1 00:00:00,020 --> 00:00:04,300 Josh: NVIDIA just held its GTC conference in San Jose, where Jensen Huang walked on 2 00:00:04,300 --> 00:00:08,240 Josh: stage in front of 30,000 people and opened with a number that's probably going 3 00:00:08,240 --> 00:00:12,600 Josh: to echo across Wall Street for weeks, a trillion dollars in expected orders through 2027. 4 00:00:12,820 --> 00:00:16,760 Josh: That's double what he predicted just six months ago from that very same stage. 5 00:00:16,880 --> 00:00:20,640 Josh: And then he spent the next two hours, this was a very long presentation, 6 00:00:21,100 --> 00:00:24,260 Josh: unveiling why even a trillion dollars is conservative. 7 00:00:24,500 --> 00:00:29,080 Josh: But a lot of people throughout this presentation seem to have missed the actual 8 00:00:29,080 --> 00:00:32,460 Josh: reveal. I think they're focused on a few specific highlights when the reality is 9 00:00:32,800 --> 00:00:36,100 Josh: The things that he presented that are going to yield this trillion dollars are 10 00:00:36,100 --> 00:00:39,040 Josh: probably much different than I think the average person expects. 11 00:00:39,300 --> 00:00:43,160 Josh: Ijaz, I know we were chatting as we were watching this two-hour movie marathon. 12 00:00:43,160 --> 00:00:46,940 Josh: What were your thoughts? Did you make it through? Was it too boring? Was it exciting? 13 00:00:47,220 --> 00:00:49,480 Josh: What were the first impressions of this presentation? 14 00:00:49,820 --> 00:00:54,200 Ejaaz: The thing that excited me the most was the announcement of the DLSS 5, 15 00:00:54,440 --> 00:00:56,620 Ejaaz: which seems to be the most controversial announcement. 16 00:00:57,000 --> 00:01:03,540 Ejaaz: It's this new 3D rendering AI model that basically refactors old games or gaming 17 00:01:03,540 --> 00:01:06,860 Ejaaz: graphics into newer, higher performance graphics. 18 00:01:07,000 --> 00:01:09,880 Ejaaz: So if you're looking on the screen right now, you're seeing a version of a video 19 00:01:09,880 --> 00:01:11,380 Ejaaz: game and then suddenly it's enhanced. 20 00:01:11,540 --> 00:01:15,180 Ejaaz: It's kind of like a Snapchat filter, which I think a lot of the gaming community 21 00:01:15,180 --> 00:01:19,200 Ejaaz: had backlash about. They thought it was just AI slop. They didn't really vibe with it. 22 00:01:19,440 --> 00:01:22,580 Ejaaz: But in my opinion, it's actually quite a good product and would make me more 23 00:01:22,580 --> 00:01:24,120 Ejaaz: engaged to play the actual game. 24 00:01:24,300 --> 00:01:27,040 Josh: So it was surprising to see DLSS 5 get the attention that it 25 00:01:27,040 --> 00:01:29,800 Josh: did just because nvidia announced some unbelievable stuff and 26 00:01:29,800 --> 00:01:32,600 Josh: seemingly this was the headline at all the news outlets and it's 27 00:01:32,600 --> 00:01:35,860 Josh: basically an ai upscaler for video games it takes existing 28 00:01:35,860 --> 00:01:38,920 Josh: graphics that are you know pretty decent and upscales them it makes the facial 29 00:01:38,920 --> 00:01:42,240 Josh: features better it increases the dynamic range the highlights the shadows and 30 00:01:42,240 --> 00:01:45,560 Josh: when i saw it i loved it i was like oh this is pretty cool but the internet 31 00:01:45,560 --> 00:01:50,040 Josh: reaction to this was far from what mine was i mean if you're looking at the 32 00:01:50,040 --> 00:01:53,080 Josh: video on screen it increases the facial features if you're And now looking at 33 00:01:53,080 --> 00:01:55,620 Josh: the meme on screen, that was the public perception. 34 00:01:55,620 --> 00:02:00,480 Josh: It was also negative for such a small feature that Jensen dropped in a two-hour presentation. 35 00:02:00,740 --> 00:02:03,580 Josh: So EJs, do you have any idea what's going on with this backlash here, 36 00:02:03,820 --> 00:02:05,040 Josh: particularly around DLSS 5? 37 00:02:05,180 --> 00:02:07,800 Ejaaz: I think the point around the gaming community is they're just very sensitive 38 00:02:07,800 --> 00:02:10,640 Ejaaz: around AI being involved in art. 39 00:02:10,840 --> 00:02:15,140 Ejaaz: And I get it, right? Like some things can be kind of cringe and doesn't seem 40 00:02:15,140 --> 00:02:18,920 Ejaaz: very human. But the point is, like, this just makes graphics of games way, 41 00:02:18,980 --> 00:02:21,220 Ejaaz: way better. I mean, the meme you're showing on the screen isn't the accurate 42 00:02:21,220 --> 00:02:23,280 Ejaaz: representation of what this thing is going to do. 43 00:02:23,460 --> 00:02:28,480 Ejaaz: We had the head of Bethesda Games make a partnership with NVIDIA just for this tool. 44 00:02:28,640 --> 00:02:31,340 Ejaaz: It's going to save him and his team hours and hours of work. 45 00:02:31,440 --> 00:02:34,540 Ejaaz: And I saw someone make a really good point online yesterday, which says, 46 00:02:34,840 --> 00:02:38,940 Ejaaz: If you're a gaming developer that's spending years designing AAA games, 47 00:02:39,340 --> 00:02:44,060 Ejaaz: this not only saves you a bunch of time, but it also helps you realize your artistic vision. 48 00:02:44,600 --> 00:02:48,480 Ejaaz: Usually when you're a game developer, you make sacrifices when you are designing 49 00:02:48,480 --> 00:02:51,700 Ejaaz: a particular character or an asset because you don't have enough money, 50 00:02:51,860 --> 00:02:53,700 Ejaaz: compute, all the tools to be able to do this. 51 00:02:53,800 --> 00:02:57,200 Ejaaz: This should just be seen as another tool to get to your actual vision. 52 00:02:57,340 --> 00:02:58,580 Ejaaz: So I think it's a good reason. 53 00:02:58,740 --> 00:03:02,800 Ejaaz: But there's another reason why this is super cool and everyone missed it, in my opinion. 54 00:03:02,800 --> 00:03:05,660 Ejaaz: This is the exact same technology that you can 55 00:03:05,660 --> 00:03:09,300 Ejaaz: use to create visual learning for robotics 56 00:03:09,300 --> 00:03:12,320 Ejaaz: and for automotive sorry autonomous driving 57 00:03:12,320 --> 00:03:17,580 Ejaaz: cars so this is the same technology that nvidia is using to build out their 58 00:03:17,580 --> 00:03:21,060 Ejaaz: partner program with i think it was byd and a bunch of other car companies which 59 00:03:21,060 --> 00:03:24,040 Ejaaz: we'll talk about in a second as well as being used in their robotics division 60 00:03:24,040 --> 00:03:28,100 Ejaaz: with their group robotics models this is the same tech so i actually think it's 61 00:03:28,100 --> 00:03:31,220 Ejaaz: cool that it's so pervasive and it's entering gaming, but that's not really 62 00:03:31,220 --> 00:03:32,280 Ejaaz: the big story for me here. 63 00:03:32,480 --> 00:03:35,320 Ejaaz: Whether you hate it or you like it, you're not gonna be able to use this thing 64 00:03:35,320 --> 00:03:37,520 Ejaaz: for the mass audience until probably next year. 65 00:03:37,660 --> 00:03:42,180 Ejaaz: This thing runs on like two RTX 5090s, which are very expensive. 66 00:03:42,340 --> 00:03:45,360 Ejaaz: It's not very accessible to the average day-to-day person. 67 00:03:45,680 --> 00:03:49,020 Ejaaz: So by the time it gets released to the mass audience, I think it's gonna be 68 00:03:49,020 --> 00:03:50,260 Ejaaz: a lot better than what we see today. 69 00:03:50,400 --> 00:03:53,500 Josh: So that's the headliner. If you are not paying attention to the headliner, 70 00:03:53,620 --> 00:03:56,720 Josh: there is a lot of other stuff that was announced that is far more interesting 71 00:03:56,720 --> 00:03:59,140 Josh: than this. And we have a lot to unpack, so buckle up. 72 00:03:59,440 --> 00:04:03,300 Josh: Starting with the Vera Rubin platform, which is the big headliner. 73 00:04:03,480 --> 00:04:07,860 Josh: I mean, this is the big boy. This is what was teased previously six months ago, 74 00:04:08,020 --> 00:04:12,220 Josh: I think, when Jensen was announcing his like $500 billion in revenue. 75 00:04:12,400 --> 00:04:13,420 Josh: Now he's up to a trillion. 76 00:04:13,640 --> 00:04:15,560 Josh: He was unveiling a little bit more information about the chip. 77 00:04:15,820 --> 00:04:17,400 Josh: Ejaz, what's new with Vera Rubin? 78 00:04:17,680 --> 00:04:24,660 Ejaaz: Yeah, so the headline metric is it's 35 times more performant than the previous generation. 79 00:04:24,840 --> 00:04:28,800 Ejaaz: For anyone who's been tracking, Typically, a new NVIDIA GPU gets you about a 80 00:04:28,800 --> 00:04:31,280 Ejaaz: 2 to 5x performance upgrade on a good day. 81 00:04:31,600 --> 00:04:37,020 Ejaaz: This is the largest jump overall. And the secret is there are about five to 82 00:04:37,020 --> 00:04:39,240 Ejaaz: seven major components of a GPU. 83 00:04:39,660 --> 00:04:42,940 Ejaaz: Typically, when you improve for the next generation of GPUs, 84 00:04:43,100 --> 00:04:44,820 Ejaaz: you just refactor one of those things. 85 00:04:44,980 --> 00:04:48,320 Ejaaz: Why? Because if you did all of them at once, that's really high risk. 86 00:04:48,560 --> 00:04:52,940 Ejaaz: Anything could go wrong and it results in delays of improving your GPUs. 87 00:04:53,140 --> 00:04:57,020 Ejaaz: Jensen said, forget, I'm just going to do it anyway. And he pulled it off. 88 00:04:57,380 --> 00:05:02,180 Ejaaz: Seven new chips make up this entire new thing and it gets implemented into five 89 00:05:02,180 --> 00:05:06,600 Ejaaz: new racks, creating what he calls on stage an AI supercomputer. 90 00:05:06,840 --> 00:05:09,620 Ejaaz: And that's why you get this massive performance increase. It's just insane. 91 00:05:09,900 --> 00:05:12,980 Josh: These chips are what's running all the AI that we use every single day. 92 00:05:13,140 --> 00:05:14,860 Josh: Previously, everyone was training on Hopper. 93 00:05:15,120 --> 00:05:18,840 Josh: Hopper was the chips that are running a lot of the AI models that you're actually using today. 94 00:05:19,160 --> 00:05:22,440 Josh: The Frontier Labs have just started to spin up the Blackwell models. 95 00:05:22,560 --> 00:05:25,860 Josh: That's what we've seen with Opus 4.6, what we've seen with GPT 5.4. 96 00:05:26,120 --> 00:05:28,920 Josh: That's the Blackwell chip. It takes a long time from these chips to 97 00:05:28,920 --> 00:05:32,200 Josh: be invented to actually roll down to data centers and then train the models what 98 00:05:32,200 --> 00:05:34,960 Josh: we're seeing next and we're not going to actually feel the effects 99 00:05:34,960 --> 00:05:41,000 Josh: of this until probably early next year is vera rubin and vera rubin i mean it's 100 00:05:41,000 --> 00:05:44,660 Josh: a 10 times performance improvement versus blackwell just in terms of performance 101 00:05:44,660 --> 00:05:50,280 Josh: per watt so for every gigawatt of energy that these data centers have this new 102 00:05:50,280 --> 00:05:53,800 Josh: chip is equivalent to 10 gigawatts worth of compute today, 103 00:05:54,260 --> 00:05:58,380 Josh: So for every gigawatt, you get a 10x improvement on intelligence. And that is huge. 104 00:05:58,660 --> 00:06:02,400 Josh: It is an absolutely massive growth because we're planning to scale the gigawatts 105 00:06:02,400 --> 00:06:05,060 Josh: of these data factories pretty significantly by the end of the year. 106 00:06:05,360 --> 00:06:09,160 Josh: Six to seven gigawatts for some of these. That's going to be equivalent to 60 107 00:06:09,160 --> 00:06:11,640 Josh: to 70 gigawatts of intelligence as of today. 108 00:06:11,800 --> 00:06:15,180 Josh: And I think that's pretty important to note is that there is a strong delay 109 00:06:15,180 --> 00:06:17,980 Josh: when it comes to these chips actually being released, actually being implemented 110 00:06:17,980 --> 00:06:19,560 Josh: on the racks, trained and deployed. 111 00:06:19,760 --> 00:06:22,000 Josh: It's hard to imagine we don't get AGI from this. Yeah. 112 00:06:22,000 --> 00:06:26,940 Ejaaz: The other major improvement that they made was a few, I think it was about a 113 00:06:26,940 --> 00:06:31,480 Ejaaz: month and a half ago, NVIDIA acquired, and I do this like this because apparently 114 00:06:31,480 --> 00:06:36,900 Ejaaz: it wasn't a formal acquisition, a company called Grok, spelled G-R-O-Q. 115 00:06:37,500 --> 00:06:41,340 Ejaaz: And the reason why they acquired them is they get the rights to a very special 116 00:06:41,340 --> 00:06:46,580 Ejaaz: type of AI chip called an LPU, which uses something called SRAM, 117 00:06:46,580 --> 00:06:48,460 Ejaaz: static random access memory. 118 00:06:48,740 --> 00:06:52,700 Ejaaz: Now, if you've been keeping up to tabs with the memory walls that are happening 119 00:06:52,700 --> 00:06:55,100 Ejaaz: right now, memory prices have skyrocketed. 120 00:06:55,200 --> 00:06:58,900 Ejaaz: In fact, it's probably going to affect a bunch of major companies releasing 121 00:06:58,900 --> 00:07:02,540 Ejaaz: their own technology devices because the cost of memory is so high, 122 00:07:02,580 --> 00:07:04,840 Ejaaz: so they can't even give it to their customers because otherwise they'll need 123 00:07:04,840 --> 00:07:06,420 Ejaaz: to charge extortion of prices. 124 00:07:06,760 --> 00:07:10,100 Ejaaz: Jensen made a really smart move by acquiring this company and integrating their 125 00:07:10,100 --> 00:07:12,200 Ejaaz: technology into Vera Rubin. 126 00:07:12,300 --> 00:07:17,020 Ejaaz: And so what you're seeing on the screen now is basically the same architecture 127 00:07:17,020 --> 00:07:20,460 Ejaaz: of Vera Rubin, but integrated with this SRAM technology. 128 00:07:20,740 --> 00:07:28,400 Ejaaz: And the resulting effects is you can inference AI models at a much larger scale. 129 00:07:28,400 --> 00:07:30,560 Ejaaz: So that 10x that you just mentioned, Josh, 130 00:07:30,880 --> 00:07:35,260 Ejaaz: partially, a bunch of that is unlocked by these new LPUs. 131 00:07:35,440 --> 00:07:39,000 Ejaaz: So we're now starting to see Jensen take two things more seriously. 132 00:07:39,240 --> 00:07:43,300 Ejaaz: One, a different type of chip architecture. Usually, NVIDIA is known for generalized 133 00:07:43,300 --> 00:07:45,160 Ejaaz: GPUs, and that's where their bread and butter is. 134 00:07:45,280 --> 00:07:49,520 Ejaaz: Now we see him branching off into these hyperspecific inference chips because 135 00:07:49,520 --> 00:07:53,660 Ejaaz: he looks over his shoulder and he sees, not close, but kind of far back, 136 00:07:54,080 --> 00:07:59,680 Ejaaz: Google's TPUs looming, AMD's chips, and Intel's CPUs and chips coming up behind them as well. 137 00:07:59,880 --> 00:08:03,400 Ejaaz: And they're all specializing in inference-specific chips. And the argument or 138 00:08:03,400 --> 00:08:07,740 Ejaaz: the reason behind that is a lot of the world isn't going to be focused on training 139 00:08:07,740 --> 00:08:10,520 Ejaaz: AI models. It's going to be prompting and querying AI models. 140 00:08:10,620 --> 00:08:12,220 Ejaaz: And that's going to grow exponentially more. 141 00:08:12,380 --> 00:08:17,020 Ejaaz: So this is NVIDIA and Jensen basically saying, we're going to make a mark here. 142 00:08:17,120 --> 00:08:19,080 Ejaaz: This is our stand. This is why we acquired Grok. 143 00:08:19,320 --> 00:08:21,980 Ejaaz: And here's the chip that we're doing. And VeroRubin is going to be that chip 144 00:08:21,980 --> 00:08:24,500 Ejaaz: for anything and everything, general purpose and inference. 145 00:08:24,760 --> 00:08:27,780 Josh: When I think about these chips and just project it out to the future it's 146 00:08:27,780 --> 00:08:30,560 Josh: so exciting because there's such a 147 00:08:30,560 --> 00:08:33,480 Josh: clear path to going to where i think every ai lab 148 00:08:33,480 --> 00:08:36,300 Josh: wants to go yes getting to that agi level and beyond and 149 00:08:36,300 --> 00:08:39,640 Josh: this chart that we're showing on screen here is a beautiful example of this 150 00:08:39,640 --> 00:08:43,620 Josh: because in addition to blackwell in addition to rubin they also teased fineman 151 00:08:43,620 --> 00:08:48,100 Josh: already even though rubin is months to years away from actually being deployed 152 00:08:48,100 --> 00:08:53,780 Josh: at scale so nvidia is essentially 18 months give or take a few ahead of what 153 00:08:53,780 --> 00:08:57,320 Josh: the current reality looks like. And I think this is really important to note. 154 00:08:57,900 --> 00:09:02,360 Josh: Is currently with the bleeding edge of AI, we're running Blackwell right now. 155 00:09:02,500 --> 00:09:03,980 Josh: And we just started running Blackwell. 156 00:09:04,140 --> 00:09:08,320 Josh: And Blackwell has about 12 months of improvements to be made before we start 157 00:09:08,320 --> 00:09:09,500 Josh: to feel the effects of Rubin. 158 00:09:09,700 --> 00:09:13,040 Josh: By the time we feel the effects of Rubin, which is that 10x performance per 159 00:09:13,040 --> 00:09:17,900 Josh: watt improvement, they already have Feynman ready to go and to be deployed into these data centers. 160 00:09:18,020 --> 00:09:23,580 Josh: And already we have two incremental steps, two exponential steps ahead of where we currently sit. 161 00:09:23,700 --> 00:09:26,720 Josh: And it's hard to imagine that with the build-out that's happening, 162 00:09:26,720 --> 00:09:29,540 Josh: with the performance per watt increase that we're seeing from all these chipsets, 163 00:09:29,700 --> 00:09:30,780 Josh: that we're not just going to have 164 00:09:30,780 --> 00:09:33,760 Josh: this completely vertical and exponential growth of AI across the board. 165 00:09:33,840 --> 00:09:37,580 Josh: And I think that's probably at the core of Jensen's thesis of a trillion dollars 166 00:09:37,580 --> 00:09:41,640 Josh: is like the spending isn't going to stop because he's already created the future. 167 00:09:41,780 --> 00:09:45,180 Josh: It's just a matter of actually deploying it and plugging it into the grid so 168 00:09:45,180 --> 00:09:48,300 Josh: you could power these chips and get the intelligence that everyone wants. And it's unbelievable. 169 00:09:48,520 --> 00:09:51,340 Josh: So Feynman is coming. They didn't announce a bunch of things about Feynman, 170 00:09:51,420 --> 00:09:53,600 Josh: but that's the name of the next chip architecture. 171 00:09:54,350 --> 00:09:57,370 Josh: Named after your favorite mathematicians favorite mathematician richard feinman 172 00:09:57,370 --> 00:09:59,850 Josh: everyone's a big fan of him very cool very excited. 173 00:09:59,850 --> 00:10:01,290 Ejaaz: His book or something i 174 00:10:01,290 --> 00:10:05,550 Josh: Did yep i'm surely joking mr feinman he has a few books that are all awesome 175 00:10:05,550 --> 00:10:09,570 Josh: so if you're into physics or math or just really admire great teachers richard 176 00:10:09,570 --> 00:10:14,930 Josh: feinman is amazing and is now the naming architecture for the future of nvidia's 177 00:10:14,930 --> 00:10:17,190 Josh: ai chips so pretty cool stuff bold. 178 00:10:17,190 --> 00:10:22,330 Ejaaz: Name big ambitions um nvidia currently sits at, what, $4.5 trillion? 179 00:10:23,110 --> 00:10:28,310 Ejaaz: Biggest, most valuable company in the world. Odds that it's the same by the end of the year? 180 00:10:28,670 --> 00:10:32,390 Josh: Is this going to be prolonged? Well, we can ask our friends at Polymarket to 181 00:10:32,390 --> 00:10:36,590 Josh: answer this for us. And it looks like there has been a strong trend signaling yes. 182 00:10:37,530 --> 00:10:40,710 Josh: And this was not always the case. I mean, it looks like Alphabet, 183 00:10:40,910 --> 00:10:43,670 Josh: Google, was at one point during the year, February, just a month ago, 184 00:10:43,970 --> 00:10:45,070 Josh: was projected to flip them. 185 00:10:45,390 --> 00:10:48,390 Josh: People thought Google was going to be the world leader. it is 186 00:10:48,390 --> 00:10:51,210 Josh: clear now that is absolutely not the case in fact 187 00:10:51,210 --> 00:10:53,970 Josh: apple who we frequently talk about looks like they have a better chance 188 00:10:53,970 --> 00:10:57,610 Josh: of doing it than google now and now nvidia is up to 70 so it 189 00:10:57,610 --> 00:11:00,530 Josh: seems highly probable that people saw this presentation people have been seeing 190 00:11:00,530 --> 00:11:05,230 Josh: progress and they are very much bullish on nvidia so the market is pricing in 191 00:11:05,230 --> 00:11:08,370 Josh: a pretty steep increase to the stock price before the end of the month i mean 192 00:11:08,370 --> 00:11:13,030 Josh: it's currently trading at 182 and it looks like there's what 25 30 it's about 193 00:11:13,030 --> 00:11:17,110 Josh: a 30 chance that it trades over 200 this month so it looks like, 194 00:11:17,680 --> 00:11:20,800 Josh: Things are looking good for NVIDIA for being the most valuable company in the 195 00:11:20,800 --> 00:11:24,300 Josh: world and also continuing to trade up on this news. It was an incredible presentation. 196 00:11:24,620 --> 00:11:27,360 Josh: Thank you to Polymarket for sponsoring this segment of the episode. 197 00:11:27,480 --> 00:11:31,380 Josh: And now we could probably get into the next most interesting thing for me, 198 00:11:31,420 --> 00:11:33,660 Josh: at least, which was the full self-driving moment. 199 00:11:33,840 --> 00:11:39,260 Josh: In fact, Jensen Huang said, this is the chat GPT moment for self-driving cars. It has arrived. 200 00:11:39,440 --> 00:11:43,600 Josh: This is a bold take because the full self-driving industry is pretty, 201 00:11:43,600 --> 00:11:45,440 Josh: pretty vicious. Hasn't it been. 202 00:11:45,440 --> 00:11:47,460 Ejaaz: Solved by tesla already at this 203 00:11:47,460 --> 00:11:50,440 Josh: Point well it depends who you ask it sounds like internally 204 00:11:50,440 --> 00:11:53,300 Josh: they feel confident in the fact they've solved it but they're currently solving 205 00:11:53,300 --> 00:11:57,300 Josh: this march of nines where they have efficacy up to 99.x percent 206 00:11:57,300 --> 00:12:02,920 Josh: and they need to get it to 9999 now waymo clearly has the most deployed version 207 00:12:02,920 --> 00:12:05,560 Josh: of this you could actually go and you could get into a waymo you can get into 208 00:12:05,560 --> 00:12:09,300 Josh: a cyber cab in some places in austin but they still have the kind of guiding 209 00:12:09,300 --> 00:12:12,160 Josh: drivers they haven't figured out the legislation to let them be fully autonomous 210 00:12:12,160 --> 00:12:15,040 Josh: but jensen is saying hey if you're not Waymo, 211 00:12:15,280 --> 00:12:17,540 Josh: if you're not Tesla, we have a solution for you. 212 00:12:17,660 --> 00:12:20,920 Josh: We are actually going to build the full self-driving stack and integrate it 213 00:12:20,920 --> 00:12:24,520 Josh: directly into your cars for you from the sensors all the way to the software stack. 214 00:12:24,660 --> 00:12:28,760 Josh: And they just recently partnered with BYD, Nissan, Hyundai, and Gili. 215 00:12:28,900 --> 00:12:32,320 Josh: And for those who aren't aware, BYD is actually the largest electric car manufacturer 216 00:12:32,320 --> 00:12:34,420 Josh: in the world, more so than Tesla. 217 00:12:34,620 --> 00:12:38,800 Josh: They're based in China and it's showing that NVIDIA is not really country. 218 00:12:39,040 --> 00:12:41,040 Josh: They're kind of country agnostic, right? Like they're just going to, 219 00:12:41,240 --> 00:12:44,020 Josh: if you want a self-driving car, come to us. We got you. 220 00:12:44,340 --> 00:12:47,840 Ejaaz: NVIDIA or Jensen just doesn't care if he aligns with China or not. 221 00:12:48,180 --> 00:12:51,880 Ejaaz: He's just like out there to expand NVIDIA into anyone and everyone's hands. 222 00:12:52,120 --> 00:12:57,420 Ejaaz: As you said, BYD is the biggest EV maker. They sell more cars than Tesla every single year. 223 00:12:57,860 --> 00:13:01,700 Ejaaz: And so that distribution, like think about that. Like imagine you put your self-driving 224 00:13:01,700 --> 00:13:03,440 Ejaaz: model into as many cars as possible. 225 00:13:03,600 --> 00:13:07,120 Ejaaz: It's probably going to get smarter way, way quicker because it's just inside 226 00:13:07,120 --> 00:13:11,760 Ejaaz: more So that's real competition against Tesla from a competitive mode. 227 00:13:12,120 --> 00:13:15,300 Ejaaz: The other thing is he's also integrating into Uber as well, right? 228 00:13:15,380 --> 00:13:20,200 Ejaaz: So it's going to be launching in 28 cities by 2028. 229 00:13:20,400 --> 00:13:23,780 Ejaaz: So through the end of next year, which seems like a long time, 230 00:13:23,900 --> 00:13:24,660 Ejaaz: but that's a lot of cities. 231 00:13:24,840 --> 00:13:28,680 Ejaaz: And Uber has a lot of reach when it comes to just a driving network in general. 232 00:13:28,940 --> 00:13:33,500 Ejaaz: So this is a really cool announcement. I don't quite know if it's apples to 233 00:13:33,500 --> 00:13:35,320 Ejaaz: apples with Tesla full self-driving. 234 00:13:35,780 --> 00:13:39,400 Ejaaz: They own the end-to-end stack there. NVIDIA doesn't really have that. 235 00:13:39,400 --> 00:13:41,940 Ejaaz: This is more of a thing that you can kind of attach onto cars. 236 00:13:42,080 --> 00:13:45,540 Ejaaz: And if I had to guess, this is not just me being an Elon fanboy. 237 00:13:45,780 --> 00:13:48,880 Ejaaz: There's a lot more friction that NVIDIA will run into. So I don't think this 238 00:13:48,880 --> 00:13:50,640 Ejaaz: is a direct one-to-one competitor. 239 00:13:51,000 --> 00:13:54,580 Josh: This is a key difference. If I'm a Tesla shareholder, I'm not really nervous 240 00:13:54,580 --> 00:13:58,520 Josh: about this because like you said, Tesla owns the full manufacturing stack and 241 00:13:58,520 --> 00:14:01,940 Josh: they have millions of cars on the road that are full self-driving capable today. 242 00:14:02,260 --> 00:14:05,200 Josh: They're just one software update away from cracking that. When that 243 00:14:05,200 --> 00:14:08,060 Josh: final software update comes when the legislation passes that is 244 00:14:08,060 --> 00:14:10,980 Josh: to be determined but they're there they're ready waymo and 245 00:14:10,980 --> 00:14:13,840 Josh: i guess uber now are kind of on the other side of this where they they've kind 246 00:14:13,840 --> 00:14:16,980 Josh: of perhaps figured out the software stack they're close at least but 247 00:14:16,980 --> 00:14:19,900 Josh: they they have nowhere near figured out the manufacturing stack 248 00:14:19,900 --> 00:14:23,220 Josh: for this at scale and manufacturing as we know designing hard 249 00:14:23,220 --> 00:14:26,080 Josh: things in the physical world is hard and that's going 250 00:14:26,080 --> 00:14:28,860 Josh: to slow these companies down a lot so i think for uber this is probably 251 00:14:28,860 --> 00:14:31,800 Josh: the best case scenario they finally have a saving grace someone 252 00:14:31,800 --> 00:14:34,720 Josh: who wants to actually work with them to help deploy the full self-driving vision 253 00:14:34,720 --> 00:14:38,540 Josh: um but they got a long way to go so it's nice that they're trying this is kind 254 00:14:38,540 --> 00:14:42,140 Josh: of like apple car play but for full self-driving where they're not going to 255 00:14:42,140 --> 00:14:44,360 Josh: make the cars they're going to sell you the software to put in the cars and 256 00:14:44,360 --> 00:14:48,280 Josh: hopefully one day make them full self-driving so we'll see how that goes that 257 00:14:48,280 --> 00:14:52,120 Josh: was the first of the robotics section of this episode let me. 258 00:14:52,120 --> 00:14:55,780 Ejaaz: Let me introduce you to the unhinged uh version of this josh 259 00:14:56,160 --> 00:15:02,220 Ejaaz: So Olaf, made popular in the fictional movie Frozen, came to life on stage. 260 00:15:02,580 --> 00:15:04,260 Ejaaz: What you're looking at is an autonomous, 261 00:15:04,760 --> 00:15:09,540 Ejaaz: self-directed robot that runs on NVIDIA. I'm not making this up. 262 00:15:09,620 --> 00:15:12,640 Ejaaz: That runs on NVIDIA's Newton robotics engine. 263 00:15:12,760 --> 00:15:17,760 Ejaaz: It also runs on their Jetson chip as well. So what you're looking at is a homegrown 264 00:15:17,760 --> 00:15:22,100 Ejaaz: NVIDIA robot and product that is autonomously interacting with Jensen. 265 00:15:22,420 --> 00:15:26,220 Ejaaz: I can't help but think that some of this must be scripted. there's no way that 266 00:15:26,220 --> 00:15:27,360 Ejaaz: the robot is this interactive. 267 00:15:27,820 --> 00:15:31,860 Ejaaz: And obviously, it's like been outfitted with the look of this Frozen character, 268 00:15:32,000 --> 00:15:33,400 Ejaaz: but pretty cool all around. 269 00:15:33,600 --> 00:15:36,840 Ejaaz: I don't know if this is going to be in everyone's home. I like I don't know 270 00:15:36,840 --> 00:15:39,880 Ejaaz: what the point of this was. Like, maybe they're going to sell rights to Disney or something. 271 00:15:40,020 --> 00:15:42,200 Ejaaz: But yeah, like, I don't really have a strong take on this. 272 00:15:42,740 --> 00:15:46,040 Josh: Yeah, well, it's just I mean, it's more of the direction that they're heading 273 00:15:46,040 --> 00:15:49,420 Josh: towards, which is real world physical AI, right? It's like we're in self driving 274 00:15:49,420 --> 00:15:50,660 Josh: cars. Now we're going to get robots. 275 00:15:50,900 --> 00:15:53,780 Josh: They're creating these small packaged computers to 276 00:15:53,780 --> 00:15:56,900 Josh: put into these things are creating the entire stack nvidia is 277 00:15:56,900 --> 00:16:00,220 Josh: becoming the tesla for the general purpose company 278 00:16:00,220 --> 00:16:03,120 Josh: it's like if you can't build it all yourself nvidia has done 279 00:16:03,120 --> 00:16:06,260 Josh: it and they will sell you all their hardware they'll sell you all their software 280 00:16:06,260 --> 00:16:09,620 Josh: they're moving a lot into open source and i guess that's probably the transition 281 00:16:09,620 --> 00:16:13,140 Josh: to the next announcement which is their nemo claw announcement the open claw 282 00:16:13,140 --> 00:16:17,980 Josh: competitor which isn't an open claw competitor at all actually it's just a basically 283 00:16:17,980 --> 00:16:21,840 Josh: enterprise solution for companies that want to use OpenClaw. 284 00:16:22,220 --> 00:16:24,080 Josh: So the founder of OpenClaw, he was there. 285 00:16:24,740 --> 00:16:27,800 Josh: Jensen gave him a nice shout-out. And basically, NemoClaw is... 286 00:16:28,330 --> 00:16:33,590 Josh: A way for companies to deploy OpenClaw in a more secure way and to run on any 287 00:16:33,590 --> 00:16:35,110 Josh: coding agent and deploy from anywhere. 288 00:16:35,430 --> 00:16:38,330 Josh: And I think a lot of people, I mean, ourselves included, thought this could 289 00:16:38,330 --> 00:16:40,590 Josh: be competition. The reality is it's complementary. 290 00:16:41,070 --> 00:16:44,210 Josh: NVIDIA wants open source AI because they want to build the hardware that you 291 00:16:44,210 --> 00:16:45,790 Josh: use to run the open source AI. 292 00:16:46,030 --> 00:16:49,190 Josh: And it seems like this was kind of like a win for everyone, including the open 293 00:16:49,190 --> 00:16:50,390 Josh: source community. It's pretty cool. 294 00:16:50,890 --> 00:16:54,050 Ejaaz: Yeah. I mean, Pete Stuy, as you mentioned, the founder of OpenClaw, 295 00:16:54,270 --> 00:16:57,730 Ejaaz: actually worked with Jensen and the NVIDIA team for months to build this out. 296 00:16:58,310 --> 00:17:03,550 Ejaaz: Their target market are enterprise customers specifically because when OpenCore 297 00:17:03,550 --> 00:17:07,370 Ejaaz: went viral, it went viral because everyone could spin up their own personal agent. 298 00:17:07,710 --> 00:17:11,910 Ejaaz: There was one glaring issue, loads of security issues. So people could lose 299 00:17:11,910 --> 00:17:16,010 Ejaaz: money, expose their credit card details or lose all their personal data or have 300 00:17:16,010 --> 00:17:17,230 Ejaaz: people hack their computers. 301 00:17:17,710 --> 00:17:22,590 Ejaaz: Not good if you are an enterprise company, but companies still want to get access to this thing. 302 00:17:22,730 --> 00:17:27,170 Ejaaz: So Jensen kind of dreamt up this platform that sits on top of open call so it 303 00:17:27,170 --> 00:17:30,430 Ejaaz: works very synonymously with it and now you can kind of use open call without 304 00:17:30,430 --> 00:17:34,170 Ejaaz: any worry you can spin up an agent that does a particular enterprise workflow 305 00:17:34,170 --> 00:17:37,510 Ejaaz: or you can use it for accounting back office stuff whatever you can dream of 306 00:17:37,510 --> 00:17:40,910 Ejaaz: it's now safe to use and it's open source which is great yeah 307 00:17:40,910 --> 00:17:44,790 Josh: Okay so two more things we have two more quick announcements one ai and space 308 00:17:44,790 --> 00:17:47,590 Josh: space gpus we're doing the damn thing it's happening, 309 00:17:48,050 --> 00:17:52,730 Josh: So Jensen got on stage. He said, we are going to build Verirubin for space. 310 00:17:53,070 --> 00:17:55,950 Josh: You're going to have Verirubin orbiting the Earth. It's going to be in these 311 00:17:55,950 --> 00:17:57,450 Josh: data centers. It's going to be fantastic. 312 00:17:57,690 --> 00:18:00,930 Josh: And then he says, well, we're not quite sure how we're going to do it, but we're going to do it. 313 00:18:01,030 --> 00:18:03,670 Josh: They still have a lot of issues that they need to solve, one of which is the 314 00:18:03,670 --> 00:18:05,890 Josh: cooling, one of which is solving the radiation. 315 00:18:06,330 --> 00:18:10,190 Josh: There are a series of issues that are going to need to be solved, 316 00:18:10,310 --> 00:18:11,910 Josh: but there is the intention to do this. 317 00:18:12,070 --> 00:18:14,930 Josh: And I suspect he didn't announce it here, but I suspect they're working with 318 00:18:14,930 --> 00:18:18,550 Josh: SpaceX to design these chips hand to hand because that's really the only company 319 00:18:18,550 --> 00:18:20,030 Josh: that's going to be getting these things up in space. 320 00:18:20,250 --> 00:18:23,170 Josh: And I think it's really exciting when we think about AI data centers in space 321 00:18:23,170 --> 00:18:27,290 Josh: and the quality of the Vera Rubin chip architecture, bringing those two things 322 00:18:27,290 --> 00:18:31,550 Josh: together and getting them in orbit by 2027, 2028, maybe the latest, 323 00:18:31,810 --> 00:18:33,790 Josh: that's going to be pretty cool. That's going to change the game. 324 00:18:34,190 --> 00:18:37,810 Josh: Elon is incredibly bullish on this. He thinks that SpaceX is now going to flip 325 00:18:37,810 --> 00:18:40,770 Josh: every company in the world when it comes to AI development. 326 00:18:40,850 --> 00:18:45,190 Josh: And he might not be wrong because if he can get these chips at scale from Jensen, 327 00:18:45,650 --> 00:18:48,830 Josh: send them up into orbit, lower the cost per watt to be, I mean, 328 00:18:49,030 --> 00:18:52,610 Josh: a marginal of a small fraction of what it is today. It's a huge upgrade. 329 00:18:52,770 --> 00:18:56,910 Ejaaz: I feel like this was just a custom announcement for Elon Musk for one individual. 330 00:18:57,110 --> 00:19:01,330 Ejaaz: He's the only guy that's really trying to launch GPUs into space at scale. 331 00:19:01,530 --> 00:19:07,050 Ejaaz: Like in this demo, he's demoing it using one of NVIDIA's investment portfolio 332 00:19:07,050 --> 00:19:12,790 Ejaaz: companies, StarCloud, which are kind of the initial startup that made GPUs in space a trend, a thing. 333 00:19:12,930 --> 00:19:15,810 Ejaaz: But then elon jumped on the wave and like completely took it over and he's the 334 00:19:15,810 --> 00:19:19,790 Ejaaz: guy that's actually going to be economically able to launch these at scale um 335 00:19:19,790 --> 00:19:24,910 Ejaaz: so it's it's it's a good day to be a tesla or spacex uh share owner equity owner 336 00:19:24,910 --> 00:19:28,870 Josh: And the final announcement that we're going to talk about is the dgx spark they 337 00:19:28,870 --> 00:19:33,550 Josh: released the new spark and it's now looking like it's going to be priced around 338 00:19:33,550 --> 00:19:38,450 Josh: forty seven hundred dollars which seems high but if you are someone who runs 339 00:19:38,450 --> 00:19:42,370 Josh: local inference at your home and you're considering buying a Mac Studio or something 340 00:19:42,370 --> 00:19:43,530 Josh: to run these tokens on your own. 341 00:19:43,650 --> 00:19:45,990 Josh: Perhaps you have an OpenClaw instance you want to run local AI. 342 00:19:46,650 --> 00:19:51,130 Josh: This is a pretty compelling option. They're basically taking a GB300, 343 00:19:51,130 --> 00:19:54,890 Josh: which is the Grace Blackwell chip, and they're turning it into a tiny little 344 00:19:54,890 --> 00:19:56,050 Josh: thing that fits on your desk. 345 00:19:56,170 --> 00:20:02,290 Josh: That's 750 gigabytes of coherent memory and 20 petaflops of AI compute, 346 00:20:02,510 --> 00:20:06,830 Josh: which allows you to run models up to a trillion parameters right from your desk. 347 00:20:06,830 --> 00:20:10,150 Josh: So it's an unbelievably dense machine. 348 00:20:10,330 --> 00:20:14,570 Josh: In fact, if this was released probably even five years ago, this probably would 349 00:20:14,570 --> 00:20:16,330 Josh: have been the most powerful supercomputer in the world. 350 00:20:16,650 --> 00:20:18,630 Josh: And now it's compressed down to something that fits on your desk. 351 00:20:18,770 --> 00:20:24,050 Josh: So it's just, it's a testament to how much efficiency improvements have been 352 00:20:24,050 --> 00:20:27,850 Josh: made every single year and how powerful the NVIDIA brand is, 353 00:20:27,950 --> 00:20:30,030 Josh: man. There's no one else building stuff like this. 354 00:20:30,270 --> 00:20:34,290 Josh: No one's even close. Looking at this holistically, this is a home run for NVIDIA, 355 00:20:34,490 --> 00:20:36,570 Josh: for shareholders, for investors, for the AI industry. 356 00:20:36,930 --> 00:20:40,170 Josh: Everyone wins because NVIDIA is just running full tilts. 357 00:20:40,290 --> 00:20:43,490 Ejaaz: It's funny, you said that back in the day, you would, this would be so much 358 00:20:43,490 --> 00:20:48,370 Ejaaz: more expensive. So you would also need like a dedicated like server room to fit this entire thing. 359 00:20:48,610 --> 00:20:51,490 Ejaaz: And now you can just sit it on your desk next to your laptop and have, 360 00:20:51,610 --> 00:20:54,890 Ejaaz: in Jensen's words, an AI supercomputer in your house. 361 00:20:55,110 --> 00:20:57,970 Ejaaz: Super cool. It comes shipped with NemoClaw as well. 362 00:20:58,070 --> 00:21:03,310 Ejaaz: So you get two products, two NVIDIA GTC 2026 announcements for the price of one. 363 00:21:03,450 --> 00:21:05,550 Ejaaz: And you said it was $4,700, Josh? 364 00:21:05,830 --> 00:21:08,410 Josh: That's super cheap. That's what it's looking like on their website right now. 365 00:21:08,410 --> 00:21:10,230 Ejaaz: I think that's for the Spark. I think that's for the Spark. 366 00:21:10,430 --> 00:21:11,290 Josh: That is for the Spark. 367 00:21:11,470 --> 00:21:14,190 Ejaaz: That's for the Spark, yeah. And then they had a separate announcement i think 368 00:21:14,190 --> 00:21:19,630 Ejaaz: on the dgx station which is like more the more powerful supercomputer which 369 00:21:19,630 --> 00:21:23,810 Ejaaz: also consequently sits on your desk as well so just two different price points 370 00:21:23,810 --> 00:21:27,330 Ejaaz: but two very powerful things um yeah just a home run for nvidia yeah 371 00:21:27,330 --> 00:21:31,170 Josh: What a great presentation that is everything those the highlights i would love 372 00:21:31,170 --> 00:21:35,310 Josh: for you to share which part you are most excited about is it DLSS 5 how many 373 00:21:35,310 --> 00:21:37,530 Josh: gamers are here that actually care. 374 00:21:37,530 --> 00:21:38,390 Ejaaz: About do you hate it 375 00:21:38,880 --> 00:21:42,060 Josh: Tell me. Because I don't. I think it's cool. I mean, I could understand why 376 00:21:42,060 --> 00:21:45,620 Josh: the artists maybe don't like their art being, you know, digitally enhanced. 377 00:21:45,920 --> 00:21:48,240 Josh: But I have good news. You could just turn it off. 378 00:21:48,560 --> 00:21:51,200 Josh: Like you could just play the vanilla game too. That's also cool. 379 00:21:51,380 --> 00:21:54,040 Josh: So I'd love to know what people are most excited about here. 380 00:21:54,140 --> 00:21:55,940 Josh: I think for me, space data centers, 381 00:21:56,040 --> 00:21:58,620 Josh: man, that's my favorite thing in the world. I want to see AI in space. 382 00:21:58,720 --> 00:22:02,940 Ejaaz: It's got to be the DLSS 5, but used for robotics. 383 00:22:03,260 --> 00:22:06,900 Ejaaz: Like, okay, I'm like nerding out over robotics right now, because I think they're 384 00:22:06,900 --> 00:22:11,340 Ejaaz: going to have their chat GBT from 2022 moment at any point this year. 385 00:22:11,520 --> 00:22:15,120 Ejaaz: They're getting good enough to move around, run, lift heavy items. 386 00:22:15,560 --> 00:22:19,300 Ejaaz: We just need a good model. And I think having something like DLSS, 387 00:22:19,380 --> 00:22:25,780 Ejaaz: that's this code name is, so DLSS 5, kind of like expand robotics models is really exciting for me. 388 00:22:26,020 --> 00:22:28,760 Josh: So yeah. And one thing's for sure, the naming of all of these is going to continue 389 00:22:28,760 --> 00:22:29,800 Josh: to be absolutely horrific. 390 00:22:30,020 --> 00:22:30,560 Ejaaz: Yeah, please come up with easier names. 391 00:22:30,560 --> 00:22:35,360 Josh: Oh my god. But yeah, that's a wrap. Thank you so much for watching this recap 392 00:22:35,360 --> 00:22:37,320 Josh: on NVIDIA's GTC. I hope you enjoyed it. 393 00:22:37,480 --> 00:22:41,640 Josh: We pursued through two hours of pretty boring presentation to bring this to 394 00:22:41,640 --> 00:22:43,960 Josh: you. So hopefully it was a little more interesting, a little more exciting. 395 00:22:44,240 --> 00:22:48,600 Josh: Very technical. Jensen's a technical guy. I hope you enjoyed Ejaz's leather 396 00:22:48,600 --> 00:22:52,040 Josh: jacket that he's rocking today in honor of Jensen and NVIDIA, 397 00:22:52,120 --> 00:22:54,140 Josh: who donned a leather jacket on stage. 398 00:22:54,340 --> 00:22:56,180 Ejaaz: I know you are. I hope you appreciate it. 399 00:22:56,920 --> 00:22:59,200 Josh: But yeah, as always, please don't forget to share with your friends. 400 00:22:59,600 --> 00:23:04,540 Josh: Like the video subscribe leave a comment rate us five stars all the great things 401 00:23:04,540 --> 00:23:08,340 Josh: thank you so much for watching and yeah we'll see you guys in the next episode.