what land to build on it takes a lot of time. And that backlog, in my opinion,

is probably going to go on for a lot more. So we're probably going to see a lot more investments of this trend. Sarah Fryer is incredible. I would encourage everyone to go and watch this. There's a strong case to me that she should be running the company because, man, she is incredibly well-spoken and thoughtful and revealed a lot of things that weren't obviously apparent to me. One of them talking about the time frames in which she deals on. Like I'm very much under the impression that these AI labs are thinking maybe 12 to 24 months out because further than that, the supply chains are nearly impossible predict. She was sharing that they think up to six years out. And she's placing guidance for 2032, which to me feels like this impossibly complicated task.

So maybe she's crazy. Maybe there's some kind of merit to these ideas. But it was a really fascinating episode from the all-in guys just to understand kind of where opening I see themselves and how they plan to navigate through this next five to six year period in the sense that like you're saying, Ejazz, there's just not enough compute. So therefore, they're not worried about any slowing demand because the demand is going to continue to outpace that compute over the next six years. And so long as that demand outpaces the expense, these companies are going to become profitable. They're going to continue to go up.

And I mean, obviously she's guiding for this positive future, but it seems like they have the numbers to back it up. A second thing she mentioned that I found very, very exciting, was she talked about my favorite, our favorite hardware device, the Johnny Ive designed device. She actually referenced it by name. Johnny Ive was mentioned. And I think it was earbud that was mentioned or earpiece. She said Johnny would kill her if she called it. an earpiece, kind of acknowledging the fact that this is actually something that goes in your ears. The things that we have been seeing are somewhat real. Jason, one of the people who host the show, he referenced the puck, the earpiece. She did not. Rebut, she kind of was just like, yep,

it's happening. It's amazing. She used it. Yes. Like, people internally have been using it. It seems like this incredible device. And she confirmed that it will be announced before the end of this year. So we have confirmation at least on timelines happening sometime. I would assume around Q4, well, we'll get more information. And that is the holy grail of hardware. I cannot wait to get those to try it out just to see what it looks like. So that was really exciting. Open AI also, they launched sites, which is a cool new feature.

Back to like practical AI use cases that you can actually use. Tell us about sites that you guys. I know you've been looking through how these kind of work. So I'm a very visual learner. And one thing that I interact a lot with when I'm creating kind of like research and agendas with Claude is like, can you create me like a visual artifact? This is basically Open AI's attempt to kind of like not only compete, but Excel for that product. So in their own words, with sites, Codex can turn your work, ideas and plans into an interactive website or app your team can explore, use, and share with a URL.

Now, why I enjoy this the most is typically when you create a visual artifact, it's static. It's like an image. It kind of, you can kind of tell that it's created by an AI because it uses the same font. It's structured in a very particular way. And what I typically do with that is I take that, and I take it to ClaudeC and I say, hey, can you create an HTML version of this and enhance it in A, B, and C types of ways? This is a one button, one plug-in, one feature fix that Open AI is doing via Codex. So it's, you know, they're creating this like super app and this is kind of like a step towards that direction.

Well, when you ask for a visual artifact, it will now do this. And what I like about this is it's easily shareable. So like, it can create your own webpage, you can share it with someone and it introduces just a new medium and way to kind of read and learn about something. Like I would love to scroll a custom website with graphics, infographics that are kind of like responsive to the way that I move my cursor that kind of explains concepts very clearly to me. I would learn much faster that way versus just reading a block of text or looking at a static image. And I just think it's a really neat feature. And I want to see more companies rolling out features like this. Yeah, it seems like it's going

out to business and enterprise plans currently. So if you have one of those congratulations, you can go to try it out. I assume all the other plans will be following shortly after. I'm excited for this. So many people want to have their own websites. They've dreamed of making websites. They've paid tens to hundreds of thousands of dollars to have other people build websites. And now it's the matter of clicking a button and getting something that looks really lovely and that is functional and that accomplishes all of your goals. So I think in terms of people, just the average person, this is such a powerful tool and use case to empower them to build these websites. So that's great. We have two more topics quickly that we could probably breeze through.

The first is GROC launching with Composer 2.5. I'm just reading the headline here, making it competitive product to ClaudeCode. When I think of CloudCode, I think of, oh my gosh, this model is unbelievable. People are actually doing real work. Awesome. When I think of GROC, I think of, oh, yeah, that's right. That's the thing I used like six months ago and haven't touched since. So what's changed over those six months?

Yeah, so it's this one layer that sits on top of the model called the 8. Agent Harness. Now, a few weeks ago, we covered X-AI or SpaceX-X-AI forming a large partnership with this company called Cursor. Curser doesn't actually create their own AI model, but they have specialized a mode known as the Agent Harness, which basically sits on top of any frontier AI model and tells it, you know, kind of how to work, where to direct prompts. It kind of like fine-tunes it in a very specific way that actually includes a lot of information that is very valuable if you're able to own it. As a company, Cursor's been able to do that. Now, in this partnership, Elon has basically been able to acquire the right to buy Cursor within 30 days of their IPO. And as part of this agreement, that means he can take Cursor's very valuable moat, their harness, slap it on GROC 4.3, which is the latest model that he's training right now. That is, I believe, publicly available now as of this week. And, hey, presto, you have a frontier coding model. That's how powerful the harness layer is over the model. It's not just how good your model is, it's how good your harnesses as well.

Like Dario and Sam have both confirmed this. And actually, if you slap Cursor's harness right now on top of Claude Opus 4.8 or GPD 5.5 high, you end up having a better coding model than those models stand on their own. So Elon has been able to buy his way to the frontier top Composer 2.5 on its own, which is a Cursor's own model that they wrapped a harness over, is comparable. and on par with 4.8 and 5.5. So if you wrap it around Grok 4.3, I can imagine the same is the case.

That's pretty amazing. They have made some unbelievable progress. And I think, again, it's a testament to never count Elon out. They are making incredible progress. And as we know, they have the training models. They have the hardware. They've currently given it out to Anthropic. Elon restated publicly this week that it is more of a shorter term deal is kind of the idea

that they both have the option to opt out at an argument moment. and I think the terms were 90 days that he said or something like that. But it is cool to see another competitor in the race. It's great to see another harness that people are excited about. I mentioned there was two topics. The last one is actually Google's $80 billion race that we covered yesterday in the episode. So unless there's net new information, I think I'm going to tease it a little bit. And then if you want to see the full thing, we could go back to our episode from yesterday.

Yeah. And the only quick, oh, sorry. Yeah, what do you have? Do you have any updates? Well, yeah, the only quick update is they doubled down. and raised it to $85 billion. So there's an extra $5 billion that they raised at the last minute. And I just wanted to spend like 10 seconds saying a lot of people's takes on this raise is it's

bearish that Google's selling stock. They're not confident in the future of Kappex. I actually think it's the opposite. I think Google has $126 billion on their balance sheet and they can use that to raise higher debt in the future. So they're being strategic. This is financial bookkeeping, creative financial bookkeeping to make sure that they have a healthy balance sheet in the future for when they need to raise even more money. So sending a bit of

stocks now won't get punished if their stock value ends up going up more for the fewer shares that they end up having. So I think this is ultimately a good move. And I just want to go on record saying that for when it eventually is proven. And there you have it. So if you have made it to this part of the episode, congrats, you're totally caught up on the week. You can go touch grass, go enjoy your weekend. You have been exposed and up to date on everything. As for the next two weeks, buckle up. It's going to get exciting. Next week, we have W. WDC from Apple early in the week, which is huge. This is make a break for them.

If they cannot deliver again Apple, we're going to have to write them off to zero. The following week is seeming like it's going to be SpaceX IPO week. That's going to be huge, biggest IPO in history. And I'm sure along the way there will be many other releases. So buckle up, very excited. EJAS, any final parting thoughts before we head into our weekend? No, that's it. I'm curious if any of you were out at any of the major Microsoft build or compute X events,

let us know if there are any different takes that we missed. But aside from that, wherever you listen to us on YouTube, Spotify, Apple Music, please subscribe if you haven't. Please give us a rating if you haven't. Please leave us a comment if you haven't. All of these are incredibly valuable in helping us grow our channel. And we have grown massive. I think it's like 12,000 people over the last month,

which is just absolutely insane. It's lovely to meet you guys. And yeah, we will see you on the next one next week. Thanks so much. See you. See you guys.