1 00:00:10,280 --> 00:00:14,360 Welcome to the Cerebral Valley Podcast on newcomer I'm Back 2 00:00:14,360 --> 00:00:17,680 with Max Child and James Wilsterman, my Cerebral Valley 3 00:00:17,680 --> 00:00:19,600 Co host and the Volley Co founders. 4 00:00:20,640 --> 00:00:22,680 We are very excited about this episode. 5 00:00:22,680 --> 00:00:29,080 The meat of it will be a draft of the most valuable AI 6 00:00:29,080 --> 00:00:32,640 companies we get. It is like, it's as close as I 7 00:00:32,640 --> 00:00:35,960 think in Silicon Valley you can get to watching sports hopefully 8 00:00:35,960 --> 00:00:37,360 or fan of. Sports. 9 00:00:37,520 --> 00:00:39,880 Sports debate TV show. Yeah, exactly. 10 00:00:42,400 --> 00:00:47,360 Yeah, I I overcomplicated it to make sure I got the. 11 00:00:47,680 --> 00:00:51,840 I won't spoil it anyway, you know we took it very seriously 12 00:00:51,880 --> 00:00:55,480 and even if all our listeners forget about it, I am sure. 13 00:00:55,840 --> 00:00:58,320 James, what was the time window? It's five, is it? 14 00:00:58,680 --> 00:01:01,800 Five years, Yeah. In five years we'll all be 15 00:01:02,040 --> 00:01:07,080 checking to see see how we did. But before that Cerebral Valley, 16 00:01:07,080 --> 00:01:13,480 the conference is on Wednesday. And you know, obviously for the 17 00:01:13,480 --> 00:01:16,040 people going in person, you, you get why you're curious. 18 00:01:16,040 --> 00:01:20,880 It's like OK, But then for everybody else, the videos from 19 00:01:20,880 --> 00:01:23,920 the conference will be posted on YouTube and in the 20 00:01:23,920 --> 00:01:27,000 newsletternewcomer.co. So this conference will be 21 00:01:27,000 --> 00:01:29,840 brought to you digitally. We have not just hyped up a an 22 00:01:29,840 --> 00:01:33,080 event you will not get to consume, you will get to see the 23 00:01:33,080 --> 00:01:35,360 talks. So look forward to that. 24 00:01:35,360 --> 00:01:39,200 I think usually we put some of the best ones in the podcast 25 00:01:39,200 --> 00:01:44,720 feed and we just wanted to sort of take stock going into the 26 00:01:44,720 --> 00:01:49,840 conference sort of big themes that that are going to be on our 27 00:01:49,840 --> 00:01:54,920 mind and then what's changed from the first Cerebral Valley 28 00:01:55,520 --> 00:01:58,440 in March. So yeah, let's get to it. 29 00:01:58,520 --> 00:02:00,640 All right. I said you guys could could go 30 00:02:00,640 --> 00:02:05,000 for themes last, I mean go for themes 1st and I would play 31 00:02:05,000 --> 00:02:08,919 clean up so. James, why don't you go still 32 00:02:08,919 --> 00:02:13,560 thinking about my themes? Yeah, I have noticed a theme 33 00:02:13,840 --> 00:02:18,240 kind of gaining some heat over the last few weeks on Twitter. 34 00:02:18,240 --> 00:02:23,680 Slash X of Kind of this big debate around open source 35 00:02:23,680 --> 00:02:28,120 versus. Close source in terms of safety, 36 00:02:28,120 --> 00:02:30,400 right. I think this ties in with the 37 00:02:30,400 --> 00:02:36,080 executive order put out by the Biden administration and kind of 38 00:02:36,080 --> 00:02:39,960 a raging debate I guess or simmering debate of whether 39 00:02:40,920 --> 00:02:45,240 these larger close source language model companies are in 40 00:02:45,240 --> 00:02:49,160 the business of regulatory capture through their efforts to 41 00:02:50,360 --> 00:02:54,000 speak around model safety and get. 42 00:02:55,280 --> 00:02:59,200 Get the Biden administration involved or whether that comes 43 00:02:59,200 --> 00:03:05,000 from a true concern and how that impacts companies who are 44 00:03:05,000 --> 00:03:07,520 putting out open source models for the community. 45 00:03:07,680 --> 00:03:10,160 OK, I'm going to jump on that. Open source versus closed 46 00:03:10,160 --> 00:03:11,760 source. I I think that's probably the 47 00:03:11,760 --> 00:03:14,760 top in my mind. And in some ways I'd almost 48 00:03:14,760 --> 00:03:18,040 frame it as like I think the first conference ended up having 49 00:03:18,040 --> 00:03:22,120 more of the pro open source people. 50 00:03:22,120 --> 00:03:25,640 You know we ended with Hugging face and Replet. 51 00:03:25,640 --> 00:03:28,560 Both of them are very pro open source. 52 00:03:29,400 --> 00:03:33,600 I mean we have data bricks back they're pretty pro open source. 53 00:03:33,880 --> 00:03:38,640 Whereas this conference we're ending the day with Vinod Khosla 54 00:03:38,920 --> 00:03:43,560 who has been somewhat critical of he he wants more sort of 55 00:03:43,560 --> 00:03:46,720 control and regulation. We'll we'll get the exact nuance 56 00:03:46,720 --> 00:03:51,120 of his open source take and then Mustafa Suleiman at inflection 57 00:03:51,120 --> 00:03:54,200 and the D Mine Co founder who's going to end the day has been 58 00:03:54,200 --> 00:03:58,640 super supportive of regulation and somewhat apprehensive about 59 00:03:58,680 --> 00:04:00,960 open source. So I think this conference in 60 00:04:00,960 --> 00:04:04,520 some ways will be the response to the really pro open source of 61 00:04:04,520 --> 00:04:06,280 the first. Of course I I think open source 62 00:04:06,280 --> 00:04:09,120 is very popular in the not open AI world. 63 00:04:09,120 --> 00:04:12,640 So I wouldn't be surprised still if probably the majority of our 64 00:04:12,640 --> 00:04:15,600 panelists are still pro open source or yeah, what do you guys 65 00:04:15,600 --> 00:04:17,880 think? I guess one thing I wanted to 66 00:04:18,120 --> 00:04:23,280 ask you guys on this topic is. Last time at the conference, a 67 00:04:23,280 --> 00:04:26,600 lot of discussion was about the letter that just came out, with 68 00:04:26,600 --> 00:04:31,200 a lot of signatories around taking a pause on AI 69 00:04:31,200 --> 00:04:33,680 development. A modest facility had signed it. 70 00:04:34,000 --> 00:04:38,080 Totally perplexingly, yeah. Yeah, which, which was? 71 00:04:38,360 --> 00:04:43,000 Confusing to a lot of people. And I guess my question is, at 72 00:04:43,000 --> 00:04:46,800 the time, it felt like there was a lot of heat on companies even 73 00:04:46,800 --> 00:04:52,040 like Open AI and Sam Altman. For being cavalier with their 74 00:04:52,040 --> 00:04:55,360 development of AI and causing, you know, an AI race and putting 75 00:04:55,360 --> 00:04:58,920 out ChatGPT. And you know, before the bigger, 76 00:04:59,240 --> 00:05:01,200 you know, more cautious companies like Google had 77 00:05:01,200 --> 00:05:05,120 released their AIS, right? And I really think in like 6 78 00:05:05,120 --> 00:05:08,480 months that has just totally changed where you know now that 79 00:05:08,480 --> 00:05:11,320 you have this open source community putting out models. 80 00:05:11,600 --> 00:05:14,760 Like Open AI is able to like look like the responsible 81 00:05:14,960 --> 00:05:18,240 stewards of AI going to the White House and you know, 82 00:05:18,240 --> 00:05:22,760 getting regulations in place. Just curious if you guys agree 83 00:05:22,760 --> 00:05:26,920 with that assessment? I do think a lot of the safety 84 00:05:26,920 --> 00:05:30,840 talk has died down. I guess the executive order is 85 00:05:30,840 --> 00:05:33,080 maybe a contrary example of that. 86 00:05:33,080 --> 00:05:35,640 Although I you know, maybe the White House timeline is like six 87 00:05:35,640 --> 00:05:37,720 months behind everyone else's timeline, but. 88 00:05:39,560 --> 00:05:43,920 It does seem like people are a lot less hung up on like is the 89 00:05:44,080 --> 00:05:47,680 AI going to kill everyone and sort of the tech industry and 90 00:05:47,680 --> 00:05:50,760 then I guess simultaneously we got this massive executive order 91 00:05:50,760 --> 00:05:53,760 banning or not banning but regulating models over a certain 92 00:05:53,760 --> 00:05:58,720 size etcetera, etcetera. So it's it is interesting to 93 00:05:58,720 --> 00:06:02,520 think about if there's going to be this sort of waves and and 94 00:06:02,520 --> 00:06:06,120 troughs or peaks and troughs of the the hype cycle around safety 95 00:06:06,120 --> 00:06:08,240 and I feel like. Burn a little bit of a through 96 00:06:08,240 --> 00:06:11,840 inside the industry right now, but six months from now that 97 00:06:11,840 --> 00:06:13,080 could change again. Just got to keep having 98 00:06:13,080 --> 00:06:16,720 conferences in March. GPD 4 was like, you know, a 99 00:06:16,720 --> 00:06:21,320 shockwave and it had come over 3.5 and I think it felt like, 100 00:06:21,320 --> 00:06:23,960 man, if this keeps going like we're going to have generalized 101 00:06:24,040 --> 00:06:27,880 AI tomorrow and then I nobody is really caught up with them, 102 00:06:27,880 --> 00:06:30,560 right? There are open source models and 103 00:06:30,760 --> 00:06:34,760 people can say, oh, this is better than open AI on cost. 104 00:06:35,160 --> 00:06:38,760 But I don't think anyone even in specialized use cases has really 105 00:06:39,160 --> 00:06:41,960 had anything as smart. So then there's just a little 106 00:06:41,960 --> 00:06:44,560 less fear of like, Oh my God, it's becoming like. 107 00:06:44,560 --> 00:06:49,160 So you know, that is just not sort of the pace hasn't been so 108 00:06:49,160 --> 00:06:51,720 out of control. And I think that's part of what 109 00:06:51,720 --> 00:06:54,200 has cooled things down. But I sort of, I mean I think 110 00:06:54,200 --> 00:06:57,520 the binding executive order is like a huge deal and shows that 111 00:06:57,520 --> 00:07:00,960 there's very active sort of government interference and like 112 00:07:00,960 --> 00:07:02,880 Europe has been talking about things. 113 00:07:03,200 --> 00:07:07,320 So I don't know, I think this question of cracking down on 114 00:07:07,320 --> 00:07:10,880 open source is very real. But yeah, but the open source 115 00:07:10,880 --> 00:07:14,360 community has certainly been vibrant and and you know, and 116 00:07:14,360 --> 00:07:17,520 the other piece of that is that firms like Andreessen and others 117 00:07:17,520 --> 00:07:20,760 have been very loud about sort of positioning themselves as 118 00:07:20,760 --> 00:07:23,360 super pro open source. Yeah, interesting. 119 00:07:23,520 --> 00:07:25,320 Next theme. All right, I have a theme for 120 00:07:25,320 --> 00:07:30,120 you guys to think about here. I think one of. 121 00:07:30,480 --> 00:07:33,800 The all time maybe the number one all time Silicon Valley 122 00:07:33,800 --> 00:07:38,600 investing philosophy for all categories of software is you 123 00:07:38,600 --> 00:07:41,240 know we like to invest in companies that sell picks and 124 00:07:41,240 --> 00:07:43,440 shovels, not the companies that dig for gold, right. 125 00:07:43,440 --> 00:07:47,560 We like to, we like to fund the tool makers, not the tool users 126 00:07:47,560 --> 00:07:51,520 in many cases. And I think what's interesting 127 00:07:51,640 --> 00:07:55,040 just in the last six months between the previous conference 128 00:07:55,360 --> 00:08:00,200 and the current conference is. The the sort of what tools were 129 00:08:00,200 --> 00:08:04,080 in vogue 6-7 months ago and then what tools are invoked today. 130 00:08:04,640 --> 00:08:08,680 And then the sort of cherry on top of that is Open AI itself, 131 00:08:09,000 --> 00:08:13,320 you know, very, very recently announced this kind of huge wave 132 00:08:13,320 --> 00:08:17,280 of first party tools improvements whether it's you 133 00:08:17,280 --> 00:08:22,440 know, retrieval improvements, context increases, speed, cost. 134 00:08:22,800 --> 00:08:26,440 Really seeing anyone able to make their own GPT a character 135 00:08:26,440 --> 00:08:28,400 to kind of go after the character AIS of the world. 136 00:08:28,400 --> 00:08:32,000 Like they really came hard on like a bunch of tools categories 137 00:08:32,000 --> 00:08:34,880 that I would say 6-7 months ago were being funded as like pretty 138 00:08:34,880 --> 00:08:36,600 compelling stand alone companies. 139 00:08:36,960 --> 00:08:41,200 And I think it's sort of a A2 way question which is 1. 140 00:08:42,080 --> 00:08:46,440 Is tools investment still good idea in an AI world like or is 141 00:08:46,440 --> 00:08:50,520 open AI going to eat this all alive and then I guess 2 like? 142 00:08:50,920 --> 00:08:54,000 You know who who is, who is still exciting in the tool space 143 00:08:54,000 --> 00:08:56,280 and and who do we think has like a strong foundation for the 144 00:08:56,280 --> 00:08:59,560 future. Yeah, I mean I think that 145 00:08:59,560 --> 00:09:04,640 question of will Open AI just eat everything applies in every 146 00:09:04,640 --> 00:09:07,680 part of the. Yeah, I mean that was a theme of 147 00:09:07,720 --> 00:09:11,440 the first conference definitely, just like how how relevant is 148 00:09:11,440 --> 00:09:16,240 everybody except Open AI and and sort of an ongoing question. 149 00:09:16,240 --> 00:09:19,440 So certainly, yeah, I think now with the developer conference, 150 00:09:20,120 --> 00:09:26,040 this thread on tools and then I think you know, given ChatGPT is 151 00:09:26,040 --> 00:09:28,600 clearly the most popular consumer AI app. 152 00:09:28,880 --> 00:09:30,720 Well, I'd be interested. I mean, you have authentic 153 00:09:30,720 --> 00:09:33,440 conversations with venture capitalists regularly, Eric, 154 00:09:33,440 --> 00:09:35,520 right behind the scenes that don't all get published. 155 00:09:35,520 --> 00:09:41,160 Like, do you ever get some authentic expression of concern 156 00:09:41,160 --> 00:09:43,600 about tools companies that they've backed in the space that 157 00:09:43,600 --> 00:09:47,280 were like, well, maybe we. Jumped on this tools idea too 158 00:09:47,280 --> 00:09:50,680 early and it actually isn't a standalone product or you know 159 00:09:51,120 --> 00:09:54,040 behind the scenes there are definitely a lot of firms that 160 00:09:54,040 --> 00:09:56,440 are shitting on foundation model bets. 161 00:09:56,560 --> 00:10:00,200 Oh, interesting. Oh yeah, that's that's like, I 162 00:10:00,200 --> 00:10:03,680 mean they're expensive you know they're very speculative. 163 00:10:03,720 --> 00:10:07,600 Open AI is in the lead. They feel like even with open AI 164 00:10:07,600 --> 00:10:09,720 open source comes to parity very quickly. 165 00:10:09,720 --> 00:10:13,120 So the, the category I hear the most dunking on is just like 166 00:10:13,120 --> 00:10:16,640 foundation models and you'll you'll see firms that haven't 167 00:10:16,640 --> 00:10:20,600 made a lot of bets there. You know, yeah, the Valley loves 168 00:10:20,600 --> 00:10:25,160 like tools, picks and shovels. So that's not something where I 169 00:10:25,160 --> 00:10:26,560 feel like I've heard a lot of regret. 170 00:10:26,560 --> 00:10:29,080 And if anything I feel like I've been, I mean I've been talking 171 00:10:29,080 --> 00:10:31,600 like vector databases all week and stuff like that. 172 00:10:32,000 --> 00:10:35,640 And literally you're getting VCs talk about chips or getting 173 00:10:35,640 --> 00:10:38,960 companies around chips, which traditionally they've been 174 00:10:38,960 --> 00:10:41,400 scared about. But you have sort of the fact 175 00:10:41,400 --> 00:10:44,880 that NVIDIA is doing so well and then you also have the fact that 176 00:10:44,880 --> 00:10:47,600 it's sort of cool to be like an emerging tech investor at the 177 00:10:47,600 --> 00:10:48,840 moment. So I think those things 178 00:10:48,840 --> 00:10:54,200 combined, you're seeing that. So no, I I think V CS dunking on 179 00:10:54,200 --> 00:10:57,120 foundation models, but bullish about everything else supporting 180 00:10:57,360 --> 00:10:59,760 AI. Yeah. 181 00:11:00,000 --> 00:11:02,360 Are there particular company types you have in mind like do 182 00:11:02,400 --> 00:11:04,120 you think? Well, I mean, I think there's 183 00:11:04,120 --> 00:11:06,000 like obvious companies that are like. 184 00:11:06,640 --> 00:11:08,880 May be less hot than they were in the last conference, right? 185 00:11:08,880 --> 00:11:11,200 I mean like Jasper. Jasper would be an example, 186 00:11:11,200 --> 00:11:13,080 right? Stability would be a huge 187 00:11:13,080 --> 00:11:14,840 example, right? But I think there's company 188 00:11:14,840 --> 00:11:17,080 stability that's just like the unique failings of. 189 00:11:17,400 --> 00:11:19,400 That sure, sure. But there are companies that we 190 00:11:19,400 --> 00:11:21,440 talked to last time, right? You know, Lang Chain, for 191 00:11:21,440 --> 00:11:24,440 example, right. That like it's a big question as 192 00:11:24,440 --> 00:11:27,720 to where Lang Chain fits in the universe, where Open AI is just 193 00:11:27,720 --> 00:11:29,160 going to take care of all this crap, right? 194 00:11:29,200 --> 00:11:31,040 And then? You know, I think we're talking 195 00:11:31,040 --> 00:11:33,560 to some companies this week that, you know, look like they 196 00:11:33,560 --> 00:11:35,560 they could totally be awesome companies. 197 00:11:35,560 --> 00:11:38,200 And I have no earthly clue, but they're providing very 198 00:11:38,200 --> 00:11:41,600 specialized tools like you know, data cleaning or something like 199 00:11:41,600 --> 00:11:43,640 that, right. Like which again I'm excited to 200 00:11:43,640 --> 00:11:45,960 hear the pitch like of of why that's a a great idea for the 201 00:11:45,960 --> 00:11:49,320 company and then probably is. But like, you know, I would be a 202 00:11:49,320 --> 00:11:51,680 little scared that opening. I was just going to like kick 203 00:11:51,680 --> 00:11:54,200 ass in my category if I were a tools company. 204 00:11:55,240 --> 00:11:57,800 And so I just think, and I mean, I think you have to agree that 205 00:11:57,800 --> 00:12:01,440 character has to be pretty upset that Open AI basically announced 206 00:12:01,440 --> 00:12:03,680 their version of character AI tools, right? 207 00:12:03,680 --> 00:12:08,320 Like, so I'm just interested if there's any real feedback of the 208 00:12:08,320 --> 00:12:10,600 fear factor on that front. But it sounds like maybe not. 209 00:12:10,600 --> 00:12:15,600 So yeah. Do you think just jumping diving 210 00:12:15,600 --> 00:12:18,680 in on the character question like there is a theme I've 211 00:12:18,680 --> 00:12:22,120 noticed of people kind of questioning whether? 212 00:12:22,680 --> 00:12:27,720 Open AI for consumers as an end point is is really sufficient. 213 00:12:27,720 --> 00:12:30,000 So there's kind of like the opposite side of that question 214 00:12:30,000 --> 00:12:32,520 from the tooling side, right? Like will they eat all tools? 215 00:12:32,800 --> 00:12:35,720 And then there's from the consumer assistant side, is 216 00:12:35,720 --> 00:12:38,800 there a super assistant that's just ChatGPT or do you end up 217 00:12:38,800 --> 00:12:43,480 using all these characters or mini GBTS that that they are 218 00:12:43,480 --> 00:12:45,720 working to enable developers to build? 219 00:12:46,040 --> 00:12:47,880 How do you guys feel about that debate? 220 00:12:48,960 --> 00:12:51,040 Yeah, I mean, I think that's, that's the open question. 221 00:12:51,040 --> 00:12:54,600 I think it depends. It depends on the quality of 222 00:12:54,680 --> 00:12:57,120 the, you know, the foundation. Models. 223 00:12:57,120 --> 00:12:58,400 I don't know. I mean, I guess like you 224 00:12:58,400 --> 00:13:01,640 selfishly we built games, James, that are built on AI tools, 225 00:13:01,840 --> 00:13:04,400 including in some cases large language models. 226 00:13:04,760 --> 00:13:09,160 So my selfish belief is, is possible to build games on LLMS 227 00:13:09,160 --> 00:13:11,680 without open AI building the exact same thing. 228 00:13:12,240 --> 00:13:16,280 It may depend on the complexity level of the consumer app built 229 00:13:16,280 --> 00:13:18,440 on top of the tool, whereas character is sort of. 230 00:13:19,160 --> 00:13:23,600 Pretty vanilla RE, you know, re prompting or you know tweaked 231 00:13:23,600 --> 00:13:26,040 version of the core, the GPT experience. 232 00:13:26,040 --> 00:13:29,120 But you know, I think it depends on the complexity level and sort 233 00:13:29,120 --> 00:13:32,200 of how much of the stack you control and and how easy it is 234 00:13:32,200 --> 00:13:35,680 to get to your customers. I mean a phrase I've been 235 00:13:35,680 --> 00:13:38,080 hearing over and over again this week, you know, just talking to 236 00:13:38,080 --> 00:13:42,200 VCs has been retrieval augmented generation. 237 00:13:42,200 --> 00:13:45,600 Rags, Rags and I think there are a lot of companies right now 238 00:13:45,600 --> 00:13:51,440 that think we can be really good at giving GPT the sort of 239 00:13:51,440 --> 00:13:57,360 information it needs and then you know Open AI or whoever can 240 00:13:57,360 --> 00:13:59,160 sort of do the thinking based on that. 241 00:13:59,160 --> 00:14:01,880 But we're not, we're not in the business of making the best 242 00:14:01,880 --> 00:14:04,720 thinking company, but we're we're really good at figuring 243 00:14:04,720 --> 00:14:08,800 out like what it needs to consider and and it's possible 244 00:14:08,800 --> 00:14:11,200 that there are a lot of what. Kind of companies do you think 245 00:14:11,200 --> 00:14:13,560 are saying that around like I guess this is kind of. 246 00:14:15,080 --> 00:14:19,160 Pointing to data and unique data sets as kind of a mode of some 247 00:14:19,160 --> 00:14:22,400 kind, right? Like, what kinds of companies do 248 00:14:22,400 --> 00:14:26,320 you think fit that category of potentially having a true data 249 00:14:26,320 --> 00:14:31,200 mode that could be valuable, having a data mode? 250 00:14:31,600 --> 00:14:38,600 Yeah, Well, well, this isn't a data mode, but I mean, I'm one 251 00:14:38,600 --> 00:14:40,880 company as you'll hear in the second. 252 00:14:40,880 --> 00:14:43,120 Isn't it a data? Isn't it a data mode that you 253 00:14:43,120 --> 00:14:45,800 are saying that? These companies are talking to 254 00:14:45,800 --> 00:14:49,560 you saying we have well, like OK, there's this company clean 255 00:14:49,560 --> 00:14:52,160 right, that lets companies run searches, right? 256 00:14:52,200 --> 00:14:54,240 Is it a data mode? They're really like good at 257 00:14:54,240 --> 00:14:58,920 hooking companies up, companies data up to data sources that 258 00:14:58,920 --> 00:15:03,240 then connect, that use rag and then plug into something like 259 00:15:03,560 --> 00:15:06,160 Open AI. So it's not necessarily like a 260 00:15:06,160 --> 00:15:08,560 Bloomberg where it's like we have all the financial data in 261 00:15:08,560 --> 00:15:10,480 the world. It's we're really good at like 262 00:15:10,480 --> 00:15:14,440 getting the relevant data four companies into into this system 263 00:15:14,440 --> 00:15:18,760 is is 1 case. But at least at Open AI Dev Day 264 00:15:18,760 --> 00:15:23,280 this week, they seem to be going after some similar use cases of 265 00:15:23,280 --> 00:15:26,560 allowing companies and enterprise to build their own 266 00:15:26,560 --> 00:15:31,240 internal GPTS or chat GPTS based on internal company data. 267 00:15:31,240 --> 00:15:35,200 So I think that that is still some area that they seem to be 268 00:15:35,200 --> 00:15:39,200 competing in as well. I mean, you know, eventually 269 00:15:39,200 --> 00:15:43,520 there's only so much one company can do unless, unless it's like 270 00:15:43,760 --> 00:15:46,560 the, I feel like James, this is your style of suggestion 271 00:15:46,560 --> 00:15:49,640 throughout this series. Unless there's AGI and like 272 00:15:49,640 --> 00:15:51,160 every way we talk about this is wrong. 273 00:15:51,160 --> 00:15:55,280 But in any sort of traditional business sense where like you 274 00:15:55,280 --> 00:15:58,320 know, Sam Ullman, you know, has joked in the past that like, oh, 275 00:15:58,320 --> 00:16:02,240 we'll just ask open it, we'll ask Chad GBT at some point, you 276 00:16:02,240 --> 00:16:04,040 know, like what our business model should be and we'll do 277 00:16:04,080 --> 00:16:06,240 that. Clearly they didn't need to wait 278 00:16:06,240 --> 00:16:08,080 for that. They figured out some business 279 00:16:08,080 --> 00:16:11,280 models. But you know, I anyway, my my 280 00:16:11,280 --> 00:16:14,040 point is just like in a traditional, normal world that I 281 00:16:14,040 --> 00:16:17,520 expect to continue, You can't do everything And so whatever, 282 00:16:17,520 --> 00:16:19,960 they're going to have to pick their strengths and then 283 00:16:20,120 --> 00:16:22,320 everybody else can do other stuff. 284 00:16:24,480 --> 00:16:28,360 I'm going to, I'm going to give a last, a last theme that's sort 285 00:16:28,360 --> 00:16:31,920 of more in the fun sort of philosophy sort of world. 286 00:16:32,120 --> 00:16:36,160 You know, fitting into that AGI question which is just how 287 00:16:36,160 --> 00:16:41,920 seriously do people take existential risk in a, you know, 288 00:16:42,160 --> 00:16:47,400 is existential risk now is seen as like this is propaganda for 289 00:16:47,680 --> 00:16:51,440 people who want to make it 1 seem really awesome and sexy and 290 00:16:51,440 --> 00:16:54,240 amazing and can do everything. And then two people who want to 291 00:16:54,240 --> 00:16:57,200 be in this world of like shutting it down, Like how much 292 00:16:57,200 --> 00:17:01,040 is existential risk really on people's minds anymore? 293 00:17:01,240 --> 00:17:05,440 Or is it much important that we just make sure it's not racist 294 00:17:05,440 --> 00:17:07,880 and make sure, you know, deal with these like, very practical 295 00:17:08,280 --> 00:17:12,359 concerns that are obviously true now versus imagining this world 296 00:17:12,359 --> 00:17:15,079 where it's. You know, you know, like our AI 297 00:17:15,079 --> 00:17:17,079 Doom episode. You know, figuring out how to 298 00:17:17,599 --> 00:17:19,400 optimize paper clips into blowing up the. 299 00:17:19,480 --> 00:17:20,640 Earth. That's kind of alluding to my 300 00:17:20,640 --> 00:17:23,000 earlier point about I feel like Silicon Valley has stopped 301 00:17:23,000 --> 00:17:24,920 taking safety as seriously, right. 302 00:17:24,920 --> 00:17:27,880 I mean, like, I think that like, yeah, sure, the executive order 303 00:17:27,880 --> 00:17:29,200 is not an example of that, right? 304 00:17:29,200 --> 00:17:33,560 But in Silicon Valley, I think the discussion of, to your 305 00:17:33,560 --> 00:17:37,200 point, the the implied seriousness with which people 306 00:17:37,200 --> 00:17:39,320 take. Existential risk seems pretty 307 00:17:39,320 --> 00:17:44,840 low to me because otherwise people would be like blowing up 308 00:17:44,840 --> 00:17:47,320 GPU factories or whatever. I don't know. 309 00:17:47,320 --> 00:17:50,360 It's like, if you actually believe this, had a 20% chance 310 00:17:50,360 --> 00:17:54,320 of killing all of society, which a lot of people say, like how 311 00:17:54,320 --> 00:17:57,360 could you morally work in this space? 312 00:17:57,360 --> 00:18:01,520 Or at least not like, you know, be actively opposed to any of 313 00:18:01,520 --> 00:18:04,040 the development here, right? And and the only people who seem 314 00:18:04,040 --> 00:18:06,680 to really act as if they truly believe existential risk. 315 00:18:07,320 --> 00:18:10,080 Is real is basically you know, Elisa Yodkowski and the sort of 316 00:18:10,080 --> 00:18:12,680 the AI doomers on Twitter, right, who are really out there 317 00:18:12,680 --> 00:18:15,080 every day saying there was, there was an interview. 318 00:18:15,440 --> 00:18:18,760 Shout out to Logan Bartlett with his podcast. 319 00:18:19,000 --> 00:18:22,520 He interviewed Dario, the Co founder. 320 00:18:22,520 --> 00:18:25,000 Of anthropic, right? And Dario gave like it was like 321 00:18:25,000 --> 00:18:28,240 20% chance, like everything. Goes there was like 10 to 20% 322 00:18:28,240 --> 00:18:30,080 chance that was like really high I think. 323 00:18:31,400 --> 00:18:32,840 Is that like in the next 100 years? 324 00:18:32,840 --> 00:18:35,000 So And again the time frames matter right? 325 00:18:35,000 --> 00:18:35,840 Like. I think it was. 326 00:18:36,040 --> 00:18:39,400 It was reasonably soon. But yeah, I I don't know. 327 00:18:39,400 --> 00:18:43,840 I sort of think you'd have to be insane to be working in an 328 00:18:43,840 --> 00:18:47,320 industry where you truly believe that progress there had a 20% 329 00:18:47,320 --> 00:18:49,360 chance of ending all humanity. I I don't know. 330 00:18:49,360 --> 00:18:51,720 Like, unless you believe in some complicated. 331 00:18:52,120 --> 00:18:53,960 Game theory about deterrence or something. 332 00:18:53,960 --> 00:18:56,760 And back to our earlier episode about, you know, the only thing 333 00:18:56,760 --> 00:18:59,640 that can stop a good guy with an AI, or sorry, a bad guy with an 334 00:18:59,760 --> 00:19:01,720 AI is a good guy with an AII. Said it wrong. 335 00:19:01,760 --> 00:19:03,600 Who coined that line? Did I get that line? 336 00:19:03,760 --> 00:19:05,040 I don't. I can't remember but. 337 00:19:05,480 --> 00:19:08,200 But the only thing that can stop a bad guy with an AI is a good 338 00:19:08,200 --> 00:19:11,680 guy with a better AI, right? Yeah, unless you can explain 339 00:19:11,680 --> 00:19:16,560 some pretty elaborate game theory on how progress in AI by 340 00:19:16,560 --> 00:19:18,760 the good guys is necessary to stop extinction. 341 00:19:19,520 --> 00:19:21,720 It just, you know, it doesn't compute that you believe this 342 00:19:21,720 --> 00:19:26,200 stuff has 20% chance of killing us like you know, and be working 343 00:19:26,200 --> 00:19:29,360 in the industry at the level of Dario, who literally is like the 344 00:19:29,360 --> 00:19:33,360 number two most powerful person in foundation models, right? 345 00:19:33,360 --> 00:19:36,760 I mean it's just, it's nonsense. So I don't know. 346 00:19:36,760 --> 00:19:39,360 I agree it it doesn't seem like people take it that seriously 347 00:19:39,360 --> 00:19:42,600 for real. As we discussed in that episode, 348 00:19:42,600 --> 00:19:45,360 though, there's a lot of, like, very bad things that could 349 00:19:45,360 --> 00:19:48,400 happen that, you know, AI kills some of us, right? 350 00:19:48,400 --> 00:19:49,360 Not. All of us, Yeah. 351 00:19:49,360 --> 00:19:52,040 That was the that was. The that was the take away all. 352 00:19:52,080 --> 00:19:53,520 You're right, right. Maybe it doesn't. 353 00:19:53,600 --> 00:19:58,720 Like, I guess like maybe maybe people have come around to sort 354 00:19:58,720 --> 00:20:02,520 of a middle ground here where it's not existential risk, like 355 00:20:02,560 --> 00:20:05,520 extinction for humanity and more, that there's a lot of bad 356 00:20:05,520 --> 00:20:08,680 things that could happen over the next 5 to 10 years with AI 357 00:20:08,680 --> 00:20:12,080 and I think all of us. Would agree with that, right? 358 00:20:12,320 --> 00:20:16,320 You know there are bad things that could occur if the, you 359 00:20:16,360 --> 00:20:20,080 know, some bad actors got access to some really powerful sure 360 00:20:20,080 --> 00:20:21,560 models or something, right? I mean. 361 00:20:22,840 --> 00:20:27,240 I think you the fact that Cruise is not active in San Francisco, 362 00:20:27,240 --> 00:20:30,680 like even if they sort of bungled their disclosures to the 363 00:20:30,680 --> 00:20:35,080 government, I mean Cruise is a better driver than humans in San 364 00:20:35,080 --> 00:20:37,560 Francisco, right Like. They really they they. 365 00:20:37,880 --> 00:20:41,320 They broke the one rule, which is don't lie to the government, 366 00:20:41,320 --> 00:20:42,840 right? I mean, like, it's just like, 367 00:20:43,280 --> 00:20:47,920 you can be, you can be bad, you can screw up, you know? 368 00:20:47,920 --> 00:20:51,720 But don't lie to the government is like freaking like #1 on the 369 00:20:51,760 --> 00:20:54,040 10 commandments of running a company, right? 370 00:20:54,040 --> 00:20:57,640 I mean, like, you know, that's why SPF travels or whatever, You 371 00:20:57,680 --> 00:21:00,040 know, like, I mean, like like just don't lie to the 372 00:21:00,040 --> 00:21:03,560 government, like, you know? Fake it so you make it can apply 373 00:21:03,560 --> 00:21:05,840 in many scenarios, but not with the government. 374 00:21:05,840 --> 00:21:07,400 There, you can't do that with them. 375 00:21:08,640 --> 00:21:11,280 So anyway can we just go back to this for a second? 376 00:21:11,280 --> 00:21:17,320 So you're you're what are you saying Eric about existential 377 00:21:17,320 --> 00:21:20,960 risk or Max like you you guys are saying companies are not 378 00:21:20,960 --> 00:21:24,040 acting like they believe their own hype or do you agree with 379 00:21:24,040 --> 00:21:26,720 that or? I think, I think people are now 380 00:21:26,720 --> 00:21:29,880 saying calling bullshit in sort of private. 381 00:21:30,280 --> 00:21:32,400 Existential risk. And some of them don't want to 382 00:21:32,400 --> 00:21:35,480 say it publicly because they don't want to be seen as sort of 383 00:21:35,480 --> 00:21:39,960 undermining legitimate concerns. But I think a lot of them are 384 00:21:39,960 --> 00:21:44,520 just like doesn't feel close and then how do you feel that ties 385 00:21:44,520 --> 00:21:50,760 into the executive order? Does that feel to you like over 386 00:21:50,760 --> 00:21:53,040 regulation or you know, pre? Premature. 387 00:21:53,040 --> 00:21:56,600 I'll be sort of waiting until after the conference to stake 388 00:21:56,600 --> 00:21:59,720 out my position partially, just like Get. 389 00:22:00,240 --> 00:22:05,760 Get more information And also, you know I sort of shift between 390 00:22:06,200 --> 00:22:09,120 information gatherer and sometimes opinion haver. 391 00:22:09,120 --> 00:22:10,720 And I don't know. I've been sort of waiting to 392 00:22:11,280 --> 00:22:13,160 come out strong, though I do have a sort of. 393 00:22:13,640 --> 00:22:17,080 Side the last theme I'm actually interested in, and this is very 394 00:22:17,080 --> 00:22:19,800 close to home for James and I in the last couple of months, is. 395 00:22:21,400 --> 00:22:24,960 We are increasingly discovering that the cost of running any 396 00:22:24,960 --> 00:22:29,000 large language model products at consumer scale is unbelievably 397 00:22:29,080 --> 00:22:30,280 high. Like, oh, interesting. 398 00:22:31,200 --> 00:22:36,400 Yeah, is so high that it's it's very hard to come up with a 399 00:22:36,400 --> 00:22:40,920 mental model for how you could build anything on top of a GPT 400 00:22:40,920 --> 00:22:43,840 for quality model right now that would be distributed to 401 00:22:43,840 --> 00:22:47,160 consumers. You could maybe do it on 3/5. 402 00:22:47,640 --> 00:22:50,000 Especially with some recent price cuts with, you know, fine 403 00:22:50,000 --> 00:22:54,560 tuning, yadda yadda yadda. But I would say that it is 404 00:22:54,560 --> 00:22:58,880 shockingly high to run a very basic application that's GPT 4 405 00:22:58,880 --> 00:23:02,880 level quality on LLM. And I think it basically makes 406 00:23:02,880 --> 00:23:06,520 it impossible to build consumer apps on these level of models 407 00:23:06,520 --> 00:23:08,760 today without incinerating money. 408 00:23:09,520 --> 00:23:13,040 And so I think it's an interesting question as to 409 00:23:13,040 --> 00:23:15,400 whether or not a lot of these companies that have raised 100, 410 00:23:15,400 --> 00:23:19,320 two $100 million dollars. I have to imagine their unit 411 00:23:19,320 --> 00:23:21,680 economics are pretty much upside down if they're building any 412 00:23:21,680 --> 00:23:25,480 sort of consumer product. As in every time someone uses 413 00:23:25,480 --> 00:23:27,520 their product they lose more money than they make. 414 00:23:27,720 --> 00:23:34,600 So I wonder if the only outcome of this is that B2B companies 415 00:23:34,600 --> 00:23:36,960 can maybe afford this because they can charge such such 416 00:23:36,960 --> 00:23:38,800 staggering prices. And then? 417 00:23:39,200 --> 00:23:41,480 Open AI is like the only one that can do consumer stuff 418 00:23:41,480 --> 00:23:44,600 because you pretty much have to have no middleman to afford 419 00:23:44,600 --> 00:23:46,000 running these applications. OK. 420 00:23:46,000 --> 00:23:49,440 So that's a great one and I'll broaden it and just say I think 421 00:23:49,480 --> 00:23:53,480 there's an the overall meta theme of that is the rubber 422 00:23:53,480 --> 00:23:55,320 meets the road, right? There's this. 423 00:23:55,560 --> 00:23:57,760 Do people really care about existential risk anymore? 424 00:23:57,760 --> 00:24:00,280 Because it's like, oh, we're just in the practical problems. 425 00:24:00,280 --> 00:24:04,200 Like it's really expensive to run consumer apps and I've heard 426 00:24:04,200 --> 00:24:06,720 people in enterprise ask like are. 427 00:24:07,080 --> 00:24:11,160 Are big companies, customers sticking around or do they try 428 00:24:11,160 --> 00:24:14,040 the demo and say we didn't like it anymore? 429 00:24:14,040 --> 00:24:17,240 And I mean, you know, Chachi, BTI think we're going to see in 430 00:24:17,240 --> 00:24:20,920 a slide during the presentation, you know, activity syncs and 431 00:24:20,920 --> 00:24:23,160 then they have new features and goes up like what's the 432 00:24:23,160 --> 00:24:25,720 resiliency? So just this, yeah, the rubber 433 00:24:25,720 --> 00:24:29,440 mean the road, all right, We've gotten all the hype like how are 434 00:24:29,440 --> 00:24:31,800 people delivering? And now it's sort of the moment 435 00:24:31,800 --> 00:24:35,160 where it's turning into sort of more regular sort of business 436 00:24:35,160 --> 00:24:39,480 style questions. I mean, this stuff is so 437 00:24:39,480 --> 00:24:42,320 expensive to run that it's hard to imagine you would not have to 438 00:24:42,320 --> 00:24:44,720 run into regular business style questions pretty quickly. 439 00:24:45,120 --> 00:24:48,960 Now I was not in the Internet industry in 1993 when you had to 440 00:24:48,960 --> 00:24:54,040 go buy your own servers or on a website, you know, so that this 441 00:24:54,040 --> 00:24:57,040 may be analogous to that time, you know, where you had to be a 442 00:24:57,040 --> 00:25:00,760 garage band server farm to afford to run these kind of 443 00:25:00,760 --> 00:25:03,720 companies. But even that would have been 444 00:25:03,720 --> 00:25:06,160 that would have been fixed cost, right, I mean now we're talking 445 00:25:06,160 --> 00:25:08,160 about. Like we're talking erasing cost. 446 00:25:08,160 --> 00:25:09,840 More customers you get, more expensive. 447 00:25:09,840 --> 00:25:12,880 Yeah, Erasing software margins like. 448 00:25:13,080 --> 00:25:17,080 So like, I'm pretty interested to see if anyone can run a 449 00:25:17,080 --> 00:25:22,680 consumer product without just incinerating money besides Open 450 00:25:22,720 --> 00:25:23,440 AI. So. 451 00:25:24,040 --> 00:25:26,040 Stay tuned or or or use I would say. 452 00:25:26,120 --> 00:25:28,960 I would say the last caveat would just be using smaller 453 00:25:28,960 --> 00:25:31,160 models or. Yeah, using smaller smallers or 454 00:25:31,160 --> 00:25:35,840 fine tuning or something. Or GBT 4 gets 20 times cheaper 455 00:25:35,840 --> 00:25:38,960 in the next year and a half. And then this is a dumb, 456 00:25:39,000 --> 00:25:41,600 irrelevant, you know, retrospective discussion, but it 457 00:25:41,600 --> 00:25:44,360 really has to go down by like a 10X to be interesting. 458 00:25:44,560 --> 00:25:49,880 So they did three XA week ago. We need two more of those to 459 00:25:49,880 --> 00:25:51,720 really be in a good category of cost. 460 00:25:53,840 --> 00:25:55,800 Yeah, we, we should. I mean, that is a good reminder 461 00:25:55,800 --> 00:25:59,600 that we should ask money. Like, how much does this cost? 462 00:25:59,600 --> 00:26:01,160 Like, how? Yeah. 463 00:26:01,160 --> 00:26:04,280 How Where do you make money if like, you know, prices change 464 00:26:04,280 --> 00:26:05,360 one way or the other. Yeah. 465 00:26:05,600 --> 00:26:06,800 Yeah. All right. 466 00:26:07,240 --> 00:26:11,360 Next up, I really think you're going to want to listen to this, 467 00:26:11,520 --> 00:26:13,800 the draft pick. Please give us your feedback. 468 00:26:14,760 --> 00:26:19,680 I want insane like memos about like what the order strategy 469 00:26:19,680 --> 00:26:23,040 should have been, and just feel free to weigh in. 470 00:26:23,600 --> 00:26:26,760 We did, you know, I I think we explained all the particulars in 471 00:26:26,760 --> 00:26:28,760 in James, right. You talked through Yes, yeah, 472 00:26:28,760 --> 00:26:32,600 we're about to jump into the rules and there were, there was 473 00:26:32,600 --> 00:26:35,520 probably more time spent behind the scenes debating the rules 474 00:26:35,520 --> 00:26:38,880 than I preparing for the draft. I was insistent about a 475 00:26:38,880 --> 00:26:42,360 particular rule change that I mean you'll, you'll hear. 476 00:26:42,440 --> 00:26:47,760 So we had a lot of fun. Sent you Eric at newcomer.co. 477 00:26:48,120 --> 00:26:51,800 You can always send me your opinions about who's an idiot I 478 00:26:52,000 --> 00:26:54,400 hopefully. I think I'll try and post the 479 00:26:54,400 --> 00:26:57,880 list. So please forgive us that it was 480 00:26:57,880 --> 00:26:59,680 a set list. There are companies. 481 00:26:59,920 --> 00:27:02,760 I asked if I could pick companies off list as sort of a 482 00:27:03,160 --> 00:27:05,680 insider with Intel. But I was I was not allowed to 483 00:27:05,680 --> 00:27:08,160 pick off list. So I you're restricted to a list 484 00:27:08,160 --> 00:27:12,680 of companies and we didn't pick you, please hire a new marketing 485 00:27:12,680 --> 00:27:13,400 officer. I don't know. 486 00:27:13,400 --> 00:27:17,080 Sorry. Yeah, find find out why your 487 00:27:17,080 --> 00:27:21,400 your valuation was also not, you know, listed in Pitchbug, right 488 00:27:21,400 --> 00:27:26,880 'cause the list was selected from companies who are or who 489 00:27:26,880 --> 00:27:31,680 have have a recent over $100 million rounds in Pitchbug. 490 00:27:32,640 --> 00:27:33,880 There's go. Listen to that. 491 00:27:37,800 --> 00:27:38,280 Welcome. Welcome. 492 00:27:38,280 --> 00:27:41,520 All right. We are gonna move into the draft 493 00:27:41,720 --> 00:27:45,720 we this is probably the thing we're most excited about at 494 00:27:45,720 --> 00:27:49,560 least informed to do taking a page out of. 495 00:27:50,320 --> 00:27:54,040 Sports Podcast, Political Podcast Playbook and we are 496 00:27:54,040 --> 00:27:59,720 going to do a draft pick of the top AI companies. 497 00:28:00,440 --> 00:28:05,800 James, do you want to explain the metrics for who wins this 498 00:28:05,800 --> 00:28:08,680 and how we know and all of that? Here's what we're doing. 499 00:28:08,960 --> 00:28:13,520 Max, Eric and I are drafting teams of AI startups. 500 00:28:13,800 --> 00:28:18,120 The startups must have risen over $100 million. 501 00:28:18,240 --> 00:28:20,720 I wish that wasn't the case, partially because I have Intel 502 00:28:20,720 --> 00:28:24,240 on like companies that are promising, but James isn't 503 00:28:24,240 --> 00:28:27,880 letting me include random. I feel like there's, you know, 504 00:28:27,880 --> 00:28:30,080 it's a risk. They are not even valued at that 505 00:28:30,080 --> 00:28:35,400 yet, but I I, yeah, I think we have, we have 43 companies in 506 00:28:35,400 --> 00:28:39,320 the list and we're only picking five each. 507 00:28:39,320 --> 00:28:42,120 So many of these will not make it on the board. 508 00:28:42,120 --> 00:28:46,920 So I'm excited to see what happens, but this really shows, 509 00:28:46,920 --> 00:28:51,000 I think, the amount of companies who have raised ginormous rounds 510 00:28:51,080 --> 00:28:54,320 in AI. Is this 100 million plus list 511 00:28:54,320 --> 00:28:57,480 you have to have raised raised over 100 million, not valuation 512 00:28:57,480 --> 00:28:59,520 raised. It's probably 100 million 513 00:28:59,520 --> 00:29:01,840 because I I do think also they're probably ones that 514 00:29:01,840 --> 00:29:05,120 aren't. So some other criteria I used to 515 00:29:05,120 --> 00:29:09,320 kind of narrow down this list. You had to be using generative 516 00:29:09,360 --> 00:29:14,280 AI or the company had to be have generated AI or generative AI 517 00:29:14,280 --> 00:29:17,880 infra kind of core to it. I kind of excluded some 518 00:29:17,880 --> 00:29:22,000 industries like bio and healthcare, defense, silicon 519 00:29:22,000 --> 00:29:26,560 chips, robotics, driverless, or a VS and China. 520 00:29:26,560 --> 00:29:30,680 I just wanted to filter down to companies we might have a chance 521 00:29:30,720 --> 00:29:34,000 of knowing something about or being predictive about, rather 522 00:29:34,000 --> 00:29:39,120 than just predicting whether some biotech company is going to 523 00:29:39,120 --> 00:29:41,760 use AI to discover a new drug or something. 524 00:29:42,960 --> 00:29:47,560 And I also wanted companies founded before excluded, 525 00:29:47,560 --> 00:29:51,360 companies founded before 2013. So most of these companies have 526 00:29:51,360 --> 00:29:53,880 been actually founded in the last three years or something, 527 00:29:54,840 --> 00:29:57,480 but there are a few who have been founded maybe earlier. 528 00:29:58,440 --> 00:30:02,320 So that's the game and we are going to determine who goes 529 00:30:02,320 --> 00:30:06,240 first by creating a handicap auction. 530 00:30:06,240 --> 00:30:09,360 So this was, I was adamant, yeah, that we do this. 531 00:30:09,640 --> 00:30:12,480 James and Max were like, oh, you know, that's part of the 532 00:30:12,480 --> 00:30:14,880 randomness. But to me, like a huge part of 533 00:30:14,880 --> 00:30:18,680 this is how how much would you pay to get to pick? 534 00:30:19,160 --> 00:30:22,040 Open AI Because I, I, I honestly think we're all going to pick 535 00:30:22,160 --> 00:30:24,640 Open AI. We should talk about what the, 536 00:30:24,720 --> 00:30:27,360 what the, what the grading criteria is. 537 00:30:27,360 --> 00:30:31,760 Yeah, here. So the way we are drafting these 538 00:30:31,760 --> 00:30:35,800 teams, we are we are trying to create a team with the highest 539 00:30:35,800 --> 00:30:39,800 total valuation on November 1st, 2028. 540 00:30:40,200 --> 00:30:44,720 And final valuations would be determined by pitch book, either 541 00:30:44,720 --> 00:30:48,880 kind of a reported M and a price or a current stock price if if 542 00:30:48,880 --> 00:30:52,920 the company is public or the last evaluation if the company 543 00:30:52,920 --> 00:30:56,640 is still going but is private and if the company is shut down, 544 00:30:56,640 --> 00:30:59,640 I think that would count as a 0 for that team. 545 00:31:00,880 --> 00:31:04,440 If if like 2 year, does it have to have raised at the price 546 00:31:04,440 --> 00:31:09,120 within the last two years or something because? 547 00:31:09,200 --> 00:31:10,680 There's whatever the last fundraise was. 548 00:31:11,000 --> 00:31:12,560 Yeah, just. But like, it might not be worth 549 00:31:12,560 --> 00:31:15,040 that anymore. One question I have is let's 550 00:31:15,040 --> 00:31:19,160 let's say there's a company on this list who has raised, who 551 00:31:19,160 --> 00:31:23,640 has raised a huge round, maybe in the Zerp era but never raises 552 00:31:23,640 --> 00:31:27,000 again, but is a zombie company in five years. 553 00:31:27,000 --> 00:31:31,720 Are we counting that at their? I think the last valuation. 554 00:31:31,960 --> 00:31:35,680 No, it needs to raise another round or do something in the 555 00:31:35,680 --> 00:31:38,040 next five years because otherwise there's a strategy of 556 00:31:38,040 --> 00:31:40,440 just like picking the most highly valued companies. 557 00:31:41,120 --> 00:31:43,240 And just betting that like nothing happens. 558 00:31:43,240 --> 00:31:44,440 That's the efficient market hype. 559 00:31:44,440 --> 00:31:47,400 Out there, I don't think. I think it's theory, current 560 00:31:47,400 --> 00:31:50,120 evaluation, other. Is the best particular of future 561 00:31:50,120 --> 00:31:52,120 value. It needs another either of a, it 562 00:31:52,120 --> 00:31:53,800 needs another fundraiser. Reporting. 563 00:31:53,920 --> 00:31:56,520 Yeah, I think so. Max, are you OK with that or. 564 00:31:57,160 --> 00:31:58,400 I mean, I don't care. I don't care. 565 00:31:58,400 --> 00:31:59,040 Yeah. Yeah, yeah. 566 00:31:59,360 --> 00:32:01,760 So whatever. It needs something, whatever, 567 00:32:01,760 --> 00:32:02,960 right? Are we agreed on that? 568 00:32:03,360 --> 00:32:07,040 Yeah, we're agreed. So if the company does not raise 569 00:32:07,040 --> 00:32:11,720 any round between now and November 1st, 2028, yeah, 570 00:32:12,320 --> 00:32:16,600 according to Pitch Book, a new round, they will count as a zero 571 00:32:16,600 --> 00:32:18,320 as well. Fair. 572 00:32:18,480 --> 00:32:19,680 OK. Yep, brutal. 573 00:32:21,960 --> 00:32:27,440 So how are we doing the bid for 1st and 2nd place? 574 00:32:28,200 --> 00:32:34,120 Well, I think the the idea here is if you you know we're gonna 575 00:32:34,120 --> 00:32:38,880 just go down the line here and name a price that you are 576 00:32:38,880 --> 00:32:42,160 willing to subtract from your total score and then in the end 577 00:32:42,160 --> 00:32:47,640 and then determine you know the order in that in that order of 578 00:32:47,640 --> 00:32:51,600 those 3 handicaps and we will deduct those scores at the. 579 00:32:51,600 --> 00:32:52,640 End. All right, let's just do a live 580 00:32:52,640 --> 00:32:54,200 auction. Like whoever gets the highest 581 00:32:54,200 --> 00:32:57,480 price and then we stop it. OK, so then it's a coin flip. 582 00:32:57,760 --> 00:32:59,360 OK, fine. Coin flip for a second? 583 00:32:59,720 --> 00:33:02,160 Yes. OK, let me pull up my digital 584 00:33:02,160 --> 00:33:05,920 coin flipper. OK, All right. 585 00:33:06,040 --> 00:33:08,200 And then it's a snake draft. Yes. 586 00:33:08,840 --> 00:33:10,280 All right. Let's let's start. 587 00:33:10,400 --> 00:33:15,000 I I bid -45 billion. For open AI. 588 00:33:15,640 --> 00:33:25,920 For first for to go first. For first take, OK, I'll go 589 00:33:26,120 --> 00:33:30,200 I'll, I'll go to 50. Sure, I'll go. 590 00:33:30,960 --> 00:33:37,800 I'll go 55. I'll go. 60 60 billion -60 591 00:33:37,800 --> 00:33:40,240 billion. How is opening eye going to have 592 00:33:40,240 --> 00:33:42,040 an exit? I just am confused by this. 593 00:33:42,480 --> 00:33:45,080 It does they they don't need to have an exit, they just need to 594 00:33:45,080 --> 00:33:46,680 have a fundraise. OK. 595 00:33:46,680 --> 00:33:58,040 I'll go. I'll go to 70. 75 OK, that's 596 00:33:58,040 --> 00:33:59,680 fine. Max, you're done. 597 00:33:59,880 --> 00:34:00,760 I'm done. I'm out. 598 00:34:00,920 --> 00:34:05,440 You got it at 75, all right. 1st pick You have a lunatic open AI. 599 00:34:05,960 --> 00:34:10,560 OK, bad pick. Bad pick now I. 600 00:34:11,000 --> 00:34:13,400 Imagine I did all that to get like anthropic or? 601 00:34:13,560 --> 00:34:16,440 I just, I just I just think the number of companies that are 602 00:34:16,440 --> 00:34:19,120 worth like hundreds of billions of dollars on the secondary 603 00:34:19,120 --> 00:34:21,320 market is just extremely low. So you're you're basically 604 00:34:21,320 --> 00:34:24,639 discounting the net equity value to like 5 or 15 billion now or 605 00:34:24,639 --> 00:34:27,360 whatever it is. So sorry, yeah, added. 606 00:34:28,040 --> 00:34:29,320 Yeah. It's currently valued at 80 or 607 00:34:29,320 --> 00:34:31,159 90 right on secondary markets and. 608 00:34:31,520 --> 00:34:35,520 You took 7075 off that. So you're basically saying I'm, 609 00:34:35,960 --> 00:34:39,400 you know 1010 is where I my starting point and I have to get 610 00:34:39,400 --> 00:34:41,760 another 100 billion or whatever on this valuation? 611 00:34:41,760 --> 00:34:44,400 If it's, is it a trillion dollar company like? 612 00:34:45,639 --> 00:34:47,320 Yeah. Then I crushed it like. 613 00:34:47,320 --> 00:34:49,760 But are they going to be a trillion dollar company as a 614 00:34:49,760 --> 00:34:52,199 subsidiary of Microsoft? I just don't believe that. 615 00:34:52,199 --> 00:34:54,239 But that's fine. This is why it's an exciting, 616 00:34:54,239 --> 00:34:55,840 contentious pick. Yeah, great. 617 00:34:56,040 --> 00:35:00,000 OK, All right. Eric's going to go 1st, and Max 618 00:35:00,000 --> 00:35:01,680 and I are flipping for 2nd place. 619 00:35:02,080 --> 00:35:04,280 Max, name your head or tails here. 620 00:35:04,760 --> 00:35:05,880 I'll take heads. Heads. 621 00:35:06,080 --> 00:35:08,840 You're up second. OK, Eric, we know what you're 622 00:35:08,840 --> 00:35:10,160 picking. Do you want to make it official? 623 00:35:10,600 --> 00:35:14,040 Open AI Congratulations knew what he was doing. 624 00:35:15,720 --> 00:35:17,120 Smart guy. Smart guy. 625 00:35:17,120 --> 00:35:20,040 Anything you want to, Anything you want to add there, or. 626 00:35:20,400 --> 00:35:24,080 Well, I I do also think the fact I think Microsoft be an investor 627 00:35:24,080 --> 00:35:27,640 helps drive up the valuation because they're getting spend 628 00:35:27,640 --> 00:35:30,400 back on their system. There's all this revenue round 629 00:35:30,400 --> 00:35:32,480 tripping going on. I think it'll definitely raise 630 00:35:32,480 --> 00:35:35,920 another round or two. So I feel like it's super safe 631 00:35:35,920 --> 00:35:37,440 bet. I think I'm getting it if we 632 00:35:37,440 --> 00:35:39,400 know it's reported to be 90 billion. 633 00:35:39,840 --> 00:35:41,800 I'm getting it like, yeah, with. With. 634 00:35:41,840 --> 00:35:46,040 But if Microsoft buys, if Microsoft buys an additional 2% 635 00:35:46,040 --> 00:35:49,520 to go from 49% ownership to 51% ownership, that's that's like an 636 00:35:49,520 --> 00:35:51,400 exit in my opinion. Like that's an exit. 637 00:35:51,480 --> 00:35:52,600 I agree. Yeah, yeah. 638 00:35:52,640 --> 00:35:55,320 What do you mean no? They have majority ownership of 639 00:35:55,320 --> 00:35:57,080 Open AI. I mean it's it's a scenario of 640 00:35:57,080 --> 00:35:59,840 Microsoft at that point. No, it's just over then, like. 641 00:36:00,360 --> 00:36:03,280 Yeah, I think. So no, these are if there are 642 00:36:03,320 --> 00:36:06,280 deals on Open AI, they value it as more. 643 00:36:06,480 --> 00:36:09,880 It's worth more, definitely. If other people, if people are 644 00:36:09,880 --> 00:36:11,720 getting shares at a higher price. 645 00:36:12,240 --> 00:36:14,880 There there will be no shares of Open AI, though in this context 646 00:36:14,920 --> 00:36:16,760 I don't know we can we can see what actually plays. 647 00:36:16,760 --> 00:36:19,960 At it, the valuation goes up and there are people who own a lot 648 00:36:19,960 --> 00:36:23,960 of it, then yeah, it it has its own valuation. 649 00:36:23,960 --> 00:36:26,760 There's no way that you're saying, well, well, I I I guess 650 00:36:27,120 --> 00:36:29,960 the shares have to dissolve is what you're saying for it to be 651 00:36:29,960 --> 00:36:32,200 an acquisition that's kind of. Yeah, yeah. 652 00:36:32,240 --> 00:36:35,120 Once they're in Microsoft, I accept that. 653 00:36:35,120 --> 00:36:38,120 Like I don't get to ride up. The Microsoft benefit. 654 00:36:38,400 --> 00:36:40,440 Of. Even though I do think that's a 655 00:36:40,440 --> 00:36:42,840 possible outcome, but. I I just don't think they ever 656 00:36:42,840 --> 00:36:46,400 make it to whatever a trillion dollar valuation is a stand 657 00:36:46,400 --> 00:36:48,160 alone. I don't know if Sam Altman would 658 00:36:48,240 --> 00:36:51,200 sell, oh, you know, away Control to Microsoft. 659 00:36:51,200 --> 00:36:53,960 That's I don't think so either And like there's antitrust and 660 00:36:53,960 --> 00:36:56,440 like I think this is the most Microsoft is probably going to 661 00:36:56,440 --> 00:36:58,120 own. So, but I I'm not willing to 662 00:36:58,120 --> 00:37:01,880 accept like if they own 60%, I'm like no, that's fair, that's 663 00:37:01,880 --> 00:37:02,360 fair. OK. 664 00:37:02,560 --> 00:37:04,280 All right, all right. All right, Mac. 665 00:37:04,280 --> 00:37:09,920 Max. I will take data bricks at #2 666 00:37:11,120 --> 00:37:15,440 Yeah smart, pick smart I. Like I like Data Bricks, I I 667 00:37:15,440 --> 00:37:20,480 like to think that I mean the goal here is obviously to really 668 00:37:20,480 --> 00:37:24,000 get a a couple huge winners or one really huge winner, right. 669 00:37:24,000 --> 00:37:27,600 And I think that Data Bricks is obviously already a very large 670 00:37:27,600 --> 00:37:30,320 company, you know 10s of billions of dollars and I think 671 00:37:30,720 --> 00:37:33,720 there's probably many 10s more or or potentially hundreds in 672 00:37:33,720 --> 00:37:36,120 headroom there and so. Yeah, I don't know. 673 00:37:36,320 --> 00:37:38,360 I think it's a good call option on a really massive company. 674 00:37:38,360 --> 00:37:41,440 So, So I'm grabbing data bricks and also my favorite speaker 675 00:37:41,440 --> 00:37:46,360 from CV1 and Ali goes to the CEO and he's coming back for CV2. 676 00:37:46,360 --> 00:37:47,800 So I'm just a little Homer pick here. 677 00:37:48,560 --> 00:37:50,640 Smart. Yeah, this is giving me now. 678 00:37:50,640 --> 00:37:52,760 My stomach's turning. I do data bricks like I've been 679 00:37:52,760 --> 00:37:54,720 super bullish about them for the long time. 680 00:37:54,720 --> 00:37:57,840 So not to have them in my list is it hurts. 681 00:37:58,160 --> 00:38:01,920 All right, James, pick 3. All right, I've got two that 682 00:38:01,960 --> 00:38:04,160 I'm. You get two in a row, Snake 683 00:38:04,160 --> 00:38:05,240 draft. You get two in a row. 684 00:38:05,240 --> 00:38:06,880 I get two. OK, I just do both of them. 685 00:38:07,280 --> 00:38:07,760 Yes. All right. 686 00:38:08,400 --> 00:38:12,920 I am going to go with hugging face as my first pick. 687 00:38:13,560 --> 00:38:15,640 Wow, that's that's high. That's high. 688 00:38:15,640 --> 00:38:18,440 I think you could have gotten the value bet later in the later 689 00:38:18,440 --> 00:38:20,040 in the draft there. No, I might have. 690 00:38:20,040 --> 00:38:23,120 No, it sounded like Eric was going was eyeing A hugging face 691 00:38:23,120 --> 00:38:25,360 as well. Yeah, I think. 692 00:38:25,360 --> 00:38:27,720 I just, I just really think Hugging Face is going to exit it 693 00:38:27,720 --> 00:38:30,280 under 20 to Microsoft basically. So that's well that's good for 694 00:38:30,280 --> 00:38:33,360 me though that's like a pretty good lock in of 20 million of 695 00:38:33,360 --> 00:38:35,600 you. Need you need to stack those 696 00:38:35,600 --> 00:38:37,440 hundreds to win this draft, James. 697 00:38:37,720 --> 00:38:40,160 We're going up against the trillion dollar open AI here. 698 00:38:40,360 --> 00:38:43,320 So then you should only pick foundation model companies, you. 699 00:38:43,320 --> 00:38:45,720 Should pick whatever you think can go to a trillion Eric, you 700 00:38:45,720 --> 00:38:48,960 know it could be anything hugging face has a lot of room 701 00:38:48,960 --> 00:38:53,000 for I think potentially getting to that 100 billion level if 702 00:38:53,320 --> 00:38:58,760 they kind of maintain with their position in in essential you 703 00:38:58,760 --> 00:39:02,240 know position in the open open source community and they 704 00:39:02,320 --> 00:39:06,640 potentially even build tooling around this that helps other 705 00:39:06,640 --> 00:39:10,640 companies use open source in their in their enterprises. 706 00:39:10,640 --> 00:39:15,880 And yeah, I just think there's a pretty high ceiling there. 707 00:39:17,000 --> 00:39:20,240 GitHub sold for 7.5 billion. Yeah I'm just, I'm just saying 708 00:39:20,240 --> 00:39:23,960 GitHub is is everything hugging faces and more in my opinion and 709 00:39:23,960 --> 00:39:26,440 they decided to exit at the top at 7, 1/2. 710 00:39:26,440 --> 00:39:28,440 So I don't know. I mean, I I love hugging face 711 00:39:28,440 --> 00:39:31,680 for the record, we're, you know, big fans, but I don't know, not 712 00:39:31,680 --> 00:39:33,080 not sure about the pick on the upside. 713 00:39:33,160 --> 00:39:35,280 And there are forum at the end of the day like. 714 00:39:36,400 --> 00:39:37,400 Yeah. All right, James, second pick, 715 00:39:37,400 --> 00:39:38,800 hosting. I I mean, we love them. 716 00:39:38,880 --> 00:39:43,480 But for my second pick, I am going with Anthropic. 717 00:39:44,760 --> 00:39:47,920 Huge second runner in the foundation. 718 00:39:47,920 --> 00:39:52,920 Model Wars, I would argue, just recently struck a pretty major 719 00:39:52,920 --> 00:40:00,360 partnership with Amazon and has a lot of the same people who 720 00:40:00,360 --> 00:40:02,160 were originally working at Open AI. 721 00:40:02,720 --> 00:40:06,880 So I think, yeah, and I've heard really, you know, positive 722 00:40:06,880 --> 00:40:09,040 things from developers about Claude in general. 723 00:40:09,040 --> 00:40:11,520 So yeah, I'm really excited about Anthropic. 724 00:40:12,520 --> 00:40:17,720 OK, this is a tough pick. I'm going to, I'm going to skip 725 00:40:17,720 --> 00:40:20,040 through the really high fundraisers here for my second 726 00:40:20,120 --> 00:40:25,600 pick, go a little more deviant. I'm going to take Pine Cone, the 727 00:40:25,680 --> 00:40:29,960 database company. I think that what I'm hearing 728 00:40:30,200 --> 00:40:34,760 from my little birdies, Little birdies in the, in the AI 729 00:40:34,760 --> 00:40:40,280 communities, Pine Cone is the the database of choice for AI 730 00:40:40,280 --> 00:40:44,320 developers of all stripes. And I think if you look at the 731 00:40:44,320 --> 00:40:48,040 history of Silicon Valley, you see really strong exits from a 732 00:40:48,040 --> 00:40:49,520 lot of different database companies. 733 00:40:50,360 --> 00:40:54,680 So yeah, I mean I I like Pine Cone as I don't know, I mean 734 00:40:54,680 --> 00:40:57,040 this would be a really good pick if I could get you know 735 00:40:57,360 --> 00:40:59,720 discounted for the fact that their valuation is way lower 736 00:40:59,720 --> 00:41:00,960 than most of the companies on this list. 737 00:41:00,960 --> 00:41:03,120 But regardless, I'll take the upside. 738 00:41:03,600 --> 00:41:07,120 I, you know, as a reporter does went went to a VC for advice 739 00:41:07,120 --> 00:41:09,360 before this. I got one because I asked at the 740 00:41:09,360 --> 00:41:13,840 last minute but he said pine cone not there on his list. 741 00:41:13,840 --> 00:41:18,360 He only gave me his five because I think her #3 enterprises are 742 00:41:18,360 --> 00:41:21,720 wasting tons of time building toy prototypes that will 743 00:41:21,720 --> 00:41:26,400 ultimately be replaced. All I all I care about is just 744 00:41:26,400 --> 00:41:29,040 the crazy upside case and I think that's that's what I'm 745 00:41:29,040 --> 00:41:30,440 going for here, so all. Right. 746 00:41:30,440 --> 00:41:32,200 I'm gonna do I get 2 now? I don't know. 747 00:41:32,200 --> 00:41:33,440 Do you get 2 now? Do you get to? 748 00:41:33,480 --> 00:41:34,920 Yeah, you get 2. Now you get to swing. 749 00:41:34,920 --> 00:41:36,480 Back. I'm like, nervous. 750 00:41:36,600 --> 00:41:43,920 I'm gonna take inflection. Please explain. 751 00:41:46,720 --> 00:41:48,480 You got to defend your picks here, Yeah? 752 00:41:52,000 --> 00:41:57,600 With the idea one sort of top talent obviously like DeepMind 753 00:41:57,600 --> 00:42:00,760 Co founder Mustafa Suleiman. I'm not picking him using 754 00:42:00,760 --> 00:42:06,280 speaking but you know I I do think super legit company. 755 00:42:06,280 --> 00:42:09,120 I mean obviously you're betting a lot on the future. 756 00:42:09,320 --> 00:42:14,080 I do think you know if it's go for big swings, huge foundation 757 00:42:14,080 --> 00:42:20,880 model potential is key smart technology I think. 758 00:42:21,240 --> 00:42:24,240 I think also just like give it, I feel like there's a floor and 759 00:42:24,240 --> 00:42:27,880 that it's also like a potential like acquisition to just like 760 00:42:27,880 --> 00:42:33,080 get some big tech company capable worst case. 761 00:42:33,080 --> 00:42:38,160 So yeah, I'm going inflection for #2, OK. 762 00:42:39,000 --> 00:42:42,800 And this is a tough one. All right. 763 00:42:42,800 --> 00:42:45,760 I'm taking character. AII Just feel oh man. 764 00:42:45,920 --> 00:42:49,840 I feel like that bad pick bad. Pick. 765 00:42:49,840 --> 00:42:51,160 You would have taken it, right? Yeah. 766 00:42:51,680 --> 00:42:53,000 Bad. You would have taken it. 767 00:42:53,000 --> 00:42:54,480 Or the pig bad. No, no, no. 768 00:42:54,480 --> 00:42:57,240 Bad pick is just a way I'd joke about every pick being bad. 769 00:42:57,240 --> 00:42:59,640 Sorry, would you have taken character like I never would 770 00:42:59,640 --> 00:43:01,360 have, but it sounds like James would have so. 771 00:43:01,760 --> 00:43:03,640 I think I would have taken it, yes. 772 00:43:04,080 --> 00:43:09,440 I'm going on a foundation model, forward strategy and I I think 773 00:43:09,440 --> 00:43:13,400 character, you know, has. Perhaps a clearer path to a 774 00:43:13,400 --> 00:43:17,280 product than most, I think. I mean they have a very talented 775 00:43:17,760 --> 00:43:21,840 CEOI mean Theresa Horowitz is going big on them. 776 00:43:21,880 --> 00:43:24,120 So what they're going to get lots of follow on funding 777 00:43:24,120 --> 00:43:26,360 rounds. Yeah, we'll see. 778 00:43:26,360 --> 00:43:31,600 I don't know. I you know something I I still 779 00:43:31,600 --> 00:43:35,720 think like the, you know, right now the bot sort of relationship 780 00:43:35,720 --> 00:43:37,520 bot thing is a little overhyped, but. 781 00:43:38,880 --> 00:43:40,640 You know, there's time. What do you think about 782 00:43:40,640 --> 00:43:44,120 character as it is today? What is your? 783 00:43:44,320 --> 00:43:47,640 Do you agree with The Take that it's all sex chat or or? 784 00:43:47,640 --> 00:43:49,720 Not, I don't, I don't know, I'm not like an expert on the 785 00:43:49,720 --> 00:43:52,120 product. But to me there there's sort of 786 00:43:52,120 --> 00:43:55,280 you got a dual bet in that they're super hardcore building 787 00:43:55,280 --> 00:43:59,360 their own foundation model deep on technology and they're 788 00:43:59,440 --> 00:44:02,040 actually have a product that's out there that people are 789 00:44:02,040 --> 00:44:05,240 finding fun ways to use. And like, I mean, you know, Mark 790 00:44:05,240 --> 00:44:07,800 Zuckerberg, that's sort of with hot or not so like anything. 791 00:44:08,080 --> 00:44:11,640 That that they have the sort of instinct to say like let's get 792 00:44:11,640 --> 00:44:15,400 users, that that to me is like a good, good muscle. 793 00:44:15,400 --> 00:44:18,720 And like, even if this first iteration isn't like the one you 794 00:44:18,720 --> 00:44:21,520 want the New York Times to be writing about, I think I think 795 00:44:21,520 --> 00:44:25,360 it's a good sign. I think it's a good sort of out 796 00:44:25,360 --> 00:44:26,760 of the money. You guys would not have. 797 00:44:26,800 --> 00:44:28,040 Picked. Inflection though? 798 00:44:29,480 --> 00:44:31,360 No, I I think. Inflection's a reasonable I like 799 00:44:31,360 --> 00:44:32,480 your picks. I like your picks. 800 00:44:32,880 --> 00:44:34,240 I don't. I honestly don't really know 801 00:44:34,240 --> 00:44:36,840 what these foundation models, which is why I'm going to pick a 802 00:44:36,840 --> 00:44:42,320 foundation model next, baby. Yeah. 803 00:44:42,320 --> 00:44:44,720 I mean, I literally think I could be choosing foundation 804 00:44:44,720 --> 00:44:48,120 models. Pass open AI and anthropic as a 805 00:44:48,120 --> 00:44:50,680 a monkey. Throwing darts at the proverbial 806 00:44:50,680 --> 00:44:53,120 board would do better than me here, but I'll just go with Co 807 00:44:53,120 --> 00:44:54,440 here. I'll just take the market 808 00:44:54,440 --> 00:44:57,720 deciding it's worth a lot. It seems like it's the number 3 809 00:44:57,720 --> 00:45:02,360 #4 best foundation model. Who the F knows is my defense of 810 00:45:02,360 --> 00:45:05,280 this pick, but I got to grab a foundation. 811 00:45:05,960 --> 00:45:08,680 Everybody wants to be in business, so I feel like they're 812 00:45:08,680 --> 00:45:10,040 going to face a ton of competition. 813 00:45:10,040 --> 00:45:13,080 They were early, sure. Yeah, I'm. 814 00:45:13,120 --> 00:45:14,080 I'm not. I'm, I'm. 815 00:45:14,080 --> 00:45:15,680 I'm freely willing to admit this with them. 816 00:45:15,680 --> 00:45:18,080 I'm I'm freely willing to admit this is my lowest conviction 817 00:45:18,080 --> 00:45:19,600 pick here. But I just feel like I got to 818 00:45:19,600 --> 00:45:21,840 have a foundation model, and that to me is the best pick on 819 00:45:21,840 --> 00:45:23,200 the board. So so be it. 820 00:45:23,440 --> 00:45:24,960 Onward and upward. Maybe I'll get lucky. 821 00:45:25,040 --> 00:45:26,720 OK, I'm. I'm ready with my 2 picks. 822 00:45:27,040 --> 00:45:28,320 All right? Please fire away. 823 00:45:28,480 --> 00:45:33,920 All right. 1st pick go go on with another Foundation model of 824 00:45:33,920 --> 00:45:35,120 course. Got to have one. 825 00:45:35,120 --> 00:45:40,040 I will say you can see why VCs are investing big in foundation 826 00:45:40,040 --> 00:45:43,520 models, because if the goal is to go for the fences, you're 827 00:45:43,520 --> 00:45:48,520 like you know, foundation models are are are you know big swings. 828 00:45:48,840 --> 00:45:51,800 And so, like nobody here is really so far trying to 829 00:45:51,800 --> 00:45:57,240 accumulate like a but like I could name some sure things but 830 00:45:57,240 --> 00:45:59,720 then. You know you want like a swing 831 00:45:59,800 --> 00:46:04,000 anyway, James. I'm going with AI 21 Labs, which 832 00:46:04,000 --> 00:46:08,840 is a Israeli foundation model company. 833 00:46:09,160 --> 00:46:13,200 Solid. Just just announced $155,000,000 834 00:46:13,200 --> 00:46:20,480 raise in August at a $1.4 billion valuation and. 835 00:46:21,680 --> 00:46:25,840 The. Participants included Google and 836 00:46:25,920 --> 00:46:29,480 NVIDIA. They create foundation models 837 00:46:29,480 --> 00:46:34,840 particularly targeted at writing, and they are known for 838 00:46:34,840 --> 00:46:37,040 their cutting edge Jurassic models. 839 00:46:37,680 --> 00:46:41,160 So yeah, I'm impressed with the founding team and the product. 840 00:46:41,640 --> 00:46:42,840 Yeah. And they they've got some, I 841 00:46:42,840 --> 00:46:44,840 think they have a partnership maybe with Amazon, like I've 842 00:46:44,840 --> 00:46:48,080 seen them come up, but it feels like they're under the the radar 843 00:46:48,200 --> 00:46:49,640 a little bit. I think it's a little bit of an 844 00:46:49,640 --> 00:46:53,680 under the radar pick, but a solid that. 845 00:46:53,680 --> 00:46:55,920 Makes sense? Foundation models, I mean I 846 00:46:55,920 --> 00:47:01,440 would say that is, you know, if if they are Israeli and they're 847 00:47:01,440 --> 00:47:04,400 valued at 1.4, probably in the US they'd be valued at 14 848 00:47:04,440 --> 00:47:08,280 billion. So I think, yeah, exactly like 849 00:47:09,560 --> 00:47:13,600 you know that that seems like a good, good, good cost of living 850 00:47:13,600 --> 00:47:16,960 adjustment right there. All right and with my next pick 851 00:47:17,120 --> 00:47:21,880 I am taking Replet. I'm Jod. 852 00:47:21,880 --> 00:47:26,920 Yeah, another CVCVAI alumni Alumnus if you will. 853 00:47:28,120 --> 00:47:31,360 So replica. You have ugly Face and replica. 854 00:47:31,360 --> 00:47:32,800 They're on the same panel together. 855 00:47:32,800 --> 00:47:36,000 You're just a you're just a big CVAI chapter here. 856 00:47:36,080 --> 00:47:38,360 I love it. I like to meet the founders and 857 00:47:38,360 --> 00:47:46,400 get to know them. Practicing for In the end, it's 858 00:47:46,400 --> 00:47:48,080 really, it's really just a bet on the founders. 859 00:47:48,080 --> 00:47:49,680 It's really about the people. It's about the people. 860 00:47:49,680 --> 00:47:52,320 It's really about the founders. What do you think about this 861 00:47:52,320 --> 00:47:54,480 stuff? Yeah, Founder market fit, if you 862 00:47:54,480 --> 00:47:55,920 will. All right, What? 863 00:47:55,920 --> 00:47:57,880 What's your? I'm definitely as founder. 864 00:47:57,880 --> 00:48:01,400 Twitter are are repla and hogging face like in the like 865 00:48:01,400 --> 00:48:03,560 trillion dollar case. They're like fighting with each 866 00:48:03,560 --> 00:48:05,400 other right? They're both like in the coding. 867 00:48:05,400 --> 00:48:06,560 World, I don't. I don't agree with that. 868 00:48:06,560 --> 00:48:10,040 I think that repla, the trillion dollar case here is that 869 00:48:10,160 --> 00:48:15,000 software engineering. Changes dramatically due to AI 870 00:48:15,440 --> 00:48:20,040 and we enter this world of everyone's an engineer, you 871 00:48:20,080 --> 00:48:25,440 you're using a cloud based IDE, you can use AI really 872 00:48:25,440 --> 00:48:28,040 aggressively as you're coding and you don't even need to 873 00:48:28,040 --> 00:48:30,640 really learn how to code. And I think they're the ones who 874 00:48:30,640 --> 00:48:34,560 are furthest along at kind of building that vision that I, I 875 00:48:34,560 --> 00:48:37,800 do believe is possible with AI that was that was not on my 876 00:48:37,800 --> 00:48:41,120 draft board, all right. Yeah, I don't think I would have 877 00:48:41,120 --> 00:48:44,200 picked it. Yeah, I'm going to go with 878 00:48:44,200 --> 00:48:47,800 another Homer Pig. I will be taking Modular, 879 00:48:48,720 --> 00:48:51,440 another founder. I will be interviewing on stage 880 00:48:51,440 --> 00:48:54,120 at CVA I2 for the for the back story here. 881 00:48:54,480 --> 00:48:57,200 For the back story here. Chris Lattner, who Co founded 882 00:48:57,200 --> 00:49:01,320 Modular, literally invented the Swift programming language that 883 00:49:01,320 --> 00:49:05,720 is on every iPhone, every Mac, every iPad in the world. 884 00:49:05,800 --> 00:49:08,760 Probably one of the top five to 10 most used programming 885 00:49:08,760 --> 00:49:12,360 languages ever invented. So he's obviously a beast. 886 00:49:12,360 --> 00:49:17,160 He has shown tremendous facility in creating really low level 887 00:49:17,160 --> 00:49:20,440 technical products that are used by literally hundreds of 888 00:49:20,440 --> 00:49:22,000 millions of people, if not billions. 889 00:49:22,040 --> 00:49:24,880 Before that he created a tool called Low Level Virtual 890 00:49:24,880 --> 00:49:26,440 Machines that is also basically everywhere. 891 00:49:26,680 --> 00:49:31,560 So I'm just saying Chris Lattner, he gets WS and he is 892 00:49:31,560 --> 00:49:34,880 trying to reinvent the programming language for AI and 893 00:49:34,880 --> 00:49:37,720 replace Python with a new language called the Mojo that 894 00:49:37,720 --> 00:49:40,480 modular is invented. And so, look, I frankly don't 895 00:49:40,480 --> 00:49:43,120 know what the business model is here or where we're going with 896 00:49:43,120 --> 00:49:46,720 this part, but you know, Chris Ladner puts points on the 897 00:49:46,720 --> 00:49:48,200 scoreboard. So that's what I'm hoping he'll 898 00:49:48,200 --> 00:49:49,760 do for my team here. So modular. 899 00:49:49,760 --> 00:49:54,120 AII am breathing like a huge sigh of relief because I took 900 00:49:54,120 --> 00:49:56,800 character instead of this company thinking the character 901 00:49:56,800 --> 00:50:01,040 was going to be a more. Competitive pick and I think you 902 00:50:01,040 --> 00:50:04,800 were right, proved true and this company did not get scooped up. 903 00:50:05,160 --> 00:50:09,200 I am taking what I would have actually probably put as my 904 00:50:09,200 --> 00:50:14,280 third pick, but strategically dropped to 4 clean which is sort 905 00:50:14,280 --> 00:50:17,800 of doing corporate search. Sequoia Back Company, 906 00:50:17,840 --> 00:50:23,600 Lightspeed, Ravi, the key guy at Lightspeed is on it, and Mamoon 907 00:50:23,600 --> 00:50:26,280 Hamid at Kleiner Perkins is an investor in this company. 908 00:50:26,560 --> 00:50:29,280 So I just feel like it's one of these, Like, I don't know. 909 00:50:29,280 --> 00:50:32,640 What do they all know? So you were really like a 910 00:50:32,640 --> 00:50:35,120 venture capital fundamentalist here. 911 00:50:35,120 --> 00:50:40,400 Like, yeah, like, you know, Eric's like, have you ever heard 912 00:50:40,400 --> 00:50:42,520 of that fallacy of appealing to authority? 913 00:50:42,520 --> 00:50:44,840 Well, I go the other way and say it's the way. 914 00:50:45,080 --> 00:50:46,440 Well. Like, you know, like you should 915 00:50:46,480 --> 00:50:50,400 always appeal to authority. I mean what we don't have the 916 00:50:50,400 --> 00:50:52,560 financials, what are you doing? You're looking for a signal 917 00:50:52,560 --> 00:50:54,280 like. I mean what is better signal 918 00:50:54,280 --> 00:50:57,200 than like? Sequoia, Mamoon and Ravi like I, 919 00:50:57,240 --> 00:50:59,120 you know I don't know what it is. 920 00:50:59,200 --> 00:51:02,760 I think they're good VCs and like I I don't have metrics So 921 00:51:02,760 --> 00:51:06,920 what am I going And honestly my VC guru that I went to here's 922 00:51:06,920 --> 00:51:10,400 his case for glean semantic search over business data is the 923 00:51:10,400 --> 00:51:14,280 biggest use case. This opportunity overshadows all 924 00:51:14,280 --> 00:51:17,520 the in house builds people are using link chain etcetera for. 925 00:51:17,600 --> 00:51:21,600 So instead of instead of cobbling it together yourself. 926 00:51:22,200 --> 00:51:26,320 Use glean like I, I, I buy that. Wasn't, wasn't this the Tanium 927 00:51:26,320 --> 00:51:29,120 pitch, which also like you know, is sort of, Oh yeah, laying 928 00:51:29,120 --> 00:51:30,440 around? You know you should. 929 00:51:30,560 --> 00:51:33,280 You need to look into the eyes of the Co founders, right? 930 00:51:33,280 --> 00:51:35,640 Like, are you gonna actually help your investors make money 931 00:51:35,640 --> 00:51:40,160 out of this company anyway? OK, well there you go. 932 00:51:40,440 --> 00:51:43,080 OK. So now I have a last one, This 933 00:51:43,080 --> 00:51:47,080 one I'm I'm debating between like product that's solid, 934 00:51:47,080 --> 00:51:51,560 protect my sort of minimum versus like upside. 935 00:51:51,920 --> 00:51:56,840 Potential man, this is like, I I have empathy for VCs. 936 00:51:56,840 --> 00:52:01,880 It's like, should I go crazy, like hoping to dream, you know, 937 00:52:02,280 --> 00:52:06,640 or should I go sort of like product that people actually use 938 00:52:07,160 --> 00:52:09,480 today. I'd like to think you would do 939 00:52:09,480 --> 00:52:11,640 more than 10 minutes of research, which is what we've 940 00:52:11,640 --> 00:52:13,400 done on this. I don't know. 941 00:52:13,400 --> 00:52:16,040 Did do VCs do that? I don't know though. 942 00:52:16,200 --> 00:52:17,880 They get, they get the financials. 943 00:52:17,880 --> 00:52:19,840 We don't get any. I guess some of these companies 944 00:52:19,840 --> 00:52:23,040 haven't like you. You already did what most VCs 945 00:52:23,040 --> 00:52:25,880 do, which is as their VC. Friends who they should invest 946 00:52:26,160 --> 00:52:30,920 in, just go with that. It works. 947 00:52:31,560 --> 00:52:34,280 Lots of people, a lot of money doing that. 948 00:52:35,680 --> 00:52:40,680 All right, I. I'm going for the the hope and 949 00:52:40,680 --> 00:52:43,120 dream of the entire European continent. 950 00:52:43,280 --> 00:52:47,040 Mistral AI Oh, my gosh. I thought you were going to say 951 00:52:47,040 --> 00:52:48,800 something else, but yeah. Yeah. 952 00:52:49,120 --> 00:52:51,680 Vladimir Zelensky. All right. 953 00:52:52,720 --> 00:52:55,640 Mistral is, you know, super hypey. 954 00:52:55,640 --> 00:53:00,840 I think Lightspeed is an investor European, you know, 955 00:53:00,920 --> 00:53:07,000 they they're very early, super, super hypey, but you know. 956 00:53:07,320 --> 00:53:10,400 Smart, smart founders. I'm really enjoying this draft 957 00:53:10,400 --> 00:53:12,720 because I really would not have picked almost any of the 958 00:53:12,720 --> 00:53:14,360 companies you guys picked. And I feel interesting. 959 00:53:14,360 --> 00:53:17,360 That must apply to me too. So yeah, we can review each 960 00:53:17,360 --> 00:53:19,880 other's teams at the end here. I know who these. 961 00:53:19,880 --> 00:53:21,640 Six would be. So I'll tell you at the end I'm 962 00:53:21,640 --> 00:53:25,920 gonna, I'm gonna I'm I am also going to go hoping to dream here 963 00:53:25,920 --> 00:53:30,760 a little bit more. So I obviously read you know 964 00:53:30,760 --> 00:53:33,360 read and researched all the founders of these companies and 965 00:53:33,720 --> 00:53:37,840 I was very taken by this founder's vision of self 966 00:53:37,840 --> 00:53:42,480 improving foundation models. So the idea of recursive 967 00:53:42,560 --> 00:53:46,200 improvement of a foundation model and so that was the pitch 968 00:53:46,200 --> 00:53:53,360 of imbue our friend Ken who on a hope and a dream and I assume 969 00:53:53,360 --> 00:53:57,720 some really solid demos has raised $234 million. 970 00:53:57,880 --> 00:54:03,440 So so but I think if you're going for true take off trillion 971 00:54:03,440 --> 00:54:07,400 dollar companies as it were the first person to figure out how 972 00:54:07,400 --> 00:54:10,960 to make AI recursively self improving Will has a good shot 973 00:54:10,960 --> 00:54:12,040 at being a trillion dollar company. 974 00:54:12,040 --> 00:54:15,480 So I'll, I'll, I'll put some some chips on Kenjon and see 975 00:54:15,480 --> 00:54:16,280 what happens. Yeah. 976 00:54:16,360 --> 00:54:18,840 I think they're very bullish on like agents, right, I mean, 977 00:54:18,880 --> 00:54:21,560 right and self improving agents as well which I thought was 978 00:54:21,560 --> 00:54:25,720 exciting. So I'll throw, I'll throw $5 979 00:54:25,720 --> 00:54:29,480 into that $234 million pot and help them get there. 980 00:54:29,680 --> 00:54:32,560 All right, James, Final pick, final pick. 981 00:54:32,880 --> 00:54:35,800 All right. With my final pick, I am taking 982 00:54:35,840 --> 00:54:42,160 adept, interesting, very highly valued cerebral values. 983 00:54:42,360 --> 00:54:46,240 CV alum, Yeah. Obviously super high valued and 984 00:54:46,600 --> 00:54:50,360 highly funded. As far as I can tell, they've 985 00:54:50,360 --> 00:54:52,880 raised, yeah, close to half a billion dollars, right? 986 00:54:53,400 --> 00:54:56,360 Woo Hoo, Hoo Hoo. Essentially they are they are 987 00:54:56,360 --> 00:55:03,080 trying to create. A new way to interact, Have AIS 988 00:55:03,200 --> 00:55:06,200 interact with computers. Essentially allow creating 989 00:55:06,200 --> 00:55:08,120 agents that can control computers. 990 00:55:08,600 --> 00:55:12,800 I don't even know I I put this as a copilot I guess, but I 991 00:55:12,800 --> 00:55:15,920 don't even know if this is like this. 992 00:55:15,920 --> 00:55:19,040 Seems like a brand new type of concept. 993 00:55:19,040 --> 00:55:22,200 You know, creating a agents, AI agents that can interact with 994 00:55:22,200 --> 00:55:26,360 computers on your behalf. It's a pretty science fictiony 995 00:55:26,360 --> 00:55:31,160 idea, so you know, could be a. Huge flap. 996 00:55:31,160 --> 00:55:33,520 Or it could change the way you know the world. 997 00:55:33,520 --> 00:55:35,720 Works founders have left in droves. 998 00:55:35,720 --> 00:55:38,400 I feel like lots of internal turmoil. 999 00:55:39,640 --> 00:55:43,080 I feel like. It was hiked earlier other found 1000 00:55:43,080 --> 00:55:46,480 you know there there was a founder break up at the company 1001 00:55:47,600 --> 00:55:51,600 which which is certainly a negative sign but little scary 1002 00:55:52,240 --> 00:55:55,400 but but you never know stop off from from the ashes rises the 1003 00:55:55,400 --> 00:55:57,600 phoenix. I mean the. 1004 00:55:58,160 --> 00:56:02,800 The CEO is, you know, David Luan, a former I think VP at 1005 00:56:02,800 --> 00:56:05,920 Open AI. Certainly is there sort of early 1006 00:56:05,920 --> 00:56:09,720 days, super smart guy. But yeah, can't. 1007 00:56:09,720 --> 00:56:11,840 Can't keep the original team together. 1008 00:56:12,280 --> 00:56:15,400 Interesting, I just think. I just I think working on AI 1009 00:56:15,400 --> 00:56:20,320 agents in some capacity that can like to your earlier point about 1010 00:56:20,320 --> 00:56:26,000 self self improving AI and you know reinforcement or maybe Max 1011 00:56:26,000 --> 00:56:29,040 you were talking about that. I think there's something there 1012 00:56:29,040 --> 00:56:32,280 that could unlock a lot of value, so I'm excited that 1013 00:56:32,280 --> 00:56:35,040 they're working in that area. Sure, got it. 1014 00:56:35,360 --> 00:56:40,640 My zoom out I'm I'm most stressed that I should have just 1015 00:56:41,320 --> 00:56:44,040 hoped for a second and gotten data bricks for. 1016 00:56:44,080 --> 00:56:46,760 Cheap because I do think next you got a great deal on this 1017 00:56:46,760 --> 00:56:49,600 Section 1 there, yeah? And data bricks is like a pretty 1018 00:56:49,840 --> 00:56:52,920 sure thing, almost more sure than opening eye. 1019 00:56:52,920 --> 00:56:58,040 I feel like the upside. Our producer says he's picking 1020 00:56:58,080 --> 00:57:00,880 the script because we're editing the podcast with that and also 1021 00:57:00,880 --> 00:57:04,720 that was one I was seriously considering for five working 1022 00:57:04,720 --> 00:57:08,840 product, but I didn't know how much technical under like what 1023 00:57:08,840 --> 00:57:11,360 their level of. Heck, sophistication was so I 1024 00:57:11,360 --> 00:57:11,960 didn't pick it. What? 1025 00:57:11,960 --> 00:57:14,640 What are, what are other ones on the board, left on the board, 1026 00:57:14,760 --> 00:57:16,360 Anything you guys are? Surprised, notable. 1027 00:57:16,360 --> 00:57:19,480 I mean, Jasper obviously is like, everybody's afraid of, 1028 00:57:19,600 --> 00:57:22,760 yeah, rapper. I feel like Snorkel has gotten a 1029 00:57:22,760 --> 00:57:25,000 lot of hype with like internal data cleaning. 1030 00:57:25,000 --> 00:57:26,760 There's companies. That's what I've heard. 1031 00:57:26,760 --> 00:57:28,720 Yeah. Yeah, I they're they're always 1032 00:57:28,720 --> 00:57:30,280 bouncing around when I'm hearing. 1033 00:57:31,000 --> 00:57:34,200 Writer raised a bunch of money at 500,000,000, so I'm 1034 00:57:34,200 --> 00:57:37,640 interested to see sort of there was sort of a good price. 1035 00:57:38,520 --> 00:57:40,000 Lots of money back. Lots of money. 1036 00:57:40,000 --> 00:57:43,200 Good price for investors. It's it's interesting, humane. 1037 00:57:43,200 --> 00:57:45,920 Which makes that little pin that's supposed to replace your 1038 00:57:45,920 --> 00:57:48,680 iPhone that you wear on your jacket or whatever that you talk 1039 00:57:48,680 --> 00:57:50,680 to. Nobody put that on the board. 1040 00:57:51,240 --> 00:57:55,400 That seems like a a tough, tough sell Hard, but. 1041 00:57:55,560 --> 00:57:57,840 Hardware is hard. I just hardware is very hard. 1042 00:57:57,960 --> 00:57:58,720 Yeah. Yeah. 1043 00:57:58,720 --> 00:58:01,000 They just launched it at Paris Fashion Week, which is odd. 1044 00:58:01,000 --> 00:58:03,360 I don't. Know I mean scale you know was 1045 00:58:03,360 --> 00:58:06,800 making bank off of. Growing humans and self driving 1046 00:58:06,800 --> 00:58:10,400 car problems and now they're trying to make bank at human 1047 00:58:10,400 --> 00:58:13,040 reinforcement learning for foundation models. 1048 00:58:13,400 --> 00:58:16,840 But it's just hard to believe, you know, it's humans. 1049 00:58:16,840 --> 00:58:18,600 Like they just don't seem like that. 1050 00:58:19,200 --> 00:58:20,120 You want? We don't need. 1051 00:58:20,120 --> 00:58:21,240 We don't need them anymore. We. 1052 00:58:21,520 --> 00:58:23,600 Don't need. I know I want a big like don't 1053 00:58:23,880 --> 00:58:27,520 need this middle ground thing. I mean stability not being in 1054 00:58:27,520 --> 00:58:30,640 the discussion, you know compared to last six months ago 1055 00:58:30,640 --> 00:58:33,760 is is quite interesting. Obviously there's been, well, 1056 00:58:33,760 --> 00:58:36,000 first of all, there's been a ton of competition and you know, Mid 1057 00:58:36,000 --> 00:58:37,760 Journey has gotten super good and all that stuff. 1058 00:58:37,760 --> 00:58:39,800 And then secondly, yeah, obviously there was a story 1059 00:58:39,800 --> 00:58:41,640 about the founder making a bunch of stuff up, which was 1060 00:58:41,640 --> 00:58:45,600 interesting. I think the journey is was left 1061 00:58:45,600 --> 00:58:48,280 off the board here because they haven't taken VC funding by the 1062 00:58:48,280 --> 00:58:50,960 way, but they could be a super valuable. 1063 00:58:51,400 --> 00:58:53,880 Yeah, I know hard to maybe never get a valuation into the cash 1064 00:58:53,880 --> 00:58:56,400 flow business, but mid journey man I would definitely have 1065 00:58:56,400 --> 00:59:01,560 taken that before Mistral I think interesting the I mean. 1066 00:59:01,880 --> 00:59:04,600 Yeah, I think weights and biases really good company. 1067 00:59:04,600 --> 00:59:07,320 But I'm just like, what sort of, is it really going to gobble the 1068 00:59:07,320 --> 00:59:10,800 world? Runway Love the founder. 1069 00:59:10,800 --> 00:59:14,240 Like he's so charming. But, you know, I feel like 1070 00:59:14,240 --> 00:59:18,360 maybe, you know, the narrative there is like, does Adobe wish 1071 00:59:18,360 --> 00:59:21,200 they could have bought Runway for like way less than it paid 1072 00:59:21,200 --> 00:59:23,360 for Figma? You know, when Runway is sort of 1073 00:59:23,360 --> 00:59:27,800 the relevant company now? Yeah, I mean, lots of great 1074 00:59:27,840 --> 00:59:30,000 companies. I wouldn't write off most of 1075 00:59:30,000 --> 00:59:32,240 them. I don't even know what go 1076 00:59:32,240 --> 00:59:33,560 student is. I didn't look at him. 1077 00:59:33,760 --> 00:59:37,360 That's the most high fund rate. Did anybody research Go Student? 1078 00:59:37,640 --> 00:59:41,280 I researched it a bit. So my understanding is they're a 1079 00:59:41,840 --> 00:59:46,000 European company who kind of got started in this AI in this 1080 00:59:46,000 --> 00:59:50,680 tutoring space and raised a lot of money and now is building LMS 1081 00:59:50,680 --> 00:59:53,000 to do tutoring. But I was a little skeptical of, 1082 00:59:53,000 --> 00:59:56,880 like, whether that was a, you know, current product or not or 1083 00:59:57,320 --> 00:59:59,120 prototype. Yeah. 1084 00:59:59,120 --> 01:00:00,640 So that's why I kind of ruled them out. 1085 01:00:00,960 --> 01:00:02,960 My last point would just be, I was really struck in doing the 1086 01:00:02,960 --> 01:00:05,960 research how unbelievably difficult it is to even tell 1087 01:00:05,960 --> 01:00:08,120 what the products of a lot of these companies are or how 1088 01:00:08,120 --> 01:00:10,720 they're differentiated. And I mean, you could really 1089 01:00:10,720 --> 01:00:15,800 just like throw buzzwords like large model learning like 1090 01:00:15,800 --> 01:00:18,360 reinforcement something something internal data, 1091 01:00:18,360 --> 01:00:21,760 external data, whatever. It just spit out a new Tech 1092 01:00:21,760 --> 01:00:24,040 Crunch article with like a fake company named him every day and 1093 01:00:24,040 --> 01:00:27,280 I would find them very similar to a lot of the research I did 1094 01:00:27,280 --> 01:00:28,960 for this. So it's. 1095 01:00:29,480 --> 01:00:31,840 I mean poolside, you know, like I've written about, it's for 1096 01:00:31,840 --> 01:00:33,760 code. I mean partially it's an 1097 01:00:33,800 --> 01:00:36,720 earlier, it's like earlier on the foundation model journey. 1098 01:00:36,720 --> 01:00:38,800 But yeah, there are a lot where it's like, man, it is. 1099 01:00:38,840 --> 01:00:43,760 It is remarkable how many similar ish like either Data or 1100 01:00:43,760 --> 01:00:47,840 Foundation model or something about enterprise plus LL Ms. 1101 01:00:47,840 --> 01:00:49,600 companies there are that are in this category. 1102 01:00:49,600 --> 01:00:53,400 It's it's amazing. To me there's my my zoom out on 1103 01:00:53,880 --> 01:00:59,240 my view is that like Max, I mean a lot will fall on data bricks 1104 01:00:59,240 --> 01:01:02,640 of course. And that anthropic could be the 1105 01:01:02,680 --> 01:01:05,600 people plausibly think anthropic could beat open AI. 1106 01:01:05,600 --> 01:01:10,440 In which case, like, James got a great pick discounted open 1107 01:01:10,480 --> 01:01:12,440 anthropic. Anthropic is the only one I like 1108 01:01:12,440 --> 01:01:13,720 from James. Yeah, really. 1109 01:01:15,200 --> 01:01:17,320 Anthropic is when I'm. I want. 1110 01:01:17,360 --> 01:01:20,040 I like your, I like your I like your Israeli one. 1111 01:01:20,160 --> 01:01:22,160 AI 21 also that's kind of interesting. 1112 01:01:22,160 --> 01:01:22,600 Yeah. What? 1113 01:01:22,720 --> 01:01:24,480 Which ones do you like from Eric? 1114 01:01:24,480 --> 01:01:27,560 I guess the, I mean opening eye obviously is untouchable, but I 1115 01:01:27,560 --> 01:01:30,960 think I I think inflection is also quite strong. 1116 01:01:31,640 --> 01:01:33,880 The rest, I'm not so sure. Yeah, character. 1117 01:01:33,960 --> 01:01:36,000 Man. And whereas your your team's 1118 01:01:36,000 --> 01:01:38,920 just rock solid, my my team is Immaculate. 1119 01:01:39,200 --> 01:01:41,880 No, I would say other than databrakes, I have no sure 1120 01:01:41,880 --> 01:01:45,880 things but I really took I took call options on hopefully giant 1121 01:01:45,880 --> 01:01:47,680 upside. Like I could see my team having, 1122 01:01:47,960 --> 01:01:49,840 I could see my team having three zeros on it. 1123 01:01:49,840 --> 01:01:53,000 But as long as the two are big, I'm, I'm, I'm going for the win 1124 01:01:53,000 --> 01:01:56,600 that way, yeah. We we really had to do a 1125 01:01:57,600 --> 01:02:02,240 lightning, a speed run of of learning VCs portfolio strategy 1126 01:02:02,240 --> 01:02:04,320 here. We we do not have. 1127 01:02:04,960 --> 01:02:07,480 The date, you know any we don't get to talk to anybody. 1128 01:02:07,480 --> 01:02:09,800 Weird is severe. This is not real VC. 1129 01:02:09,800 --> 01:02:12,480 We were playing basically with blindfolds on, but it was very 1130 01:02:12,480 --> 01:02:15,720 enjoyable conversation piece. It's kind of funny that we had 1131 01:02:15,720 --> 01:02:17,960 to like pick companies that could return the fund. 1132 01:02:17,960 --> 01:02:20,960 You know, they all have to. Like, yeah, runs, right. 1133 01:02:21,000 --> 01:02:21,760 Yeah. All right. 1134 01:02:22,880 --> 01:02:24,600 Very excited to see you in five years. 1135 01:02:24,600 --> 01:02:27,200 Hopefully we'll follow up and like some of us will look like 1136 01:02:27,200 --> 01:02:29,640 maybe all of us will look like idiots in this space will be 1137 01:02:30,480 --> 01:02:34,680 wildly overhyped and maybe my, you know, -75 billion will like 1138 01:02:34,680 --> 01:02:37,920 swamp the total value and then we can just. 1139 01:02:38,120 --> 01:02:41,200 I don't think that's likely, but as a closer or do you just want 1140 01:02:41,200 --> 01:02:44,280 to read off everyone's team? Oh yeah, For people not 1141 01:02:44,280 --> 01:02:49,440 watching, the teams are. Team Eric with a negative $75 1142 01:02:49,440 --> 01:02:52,920 billion handicap that I paid to go first so I could scoop up 1143 01:02:53,080 --> 01:02:58,960 Open AI #1, Inflection #2, Character A, Character AI #3, 1144 01:02:59,320 --> 01:03:04,600 Glean #4, and Mistral AI #5. Max you want to read yours? 1145 01:03:05,000 --> 01:03:07,640 Yeah. My team is Databricks, Pine 1146 01:03:07,640 --> 01:03:12,800 Cone, Cohere Modular, and IMBU. Also known as the best team. 1147 01:03:13,480 --> 01:03:16,600 And James. I went with Hugging Face, 1148 01:03:16,600 --> 01:03:21,640 Anthropic, AI, 21 Labs, Replet, and Adept. 1149 01:03:22,240 --> 01:03:27,120 That's our episode, and that's the Cerebral Valley AI series. 1150 01:03:28,000 --> 01:03:31,880 Watch for YouTube videos. I'm sure I'll have Max and James 1151 01:03:31,880 --> 01:03:35,760 once I'm like not fatigued of talking back to sort of reflect 1152 01:03:35,760 --> 01:03:38,760 on it, but I'm not promising it immediately. 1153 01:03:38,760 --> 01:03:42,800 But this has been a great. Run Thank you for sticking with 1154 01:03:42,800 --> 01:03:44,200 us. And you missed some of the 1155 01:03:44,200 --> 01:03:47,080 earlier episodes. Go back and listen to them. 1156 01:03:48,120 --> 01:03:51,080 Thank you so much to Scott Brody, who's been the producer 1157 01:03:51,080 --> 01:03:54,160 through all this and help help us figure out the show. 1158 01:03:54,160 --> 01:03:57,720 So shout out to him. Thanks to Riley Kinsella, Gabby 1159 01:03:57,720 --> 01:04:01,960 Caliendo, both key and sort of getting the conference and 1160 01:04:02,240 --> 01:04:05,560 everything together. Thank you for everyone who's 1161 01:04:05,560 --> 01:04:10,400 going to speak. And yeah, stick around. 1162 01:04:10,880 --> 01:04:14,800 Please, like, comment, subscribe on YouTube, subscribe to 1163 01:04:14,800 --> 01:04:19,520 newcomer.co and go play song quiz on your Alexa. 1164 01:04:19,520 --> 01:04:20,280 All right? Yeah. 1165 01:04:20,960 --> 01:04:22,880 Thanks so much. Thanks guys. 1166 01:04:23,080 --> 01:04:24,360 Thank you all. Right. 1167 01:04:24,480 --> 01:04:24,640 Bye.