WEBVTT 00:00:04.000 --> 00:00:08.840 You're surrounded by journals. Hold your position. 00:00:18.400 --> 00:00:22.200 Overnight success. 00:00:30.400 --> 00:00:33.800 Double leg, 00:00:35.120 --> 00:00:38.120 >> right? 00:00:48.240 --> 00:00:51.560 That's misinformation. 00:00:55.040 --> 00:00:58.920 >> Clearing order inbound. 00:00:59.840 --> 00:01:03.320 >> Let's just roll. 00:01:03.920 --> 00:01:05.757 >> We are surrounded by journals. Hold your 00:01:05.840 --> 00:01:11.320 position. >> Come, get up. 00:01:13.600 --> 00:01:17.240 >> Trust the experts. 00:01:18.480 --> 00:01:21.480 Five. 00:01:23.360 --> 00:01:26.840 We are 00:01:28.800 --> 00:01:31.800 founder. 00:01:34.080 --> 00:01:37.400 >> Five good. 00:01:37.520 --> 00:01:43.080 >> I see multiple journalists on the horizon. Stand by. 00:01:47.440 --> 00:01:50.100 UAV online. [music] 00:01:52.000 --> 00:01:55.000 >> Blaze. 00:01:56.800 --> 00:02:00.200 Double blaze. 00:02:01.680 --> 00:02:05.240 Triple blaze. 00:02:05.760 --> 00:02:09.160 Double kill. 00:02:24.320 --> 00:02:27.800 Cook is 00:02:35.600 --> 00:02:38.920 team deathmatch. 00:02:43.760 --> 00:02:47.400 We are experts. 00:02:49.040 --> 00:02:52.600 Triple blades. 00:02:53.600 --> 00:02:59.160 Let's just roll. Right. 00:03:01.840 --> 00:03:06.040 Mark clearing order inbound. 00:03:08.480 --> 00:03:11.040 Get up. 00:03:16.080 --> 00:03:20.840 We are surrounded by journalists. Hold your position. 00:03:25.600 --> 00:03:28.920 >> Strike one. 00:03:30.480 --> 00:03:37.480 >> Strike two. Activate 00:03:38.720 --> 00:03:43.410 golden retriever mode. [music] 00:03:52.800 --> 00:03:57.080 >> Mark clearing order inbound. 00:03:58.160 --> 00:04:01.560 >> Five good. 00:04:13.520 --> 00:04:19.040 I see multiple journalists on the horizon. Stand by. 00:04:21.040 --> 00:04:24.040 founder. 00:04:30.160 --> 00:04:32.157 >> You're watching >> You're watching TBPN. 00:04:32.240 --> 00:04:35.197 >> Today is Thursday, June 4th, 2026. We 00:04:35.280 --> 00:04:40.317 are live from Palunteer AIPCON. Uh with the Temple of Technology is back in 00:04:40.400 --> 00:04:44.077 Fortress of Finance. We will return to it, but it is also a state of mind. Um 00:04:44.160 --> 00:04:47.677 that's right. We are also sponsored by RAMP. Time is money. Save both. Easy to 00:04:47.760 --> 00:04:50.317 use corporate cards, bill pay, accounting, and a whole lot more all in 00:04:50.400 --> 00:04:53.117 one place. Big news from RAMP today. 00:04:53.200 --> 00:04:57.037 Massive fundraising a little bit. But first, we got to talk 00:04:57.120 --> 00:05:00.557 >> Oh, is it still going? I like it. The Ramp song's back. 00:05:00.640 --> 00:05:04.957 >> This was This was early days. We really talked about RAM so much. Turned it into 00:05:05.040 --> 00:05:07.517 a song. Anyway, uh the topic of 00:05:07.600 --> 00:05:10.957 conversation in DC has uh it's still in 00:05:11.040 --> 00:05:15.677 AI world, but instead of talking about uh approving models before they're 00:05:15.760 --> 00:05:19.757 released today, it's about the bio threat. Brandon Guerell wrote in the 00:05:19.840 --> 00:05:24.717 TBPN newsletter today, the great houses of AI have united behind the bio threat. 00:05:24.800 --> 00:05:28.157 There's actually a lot more to that because it was a big long list of 00:05:28.240 --> 00:05:33.597 signitories from AI, but also from the bio world and biotech and even startups. 00:05:33.680 --> 00:05:36.717 We've seen former guests of the show sign on. I'm excited to bring some of 00:05:36.800 --> 00:05:39.197 those folks back on the show in the coming weeks and hear more about this 00:05:39.280 --> 00:05:42.877 because I have this belief that as AI 00:05:42.960 --> 00:05:46.717 advanced, we got cyber because it was such a tight feedback loop, such a tight 00:05:46.800 --> 00:05:50.157 verifiable reward. Reinforcement learning works really well in that 00:05:50.240 --> 00:05:53.117 context. Bio has some similar characteristics. 00:05:53.200 --> 00:05:56.317 >> And it was a very tangible Y2K style 00:05:56.400 --> 00:06:01.277 moment. Exactly. Where there was it was a let's just say powerful business 00:06:01.360 --> 00:06:04.557 strategy. >> Yeah. Is it Yeah. It was like is it over? You start thinking about the 00:06:04.640 --> 00:06:08.877 consequence of this and you don't need to get to AGI super intelligence god. 00:06:08.960 --> 00:06:12.557 You can just have a really powerful tool that creates a new problem and that uh 00:06:12.640 --> 00:06:15.997 creates full employment for Nicashura over at Palo Alto Networks who we had a 00:06:16.080 --> 00:06:20.077 chance to talk to yesterday and uh he's been very fortunate in implementing the 00:06:20.160 --> 00:06:24.157 solutions to the cyber security threats posed by new AI systems some of the new 00:06:24.240 --> 00:06:26.397 AI capabilities they're rolling out. Um, 00:06:26.480 --> 00:06:28.637 but bio might be next. And so it's 00:06:28.720 --> 00:06:31.517 exciting to see that uh the great houses 00:06:31.600 --> 00:06:35.757 of AI are uniting behind the bio threats. So let's take you through this. 00:06:35.840 --> 00:06:38.877 First I'm going to tell you about console.com. Console builds AI agents 00:06:38.960 --> 00:06:42.637 that automate 70% of IT HR and finance support giving employees instant 00:06:42.720 --> 00:06:47.117 resolution for access requests and password resets. So uh in 1981 a group 00:06:47.200 --> 00:06:52.157 of researchers published the primary structure of the polio virus genome in 00:06:52.240 --> 00:06:53.757 the journal Nature. So they're basically 00:06:53.840 --> 00:06:57.197 open sourcing the sequence for making 00:06:57.280 --> 00:07:00.157 polio which just a few years earlier 00:07:00.240 --> 00:07:04.957 polio I think was on the decline by 1981 but uh a very very problematic virus. 00:07:05.040 --> 00:07:06.957 It's an RNA virus meaning that its 00:07:07.040 --> 00:07:11.117 nucleiobases or building blocks are ACGU 00:07:11.200 --> 00:07:13.997 if you're familiar with uh RNA adsine 00:07:14.080 --> 00:07:16.477 cytosine guanine and uricil. Uh put more 00:07:16.560 --> 00:07:21.117 plainly, thanks Brandon Guerell, uh he says when the researchers published the 00:07:21.200 --> 00:07:24.397 primary structure of the polio virus, they gave the world the literal sequence 00:07:24.480 --> 00:07:28.637 of polio virus building blocks in order from start to finish. By the mid- 20th 00:07:28.720 --> 00:07:32.797 century, before mass vaccination, polio was paralyzing and killing more than 00:07:32.880 --> 00:07:36.557 half a million people per year worldwide. So you have this pretty 00:07:36.640 --> 00:07:39.677 deadly virus killing more than half a million people per year worldwide. And 00:07:39.760 --> 00:07:43.437 you have just open sourced it. What happens? So in 2002, researchers 00:07:43.520 --> 00:07:48.397 synthesized infectious polio virus from its publicly available sequence data. So 00:07:48.480 --> 00:07:52.477 they didn't actually need any of the polio virus RNA to start. They didn't 00:07:52.560 --> 00:07:53.917 need it on hand. They didn't need to. 00:07:54.000 --> 00:07:56.637 It's not like they took a little sample and they just cloned it up and made it 00:07:56.720 --> 00:08:00.637 bigger. They they just took the data and they made the actual virus. So this is 00:08:00.720 --> 00:08:04.637 the this is the shape of the threat. If there's a new uh if there's a new virus 00:08:04.720 --> 00:08:08.317 or an existing virus or a forgotten about virus and you have the code to it, 00:08:08.400 --> 00:08:12.477 you can potentially print that RNA and then have the virus in your hands. Even 00:08:12.560 --> 00:08:16.397 if you don't have a sample, you weren't able to kill a sample. So instead, these 00:08:16.480 --> 00:08:20.477 researchers in 2002, they were able to take the published sequence, chemically 00:08:20.560 --> 00:08:25.277 synthesize short DNA fragments, assemble them into full length a full length DNA 00:08:25.360 --> 00:08:30.237 copy of the polio virus genome, and then use the DNA to make the viral RNA to 00:08:30.320 --> 00:08:34.157 fully recover the infectious virus. So in 2005, researchers used these same 00:08:34.240 --> 00:08:35.757 technologies to reconstruct the Spanish 00:08:35.840 --> 00:08:40.077 flu, a virus in 1918 that killed 675,000 00:08:40.160 --> 00:08:44.717 Americans and had a 2 to 3% mortality rate among those infected. Very, very 00:08:44.800 --> 00:08:48.957 dangerous stuff. So basically, these two reconstructed viruses showed that having 00:08:49.040 --> 00:08:53.117 a physical virus on hand was no longer necessary as source material to create 00:08:53.200 --> 00:08:55.837 viruses. All you needed was the blueprints. as long as you have the code 00:08:55.920 --> 00:08:59.837 literally just like texted in a text file a bunch of ATGU 00:08:59.920 --> 00:09:03.517 uh you can go and make this as long as you have the equipment on hand but that 00:09:03.600 --> 00:09:07.837 is getting democratized as well and that's what this uh AI letter is all 00:09:07.920 --> 00:09:11.517 about. So that's a situation that we're still in today except now that we have 00:09:11.600 --> 00:09:14.557 AI there are easier ways to potentially 00:09:14.640 --> 00:09:18.637 reconstruct DNA sequences that could create new viruses. So yesterday, Demis 00:09:18.720 --> 00:09:22.797 Hassabis, Sam Alman, Dario Amade, Alex Wang, and dozens of other high-profile 00:09:22.880 --> 00:09:26.797 leaders across AI, tech policy, nucleic acid synthesis, and biotech signed an 00:09:26.880 --> 00:09:31.197 open letter called in support of mandatory nucleic acid synthesis, 00:09:31.280 --> 00:09:35.277 screening, and recordkeeping. You might have seen it on the timeline. And at 00:09:35.360 --> 00:09:39.437 first glance, Brandon here assumed, and I assumed the same thing, assumed it was 00:09:39.520 --> 00:09:42.317 another press release from a Frontier Lab claiming it had just discovered new 00:09:42.400 --> 00:09:45.677 capabilities in one of its internal models that would ultimately lead to 00:09:45.760 --> 00:09:49.277 cata catastrophe. A lot of this like doom fear-based marketing has been 00:09:49.360 --> 00:09:51.837 happening. So, that was sort of the natural reaction. Uh, and that's what 00:09:51.920 --> 00:09:54.157 Brandon >> Some people's reaction would be, were we 00:09:54.240 --> 00:09:56.637 not doing recordkeeping here already? 00:09:56.720 --> 00:09:59.917 >> That's a great question and Brandon actually did answer that. But, it's not 00:10:00.000 --> 00:10:03.277 just a PR stunt and it's not a new capability. They're not saying that the 00:10:03.360 --> 00:10:05.757 models can just create a novel virus, 00:10:05.840 --> 00:10:10.397 you know, one shot like that that that that is solved yet. Uh it's not there, 00:10:10.480 --> 00:10:14.397 but they see it as something that's coming down the pipe. And this letter is 00:10:14.480 --> 00:10:18.957 not this dangerous new capability. It's more asking the US government to force 00:10:19.040 --> 00:10:21.277 nucleic acid synthesis companies to 00:10:21.360 --> 00:10:24.077 screen orders for sequences of concern. 00:10:24.160 --> 00:10:26.557 So, hey, somebody just ordered this. 00:10:26.640 --> 00:10:30.957 Looks a lot like a virus. Like, what are we doing here? uh you said that you were 00:10:31.040 --> 00:10:34.317 trying to treat cancer or you said that you were, you know, trying to make a new 00:10:34.400 --> 00:10:37.037 peptide and all of a sudden you're asking for polio virus or something that 00:10:37.120 --> 00:10:40.637 looks like polio virus. Like let's let's dig into this. Uh that's where they're 00:10:40.720 --> 00:10:45.117 going with that. And so uh they also need to verify the legit the legitimacy 00:10:45.200 --> 00:10:49.517 of the customer and to keep a record of what they're sending and to whom. That's 00:10:49.600 --> 00:10:53.277 a crazy one that I'm sure you're like, wait, they weren't keeping records. They 00:10:53.360 --> 00:10:55.517 were a little bit. Uh he gets into this. 00:10:55.600 --> 00:10:58.157 So he says the reason the letter is coming out now is that the threat of 00:10:58.240 --> 00:11:02.317 nucleic acid synthesis sequences sequencing getting into the wrong hands 00:11:02.400 --> 00:11:07.277 has been enhanced by AI. So anyone with an AI tool in the future could in theory 00:11:07.360 --> 00:11:11.757 if the models don't have safeguards on them could uh synthes could create a 00:11:11.840 --> 00:11:15.597 sequence that then they go to a nucleic acid sequence company get printed send 00:11:15.680 --> 00:11:20.077 it to them mix it up boom they got a virus not good. So uh most of the global 00:11:20.160 --> 00:11:23.837 nucleic acid synthesis industry has already signed up to do some of this. 00:11:23.920 --> 00:11:26.477 They did they started this in 2009 with 00:11:26.560 --> 00:11:28.317 what's called the international gene 00:11:28.400 --> 00:11:31.517 synthesis consortium and roughly 80% of 00:11:31.600 --> 00:11:33.917 commercial synthesis capacity worldwide 00:11:34.000 --> 00:11:40.237 is on is is on board but membership >> 20% [laughter] still just hanging out 00:11:40.320 --> 00:11:43.997 we're good >> 80% of nuclear weapons are safely 00:11:44.080 --> 00:11:46.077 stored. Don't ask about the other 20%. 00:11:46.160 --> 00:11:47.597 That's kind of what this letter is 00:11:47.680 --> 00:11:49.997 getting at because 80% it was a good 00:11:50.080 --> 00:11:53.917 first effort 2009 it's been 16 17 years. 00:11:54.000 --> 00:11:56.877 Uh we haven't had >> Yeah. But there's a new reason to go 00:11:56.960 --> 00:12:01.357 further. Let's get that last 20%. That's what they're asking for. So um 00:12:01.440 --> 00:12:05.037 >> me membership is not a strong guarantee that they're actually screening or 00:12:05.120 --> 00:12:08.557 keeping records of their customers because it's voluntary. The 80% number 00:12:08.640 --> 00:12:12.717 is also self-reported, for example, and a bunch of other factors contribute to 00:12:12.800 --> 00:12:15.917 the relative climsiness of the agreement. 00:12:16.000 --> 00:12:18.797 >> So, it's not >> you can opt into this program by just 00:12:18.880 --> 00:12:22.317 saying that you're opting into it, but then even the reporting once you're 00:12:22.400 --> 00:12:25.917 opted in is voluntary. So I I think the way this works is the international gene 00:12:26.000 --> 00:12:27.837 synthesis consortium is probably a 00:12:27.920 --> 00:12:31.117 nonprofit NGO you know non-governmental 00:12:31.200 --> 00:12:34.797 organization and every all the companies 00:12:34.880 --> 00:12:37.277 uh they they volunteer 80% of commercial 00:12:37.360 --> 00:12:40.157 synthesis volume has opted into this and 00:12:40.240 --> 00:12:44.877 then this organization the international gene synthesis consortium they say hey 00:12:44.960 --> 00:12:49.117 we've looked at the market and we're covering about 80% has opted into this 00:12:49.200 --> 00:12:52.957 we're we're on board with 80% and and the government isn't coming in and 00:12:53.040 --> 00:12:55.757 checking the records. They're not actually saying, "Okay, well, we we have 00:12:55.840 --> 00:12:58.957 a different number because we're the government and you have your this 00:12:59.040 --> 00:13:02.397 number. Let's verify this number." It's self-reported by that organization, but 00:13:02.480 --> 00:13:05.997 there's no reason not to trust that organization necessarily. Um, so what 00:13:06.080 --> 00:13:08.397 else? >> Bunch of other factors contribute to the 00:13:08.480 --> 00:13:10.717 relative flimsiness of this agreement. 00:13:10.800 --> 00:13:13.997 Uh, HHS also has guidance in place around the issue, but again, it's 00:13:14.080 --> 00:13:17.117 voluntary, meaning that the possibility of bad actors getting their hands on 00:13:17.200 --> 00:13:20.477 dangerous nucleic acid sequences, at least from American companies, still 00:13:20.560 --> 00:13:22.397 cannot be ruled out. Overall, it's good 00:13:22.480 --> 00:13:24.397 to see industry leaders signing this 00:13:24.480 --> 00:13:27.677 letter and doubly refreshing uh that the 00:13:27.760 --> 00:13:31.917 letter is not yet another warning of apocalyptic AI doom, which I think the 00:13:32.000 --> 00:13:35.357 public has unfortunately be come to expect from announcements like this. 00:13:35.440 --> 00:13:38.557 Hopefully, the relevant legislators are paying attention and can make this 00:13:38.640 --> 00:13:42.237 happen in short order. So, I thought that was a good uh a good breakdown and 00:13:42.320 --> 00:13:46.157 I agree with a lot of that. Um Andrew Curran also has some uh deep dive on 00:13:46.240 --> 00:13:49.677 this with some more uh of the signitories. He shares screenshots of 00:13:49.760 --> 00:13:51.837 all of these and it really is everyone. 00:13:51.920 --> 00:13:54.077 Uh yeah, Y Combinator, 00:13:54.160 --> 00:13:56.637 DeepMind, Microsoft, Interconnect AI, 00:13:56.720 --> 00:13:59.117 Harvard, uh tons of stuff. And then over 00:13:59.200 --> 00:14:01.997 in uh in the nucleic acid synthesis 00:14:02.080 --> 00:14:04.717 industry, you have Twist Bioscience, uh 00:14:04.800 --> 00:14:09.757 ANZA, Emerald Cloud Lab, and Kathleen McMahon from Boss is on here, former 00:14:09.840 --> 00:14:14.797 guest of the show. >> Um so, so uh good news, but uh obviously 00:14:14.880 --> 00:14:18.237 and just a early step. This is just an open letter to the government saying hey 00:14:18.320 --> 00:14:22.557 we think you should we want to support this we think that the government should 00:14:22.640 --> 00:14:26.077 start thinking about this. Uh the other news in the bio world, 00:14:26.160 --> 00:14:27.517 >> yeah, I mean the news is just that 00:14:27.600 --> 00:14:30.957 there's incredible momentum in biotech, 00:14:31.040 --> 00:14:35.597 early stage biotech. Uh after >> momentum, but not like volume, not scale 00:14:35.680 --> 00:14:37.837 yet because you're looking at $3 00:14:37.920 --> 00:14:40.477 trillion IPOs going out this year 00:14:40.560 --> 00:14:44.077 potentially. So much news in AI, microns 00:14:44.160 --> 00:14:48.237 out of trillion. Every chip stock is, you know, in the hundreds of billions, 00:14:48.320 --> 00:14:52.797 trillions. This is much smaller, but >> but it's notable because biotech had 00:14:52.880 --> 00:14:56.957 been uh left for dead in some ways. We had a 00:14:57.040 --> 00:15:01.757 biotech investor on >> probably 14 months ago at this point who 00:15:01.840 --> 00:15:06.077 said I don't even know I mean just looking at the returns so far I don't 00:15:06.160 --> 00:15:08.317 know why you would invest in this asset 00:15:08.400 --> 00:15:13.357 class. Uh but of course every asset class kind of go throughs go uh goes 00:15:13.440 --> 00:15:16.157 through that kind of phase and clearly there's a lot of momentum 00:15:16.240 --> 00:15:19.837 >> and they should be you would expect that biotech would be similarly power law 00:15:19.920 --> 00:15:24.637 driven maybe not as extreme but if you pull out SpaceX open AI anthropic from 00:15:24.720 --> 00:15:28.237 capital >> yes but I feel like the biotech 00:15:28.320 --> 00:15:29.997 community has a little bit more of like 00:15:30.080 --> 00:15:35.277 a culture of like base hits doubles triples where they flip companies pretty 00:15:35.360 --> 00:15:39.837 pretty frequently in the >> Yeah we had that didn't we have a guy on that had sold 00:15:39.920 --> 00:15:43.277 like three companies. We didn't have when we were billion dollar exits. 00:15:43.360 --> 00:15:46.237 >> Yep. And then he joined another company and sold it for three billion like the 00:15:46.320 --> 00:15:47.837 next day. Uh and so 00:15:47.920 --> 00:15:49.757 >> so anyways, you have Isomorphic Labs 00:15:49.840 --> 00:15:53.677 spun out of uh Deep Mind uh Coinbase or 00:15:53.760 --> 00:15:58.637 not Coinbase, but but Brian spun out or like founded New Limit. Uh you have 00:15:58.720 --> 00:16:02.397 Retro >> a new round. We're gonna get Jacob on the show >> as well for that. 00:16:02.480 --> 00:16:05.197 >> Um Altos Labs uh 00:16:05.280 --> 00:16:08.477 >> from Jeff the Chad from Amazon. funny 00:16:08.560 --> 00:16:13.037 >> as this post puts it >> uh Anthropic obviously acquired 00:16:13.120 --> 00:16:16.637 coefficient bio as well >> but Jensen and Larry Ellison at Oracle 00:16:16.720 --> 00:16:20.317 are also doing stuff so there's a lot of activity it's very fun and I hope we're 00:16:20.400 --> 00:16:22.077 going to be able to cover this a lot more in the near 00:16:22.160 --> 00:16:26.557 >> future at what point do the 00:16:26.640 --> 00:16:28.797 uh at what point does like a Fizer or 00:16:28.880 --> 00:16:34.317 Johnson and John Johnson and Johnson start joining the press release economy 00:16:34.400 --> 00:16:37.517 of just coming I'm not I'm not saying it' be a good thing but coming coming 00:16:37.600 --> 00:16:42.317 out and saying, uh, we believe we're, you know, right at [laughter] the 00:16:42.400 --> 00:16:45.837 because there are partnerships all the time that happen and they're always just 00:16:45.920 --> 00:16:48.877 like tucked a little bit deeper in the Wall Street Journal because AI is 00:16:48.960 --> 00:16:52.477 dominating and even private credit takes the front seat to the bio news, but 00:16:52.560 --> 00:16:55.117 there there's a whole bunch of deal making going on. Anyway, there's other 00:16:55.200 --> 00:16:58.317 deal making going on in fintech. We're going to talk about ramps raised today. 00:16:58.400 --> 00:17:01.117 But first, I'm going to tell you about railway. Railway is the all-in-one 00:17:01.200 --> 00:17:04.157 intelligent cloud provider. Use your favorite agent to deploy web app 00:17:04.240 --> 00:17:07.197 servers, databases, and more. while railway automatically takes care of 00:17:07.280 --> 00:17:11.917 scaling, monitoring and security. Um, so ramp, >> what's going on in ramp land? 00:17:12.000 --> 00:17:14.157 >> 44 billion valuation. 00:17:14.240 --> 00:17:19.277 >> Really, really solid traction just, you know, every 12 18 months, sometimes much 00:17:19.360 --> 00:17:23.597 quicker. Sometimes they do two rounds in two weeks, but uh uh really solid 00:17:23.680 --> 00:17:26.237 progress. They raised $750 million at a 00:17:26.320 --> 00:17:32.397 $44 billion valuation. Last time we grew this fast, we were 120th of the size. So 00:17:32.480 --> 00:17:35.517 accelerating. >> This is the most notable thing to me. uh 00:17:35.600 --> 00:17:39.117 lots of chatter on the timeline around 00:17:39.200 --> 00:17:41.757 uh you know other fintech valuations. 00:17:41.840 --> 00:17:44.317 You compare them >> apocalypse. Well, yeah. You compare them 00:17:44.400 --> 00:17:47.437 to you know ramp is now worth more than PayPal. 00:17:47.520 --> 00:17:49.677 >> Okay. >> PayPal has 32 billion of revenue. 00:17:49.760 --> 00:17:52.157 >> Yeah. >> Uh but PayPal 00:17:52.240 --> 00:17:55.677 >> uh certainly has I would say ne you know 00:17:55.760 --> 00:17:57.517 probably negative momentum. Yeah. 00:17:57.600 --> 00:18:01.997 >> Uh whereas RAMP has incredible momentum and this this is the standout line. Uh 00:18:02.080 --> 00:18:06.637 they were 120th the size the last time they were growing this fast and so 00:18:06.720 --> 00:18:10.317 >> um yeah just really really really impressive execution. Yeah. 00:18:10.400 --> 00:18:12.477 >> Uh and 00:18:12.560 --> 00:18:14.637 incredible opportunity still. 00:18:14.720 --> 00:18:18.397 >> Yeah. Uh so Eric uh took to the timeline 00:18:18.480 --> 00:18:21.117 posted an essay about the third pillar 00:18:21.200 --> 00:18:24.717 comparing uh the previous eras of value 00:18:24.800 --> 00:18:27.357 creation. The two pillars uh people and 00:18:27.440 --> 00:18:30.877 vendors dating back to 600 BCE. If 00:18:30.960 --> 00:18:32.477 you're not thinking in millennia, what 00:18:32.560 --> 00:18:35.037 are you doing here? Tokens emerged as 00:18:35.120 --> 00:18:37.997 the third pillar in 2026 AD. And he 00:18:38.080 --> 00:18:40.397 calls it the quadrillion token blind 00:18:40.480 --> 00:18:43.277 spot. Boil down 500 years of finance. 00:18:43.360 --> 00:18:45.437 And it's really just three questions. 00:18:45.520 --> 00:18:48.477 Who spent what? Was it worth it? What's 00:18:48.560 --> 00:18:53.357 the bill next month? I [laughter] mean, people get caught up in all these crazy 00:18:53.440 --> 00:18:56.477 things. I mean, you see this in like marketing, I'm sure, and uh and ad 00:18:56.560 --> 00:18:58.477 buying where people will do all these 00:18:58.560 --> 00:19:02.637 crazy analyses and ROI ROAS and all this 00:19:02.720 --> 00:19:06.557 other stuff and like and it's always useful to zoom out and just be like, 00:19:06.640 --> 00:19:10.157 okay, we spent a bunch of money. Did the bank balance go up in this company or not? 00:19:10.240 --> 00:19:12.557 >> All all personal and business finance 00:19:12.640 --> 00:19:17.197 finance at the end eventually comes down to are we making more money than we're 00:19:17.280 --> 00:19:20.077 spending? >> Yeah. And I think uh yeah, Eric is is 00:19:20.160 --> 00:19:24.637 right to dive super deep into like token optimization and thinking about the 00:19:24.720 --> 00:19:27.757 tools that they're building, but then at the same time like not not don't get 00:19:27.840 --> 00:19:31.837 lost in the sauce and like actually zoom out and try and understand like what is 00:19:31.920 --> 00:19:33.837 the core value that you're delivering to 00:19:33.920 --> 00:19:38.957 your customer. Uh it is answering that question. Um so fantastic news over 00:19:39.040 --> 00:19:41.917 there. Let me tell you about the New York Stock Exchange. Want to change the 00:19:42.000 --> 00:19:46.397 world? Raise capital at the New York Stock Exchange. You got to do it. It's 00:19:46.480 --> 00:19:48.317 my number one advice for founders these 00:19:48.400 --> 00:19:50.797 days. Um, uh, there's some other 00:19:50.880 --> 00:19:53.677 fundraising news. Sabi, the Beanie BCI 00:19:53.760 --> 00:19:58.637 company is getting preempted at 35 million at 500 million post. Uh, this is 00:19:58.720 --> 00:20:01.117 a leak from R for Rock. We'll see where it goes. 00:20:01.200 --> 00:20:02.637 >> This is huge for you. >> Why? 00:20:02.720 --> 00:20:04.077 >> Because you are a beanie guy. 00:20:04.160 --> 00:20:07.597 >> I do like beanies. >> A beanie in the morning. 00:20:07.680 --> 00:20:09.357 >> Keeps it together. Yeah, I like a 00:20:09.440 --> 00:20:11.277 beanie. Uh very very 00:20:11.360 --> 00:20:16.797 >> I it's interesting. I think that uh this format 00:20:16.880 --> 00:20:20.557 >> Mhm. >> of course I'm sure they can adapt it to other types of hats. 00:20:20.640 --> 00:20:23.117 >> Yeah. >> But this format certainly 00:20:23.200 --> 00:20:25.437 >> maybe makes it harder to build momentum 00:20:25.520 --> 00:20:30.157 in places like California, at least Southern California, Arizona. 00:20:30.240 --> 00:20:31.837 >> Be big amongst creative directors though. 00:20:31.920 --> 00:20:35.197 >> Yeah. >> Huge. >> Huge. Huge potentially. Silver Lake. 00:20:35.280 --> 00:20:38.797 Silver Lake. Yeah. Every >> It's not too hard to change a beanie 00:20:38.880 --> 00:20:42.717 into a hat. cowboy hat. Like that's just extra leather around it. You could wrap 00:20:42.800 --> 00:20:44.717 the beanie in the in the cowboy hat. You can wear 00:20:44.800 --> 00:20:46.877 >> Here's what's interesting though. So R for Rock. 00:20:46.960 --> 00:20:50.957 >> Yeah. >> Uh usually >> it's pretty dialed. 00:20:51.040 --> 00:20:52.397 >> Pretty dialed. >> Pretty dile. 00:20:52.480 --> 00:20:55.437 >> Uh pretty dial. It's almost like he has inside information. [laughter] It's 00:20:55.520 --> 00:20:58.477 almost like he somehow got >> Yeah. But I mean, we've talked about the 00:20:58.560 --> 00:21:02.637 game theory of like do does he work at a real like tier one venture capital firm 00:21:02.720 --> 00:21:06.477 that's is like what's the benefit of leaking everything? Is he a lawyer 00:21:06.560 --> 00:21:08.957 that's seeing all the docs turn around? 00:21:09.040 --> 00:21:10.797 Zero benefit for a lawyer, >> right? 00:21:10.880 --> 00:21:15.837 >> Um the rush of getting likes on the timeline is pretty universal. You're a 00:21:15.920 --> 00:21:18.477 lawyer, you're just like, "Ah, I need a banger." 00:21:18.560 --> 00:21:20.317 >> Out of fund. Okay. >> For sure. 00:21:20.400 --> 00:21:24.397 >> Yeah. >> And uh I don't I don't know anything 00:21:24.480 --> 00:21:25.277 else. 00:21:25.360 --> 00:21:28.477 >> Um but uh he's always taken the view 00:21:28.560 --> 00:21:33.837 that it can be helpful to the founder to build because a bunch of people are 00:21:33.920 --> 00:21:37.757 going to see this. >> Sure. that that this didn't sort of land 00:21:37.840 --> 00:21:41.277 in their deal flow or land on their desk and they're going to reach out, right? 00:21:41.360 --> 00:21:43.997 So, it does create momentum. Um, 00:21:44.080 --> 00:21:48.397 >> but uh can certainly be annoying for teams as well. Uh, this was notable 00:21:48.480 --> 00:21:50.877 though. So, 200 million of of LOI from 00:21:50.960 --> 00:21:52.797 B2B customers 00:21:52.880 --> 00:21:57.597 >> and so very curious what the enterprise 00:21:57.680 --> 00:22:01.677 >> play is here. I don't know 00:22:01.760 --> 00:22:06.477 >> but uh we can work on getting a whole >> Does that mean like uh like through 00:22:06.560 --> 00:22:08.557 hospital networks or through like the 00:22:08.640 --> 00:22:13.677 health care system or is it like Mark Zuckerberg wants to go further? He wants 00:22:13.760 --> 00:22:17.677 to track the brain waves of the employees not just track your screen. 00:22:17.760 --> 00:22:21.517 >> We're also going to track your brain. I I mean it could go either way because 00:22:21.600 --> 00:22:23.837 you you imagine like Neurolink has had a 00:22:23.920 --> 00:22:25.837 bunch of traction and bunch of amazing I 00:22:25.920 --> 00:22:29.597 saw uh Nolan the uh the first patient P0 00:22:29.680 --> 00:22:34.797 uh on Rogan talking about playing COD with the Neurolink. Amazing. Uh and you 00:22:34.880 --> 00:22:37.837 can imagine that at a certain point like some sort of partnership. 00:22:37.920 --> 00:22:39.757 >> They have multiple hat form factors. 00:22:39.840 --> 00:22:43.197 >> There we go. Hats coming. >> We're good. I was getting really hung up 00:22:43.280 --> 00:22:47.357 on the beanie. I'm like there's so many different enterprise or or B2B contexts. 00:22:47.440 --> 00:22:51.437 You're in a You're in a warehouse in Dallas, Texas in the summer. 00:22:51.520 --> 00:22:56.397 >> Yeah. You don't know. Maybe this maybe this $200 million LOI is from REI or 00:22:56.480 --> 00:23:00.797 Patagonia. You know, you don't know who makes beanies. Uh what's the Carheart? 00:23:00.880 --> 00:23:02.477 Carheart makes a great beanie. 00:23:02.560 --> 00:23:04.717 >> Go. You don't know any of this stuff. 00:23:04.800 --> 00:23:09.197 You're you're completely out to lunch. I went through the beanie economy. 00:23:09.280 --> 00:23:12.317 >> Beanie economy. Anyway, beanie market map. Work on it. 00:23:12.400 --> 00:23:15.757 >> Let me tell you about public.com. 00:23:15.840 --> 00:23:18.877 Public.com. investing for those who take it seriously, stocks, options, bonds, 00:23:18.960 --> 00:23:22.397 crypto, treasuries, and more. All with great customer service. 00:23:22.480 --> 00:23:23.997 >> They just launched a feature today that 00:23:24.080 --> 00:23:29.917 allows you to connect uh your favorite chat app to public. 00:23:30.000 --> 00:23:32.957 >> Yes. >> And more important than ever, because 00:23:33.040 --> 00:23:34.717 with public, you're going to be able to 00:23:34.800 --> 00:23:38.557 go and create the S&P 499 if you don't 00:23:38.640 --> 00:23:42.877 like SpaceX or the S&P 1 if you love SpaceX. You can express your opinion 00:23:42.960 --> 00:23:45.917 about SpaceX however you want. Can you please help me build an index for one 00:23:46.000 --> 00:23:49.677 company? >> Yes, index for one company or index for 00:23:49.760 --> 00:23:52.237 everything but one company. Uh SpaceX is 00:23:52.320 --> 00:23:56.717 very uh divisive. People are extremely optimistic in certain camps. Extremely 00:23:56.800 --> 00:24:00.797 pessimistic. >> Very optimistic. Uh Goldman expects 00:24:00.880 --> 00:24:05.437 SpaceX's AI revenue to surge a hundred times by 2030. 00:24:05.520 --> 00:24:09.197 >> Huge. >> Uh big big number. I looked at this 00:24:09.280 --> 00:24:13.117 title and I was thinking like, okay, what's Grock's actual revenue today? If 00:24:13.200 --> 00:24:16.717 you take out >> Yeah. X, >> what what is their AI revenue today? Is 00:24:16.800 --> 00:24:18.717 it just Grock subscriptions plus GR 00:24:18.800 --> 00:24:20.557 Grock tokens? Do you include X 00:24:20.640 --> 00:24:26.557 subscriptions? Do you include uh cloud vendor and and neo cloud contracts? 00:24:26.640 --> 00:24:29.197 There's a bunch of different ways to measure it. The smaller the number, the 00:24:29.280 --> 00:24:32.157 easier it is to 100x. But we have seen 00:24:32.240 --> 00:24:34.557 other AI companies 100x revenues over 00:24:34.640 --> 00:24:36.397 two years, over three years, four years. 00:24:36.480 --> 00:24:41.037 Like the 100x has become, it's not a one-of-a-one scenario. 00:24:41.120 --> 00:24:43.037 >> Yeah. >> It's happened multiple times. And so uh 00:24:43.120 --> 00:24:45.197 we have seen these charts many times and 00:24:45.280 --> 00:24:48.637 uh if if they execute well there this is 00:24:48.720 --> 00:24:51.277 entirely possible. It is it is extremely 00:24:51.360 --> 00:24:54.317 >> other other notable data points from the road show. 00:24:54.400 --> 00:24:57.357 >> Yeah. >> They the forecast anticipate SpaceX 00:24:57.440 --> 00:24:59.677 making about 360 billion of capital 00:24:59.760 --> 00:25:03.277 expenditures through 2028. Uh Jensen 00:25:03.360 --> 00:25:06.157 somewhere fist pumping very excited 00:25:06.240 --> 00:25:09.597 about that number be a new hyperscaler. 00:25:09.680 --> 00:25:12.237 Um and uh anyways very 00:25:12.320 --> 00:25:14.717 >> should be unsurprising but very 00:25:14.800 --> 00:25:17.837 >> aggressive. Yeah. And um yeah the 00:25:17.920 --> 00:25:20.637 enterprise story is also live too. I saw 00:25:20.720 --> 00:25:23.117 the new Nvidia foundation model is also 00:25:23.200 --> 00:25:25.277 live. Uh we'll have to go check it out 00:25:25.360 --> 00:25:27.677 and look at the model card soon. See how 00:25:27.760 --> 00:25:32.637 it's benchmarking. But we got to move on to benchmark because there's new news in 00:25:32.720 --> 00:25:36.317 the benchmark world. First I'm going to tell you about MongoDB. What's the only 00:25:36.400 --> 00:25:40.397 thing faster than the AI market? your business on MongoDB. Don't just build 00:25:40.480 --> 00:25:44.397 AI, own the data platform that powers it. So, 00:25:44.480 --> 00:25:47.517 >> moment of silence. >> Moment of silence. Why is that sense? 00:25:47.600 --> 00:25:50.637 >> Uh the the end of an era, the last >> I guess I guess they have been very 00:25:50.720 --> 00:25:55.997 focused for decades. >> The last tier one that was a pure venture capital. 00:25:56.080 --> 00:25:59.117 >> What did they get called again? Internet boys or something. 00:25:59.200 --> 00:26:02.717 >> Oh, >> soft boys. [laughter] There was there some book about them that was very 00:26:02.800 --> 00:26:08.397 funny. But e boys is a hit piece of a book book title. 00:26:08.480 --> 00:26:11.197 >> It's a [laughter] fantastic book, >> but the subtitle makes up makes up for 00:26:11.280 --> 00:26:13.997 it. And it's a fantastic book and it's a very interesting story where they 00:26:14.080 --> 00:26:16.397 actually let a journalist come in and 00:26:16.480 --> 00:26:19.597 see how they true story of the six tall 00:26:19.680 --> 00:26:24.077 men. He clear he clearly wrote the subtitle 00:26:24.160 --> 00:26:25.837 and was like, I got to take the edge off 00:26:25.920 --> 00:26:27.997 of this. It's too glazy. I got to I got 00:26:28.080 --> 00:26:33.517 to take it down a notch. And so he uh uh he he threw the e boys in there. But uh 00:26:33.600 --> 00:26:36.877 >> anyways, [laughter] big moves from Benchmark. Tate Clark has a scoop in the 00:26:36.960 --> 00:26:40.717 journal. Benchmark has raised >> two billion across two new funds. 00:26:40.800 --> 00:26:45.357 >> Wow. >> And uh most notably their first ever dedicated growth fund. 00:26:45.440 --> 00:26:47.437 >> H do they hire anyone who has experience 00:26:47.520 --> 00:26:51.677 growth investing? Who could possibly do growth investing there? Someone who's 00:26:51.760 --> 00:26:55.837 maybe like a bond capital and then founders fund then maybe Kleiner. Like 00:26:55.920 --> 00:26:58.797 someone with that. Somebody with that kind of background I think would be fantastic. 00:26:58.880 --> 00:27:00.957 >> Pretty good for growth investing. 00:27:01.040 --> 00:27:02.557 >> Now that you say that though. 00:27:02.640 --> 00:27:05.597 >> Yeah. >> Ev Randall. >> Ev Randall. That's right. 00:27:05.680 --> 00:27:07.917 >> They did pick up Ed. >> They did pick him up. They're almost 00:27:08.000 --> 00:27:09.917 thinking two steps ahead there. 00:27:10.000 --> 00:27:13.517 >> Are they building their fund strategy now? Their entire platform strategy 00:27:13.600 --> 00:27:17.597 around Ed Randall >> potentially. Potentially. Anyway, uh let 00:27:17.680 --> 00:27:20.797 me tell you about Shopify. Shopify is the commerce platform that grows with 00:27:20.880 --> 00:27:23.997 your business, lets you sell in seconds, online, in store, on mobile, on social, 00:27:24.080 --> 00:27:26.077 on marketplaces, and now with AI agents. 00:27:26.160 --> 00:27:27.757 And we are very fortunate to be joined 00:27:27.840 --> 00:27:30.797 by Alex Karp in just a minute. He's 00:27:30.880 --> 00:27:35.837 coming in to speak with us at AIP Con here. We're going to bring him in in 00:27:35.920 --> 00:27:39.357 just a minute. >> While we wait, Austin based podcaster 00:27:39.440 --> 00:27:43.277 Joe Rogan reportedly being considered for >> 60 Minutes. 00:27:43.360 --> 00:27:48.077 >> 60 Minutes. Uh, looks like >> they're gonna have to call it 200 minutes 00:27:48.160 --> 00:27:51.037 because he records long podcasts in 60 minutes isn't enough for him. 00:27:51.120 --> 00:27:55.117 >> Hundreds. It'll just be called >> Barry, if you're listening, put us in. 00:27:55.200 --> 00:27:57.197 Put us in the ring. We're ready to go. 00:27:57.280 --> 00:28:00.317 You need tech, business correspondent, 00:28:00.400 --> 00:28:04.637 someone who can just chop it up for 60 minutes. We do 60 minutes three times a 00:28:04.720 --> 00:28:08.637 day. We're ready to go. This is going to be light work for us. Barry, 00:28:08.720 --> 00:28:11.597 >> I'm ready. I'm ready. You could do 60 minutes right now. You could do 60 00:28:11.680 --> 00:28:15.357 minutes tomorrow. You could do 60 minutes. You could do an extra 60 minutes easily 00:28:15.440 --> 00:28:18.157 >> when you're putting up. We're putting up a thousand minutes a week. It's no 00:28:18.240 --> 00:28:20.477 problem. >> We did consider that at one point early 00:28:20.560 --> 00:28:23.917 on. We thought, should we do basically a morning show? 00:28:24.000 --> 00:28:26.877 >> Oh, yeah. >> Take a two-hour break and come back and do another show. 00:28:26.960 --> 00:28:30.717 >> Late night show. Yeah, late night show, maybe. Anyway, we have Alex Garp here 00:28:30.800 --> 00:28:33.677 with us. >> Welcome to the show. Welcome back. Thank 00:28:33.760 --> 00:28:37.117 you so much for taking the time. Uh we're going to have you grab these 00:28:37.200 --> 00:28:40.157 headset. These headphones, right? Not these. 00:28:40.240 --> 00:28:45.197 >> You can sit close. Get in here. We got >> We liked it last time. The three of us 00:28:45.280 --> 00:28:48.397 were sitting here. >> No, we can. We can put the giant. 00:28:48.480 --> 00:28:50.317 >> Let's put this up here. Right here. I'll 00:28:50.400 --> 00:28:52.237 sit. You can stand. This always works. 00:28:52.320 --> 00:28:53.837 Makes me feel Yeah. Get in here close. 00:28:53.920 --> 00:28:57.757 This is good. Feel good. Okay. Feel good. Uh, how is it going? How is AIPCON 00:28:57.840 --> 00:29:00.077 this time around? What's changed? 00:29:00.160 --> 00:29:04.477 >> Um, well, we we're in a phase. All each one of these things like marks a time. 00:29:04.560 --> 00:29:08.157 First of all, you guys are even more baller, more successful. 00:29:08.240 --> 00:29:09.837 >> Thank you. >> Some tendies in your pocket. 00:29:09.920 --> 00:29:12.797 >> I think it might have been been part two. We got to say thank you. You know, 00:29:12.880 --> 00:29:16.647 it's like blew us up. You're looking you're looking bigger and stronger. 00:29:16.730 --> 00:29:19.757 [laughter] So, hey, are you more attractive in your personal life now? 00:29:19.840 --> 00:29:23.437 Randomly. >> Well, here's what we're actually focused on. Dead hangs. 00:29:23.520 --> 00:29:26.557 >> So, you came on last time. You said your dead hangs around like five 00:29:26.640 --> 00:29:30.157 >> Oh, no. It's It's Well, it's plateaued in the last couple months at 5:30. 00:29:30.240 --> 00:29:34.717 >> 5:30. Okay. So, we The thing is like people are going to hear that. They're 00:29:34.800 --> 00:29:38.237 going to think hanging on a bar. How hard could it be? You got to go and do 00:29:38.320 --> 00:29:39.837 it. The audience has to go try to do it. 00:29:39.920 --> 00:29:42.717 We've started doing it. We're still in the 00:29:42.800 --> 00:29:47.197 >> We're still between Yeah. between and somewhere around a minute 30. You feel 00:29:47.280 --> 00:29:48.877 like your tendons are going to rip. 00:29:48.960 --> 00:29:52.237 >> 300 dead hang is respectable. 00:29:52.320 --> 00:29:54.237 >> Okay. >> 2 minutes is super elite. 00:29:54.320 --> 00:29:57.597 >> It does feel respectable when you have the five minute number. You're looking 00:29:57.680 --> 00:30:01.117 at the [laughter] >> think strength matters. No. Um the the 00:30:01.200 --> 00:30:04.557 the thing is I don't want to go into rabbit hole in training. The single 00:30:04.640 --> 00:30:08.477 biggest mistake people make is they try to hang every day. You need recovery. 00:30:08.560 --> 00:30:12.797 It's like anything else. So if you want to mimic and get progress, you just do 00:30:12.880 --> 00:30:16.077 what I do, which is once a week you hang as long as you can. Doesn't have to be 00:30:16.160 --> 00:30:18.477 super macho. And then that's your day. 00:30:18.560 --> 00:30:21.117 So like just say you can do 130, >> but multiple sets or 00:30:21.200 --> 00:30:23.677 >> No, no, no. One day a week you do your maximum >> max. 00:30:23.760 --> 00:30:26.237 >> Wow. >> So like let's say you could do two minutes. Okay. 00:30:26.320 --> 00:30:29.437 >> You try to do at least 1:30. You fight to get to 1:30, but you don't fight to 00:30:29.520 --> 00:30:31.997 get to two minutes. Got it. >> That's your dead hang day. And then you 00:30:32.080 --> 00:30:33.597 can basically around the next day. 00:30:33.680 --> 00:30:37.437 Do whatever you want. Don't overdo it, but you could do two times one minute 00:30:37.520 --> 00:30:40.637 with a long break. >> Mhm. And then can't you just go around 00:30:40.720 --> 00:30:44.557 do less and less and less two days before if you two-minute dead hang you 00:30:44.640 --> 00:30:48.877 do like four times 15 seconds the day before you take off 00:30:48.960 --> 00:30:52.877 >> and you do that just keep doing that and you're dead. What the mistake people 00:30:52.960 --> 00:30:54.897 make is they hear my ball or time. 00:30:54.980 --> 00:30:59.277 [laughter] that guy. I mean the mistake you're 00:30:59.360 --> 00:31:02.397 making is not doing a course. This could be a whole new [laughter] revenue line. 00:31:02.480 --> 00:31:05.517 This could be a whole new revenue line of >> pound. You guys have it more for you 00:31:05.600 --> 00:31:09.357 guys. be like, "Hey, call in." So, um, 00:31:09.440 --> 00:31:13.037 yeah, I mean, the dead hang is a it's like and also some of it's just genetic. 00:31:13.120 --> 00:31:17.997 Like my other metrics are elite, but that this is somehow alien territory. 00:31:18.080 --> 00:31:21.517 >> God-given gift. >> What about uh what about breathold underwater? 00:31:21.600 --> 00:31:25.197 >> I don't do that. And I'm not sure I have like I grew up swimming. I think I'm 00:31:25.280 --> 00:31:29.197 weaker at that. Like I bet you I'd be in your guys range. [laughter] 00:31:29.280 --> 00:31:32.237 >> I I'm a dive master. I can hold my breath for three minutes. 00:31:32.320 --> 00:31:35.277 >> And I'm just Yeah. I'm just Yeah. So, I think you would be I think you'd be 00:31:35.360 --> 00:31:38.317 crushing me on that, honestly. But you have like the lung capacity of a whale. 00:31:38.400 --> 00:31:40.637 That's true. I mean, like you got like 00:31:40.720 --> 00:31:45.437 I'm not moving. I'm not using. It's like you're like like you're like a like a 00:31:45.520 --> 00:31:50.237 like a whale floating out [laughter] there under the ocean waiting to surface. True. 00:31:50.320 --> 00:31:54.237 >> So, no, I say I'll tell you the difference. Uh, God, they're always 00:31:54.320 --> 00:31:56.157 minding me out there. But like, okay, 00:31:56.240 --> 00:31:59.037 when we first met, it was like AI may be 00:31:59.120 --> 00:32:02.477 real. Okay. Then I would say somehow 00:32:02.560 --> 00:32:07.117 until about two weeks ago there was like a holy this is real but 00:32:07.200 --> 00:32:09.997 >> somehow it's not working but we're not allowed to say it publicly because we'll 00:32:10.080 --> 00:32:12.957 look stupid. Yeah. >> And then there's a lot of investor hype. 00:32:13.040 --> 00:32:16.717 Yep. >> Um there still is like investors printing attendees. So you have the 00:32:16.800 --> 00:32:20.637 investors on one side. I think people realize it's real but 00:32:20.720 --> 00:32:24.477 >> you know it's like you know you have the whole token maxing. Yep. 00:32:24.560 --> 00:32:26.797 >> And people are on to that and then so 00:32:26.880 --> 00:32:28.957 there's a whole value lecture there. Y 00:32:29.040 --> 00:32:33.197 >> um uh you have a political situation 00:32:33.280 --> 00:32:35.677 um where you know people who do not 00:32:35.760 --> 00:32:40.237 understand basic economics are winning the political argument. So you can you 00:32:40.320 --> 00:32:43.437 could we could talk about where >> there's a lot here. Let's break it let's 00:32:43.520 --> 00:32:46.797 break it up. Let's start with the token maxing thing. Let's start with what's 00:32:46.880 --> 00:32:51.837 what's real. Uh how are you actually thinking about deploying AI? 00:32:51.920 --> 00:32:55.677 >> First of all, what what is your what is what is Palanteer's philosophy around 00:32:55.760 --> 00:32:58.797 token consumption? >> Well, like we Okay. We have a product 00:32:58.880 --> 00:33:03.197 that will allow you to ba I mean internally it's called something but 00:33:03.280 --> 00:33:08.157 externally but really we call it the demastatory like get off masturbation 00:33:08.240 --> 00:33:12.477 thing internally. Sure. It's like people are just like print like sitting there 00:33:12.560 --> 00:33:14.717 all day kind of like a porn addiction 00:33:14.800 --> 00:33:17.357 and enterprises are like okay we knew 00:33:17.440 --> 00:33:21.837 this we believe this will create value but we cannot have people just like some 00:33:21.920 --> 00:33:24.237 people >> checking the weather with it just like 00:33:24.320 --> 00:33:27.837 and just re rearranging deck chairs on their personal Titanic 00:33:27.920 --> 00:33:30.957 >> literally like porn like people are like full on 00:33:31.040 --> 00:33:34.957 >> it feels it feels so >> yeah toolshaped objects tool-shaped 00:33:35.040 --> 00:33:38.477 objects you're looking at more than you want you hope no one notices you're kind 00:33:38.560 --> 00:33:42.397 of It feels productive to have every email classified with 00:33:42.480 --> 00:33:47.037 >> here's what it comes down to like business problems can never be I mean 00:33:47.120 --> 00:33:50.957 sometimes they can be solved purely with money and just spending more but very 00:33:51.040 --> 00:33:53.917 often >> actually I think it's the opposite so just to give you a weird an 00:33:54.000 --> 00:33:56.957 >> no and I was going to say very very often it's more the opposite where it's 00:33:57.040 --> 00:34:00.717 about figuring out the right way to do something and then you can use capital 00:34:00.800 --> 00:34:04.317 to fuel that process but >> let me give you a thing that's too 00:34:04.400 --> 00:34:09.197 generous for you guys >> okay it it it's taste plus money. 00:34:09.280 --> 00:34:13.197 >> Okay. Yeah. >> And there is no like AI like if you look 00:34:13.280 --> 00:34:14.477 at like pick any issue you want to talk 00:34:14.560 --> 00:34:18.157 about token mechanism maxim maxing uh 00:34:18.240 --> 00:34:20.477 what's going on with deploy codes are 00:34:20.560 --> 00:34:22.877 other people going to build ontologies 00:34:22.960 --> 00:34:25.597 why are why why does our political class 00:34:25.680 --> 00:34:30.237 not understand AI especially in Europe it's like yes because all these things 00:34:30.320 --> 00:34:32.317 can be scaled in a very valuable but 00:34:32.400 --> 00:34:35.357 largely going to commodified way but you 00:34:35.440 --> 00:34:39.357 can't scale the taste of like what is the business problem you want to have to 00:34:39.440 --> 00:34:43.277 solve and need to solve at the end of the day whether it's in the for whether 00:34:43.360 --> 00:34:47.277 it's the Ukrainians fighting the Israelis commercial entities it's 00:34:47.360 --> 00:34:50.877 there's somebody sitting there who's like okay but this problem is valuable 00:34:50.960 --> 00:34:53.277 this problem isn't and once that value 00:34:53.360 --> 00:34:58.237 that problem is always has that problem >> always almost always but not always has 00:34:58.320 --> 00:35:01.117 attributes so there are some problems you could solve with this like I want to 00:35:01.200 --> 00:35:05.437 write a report on GDP growth in China right okay but if it's a problem that 00:35:05.520 --> 00:35:09.757 requires a knowledge store like I want to understand the specialized way I 00:35:09.840 --> 00:35:12.637 underwrite. We're going to have a guest here. I want to understand the 00:35:12.720 --> 00:35:17.037 specialized way I drill for oil and gas that's both legal, ethical, and reduces 00:35:17.120 --> 00:35:19.357 the cost of production. I want to change 00:35:19.440 --> 00:35:24.077 the the the supply chain of my industry, whether that's military or whether 00:35:24.160 --> 00:35:25.597 that's building boxes or whether that's 00:35:25.680 --> 00:35:28.477 cars. These things require actual 00:35:28.560 --> 00:35:34.077 precise ongoing processes. They are enhanced by large language models. They 00:35:34.160 --> 00:35:37.037 are not replaced by large language models. And then you get to security 00:35:37.120 --> 00:35:39.277 issues like but for us like the whole 00:35:39.360 --> 00:35:41.117 myth of things is just a boon because 00:35:41.200 --> 00:35:44.077 like yeah we could take any model their 00:35:44.160 --> 00:35:46.397 model open AI model sure 00:35:46.480 --> 00:35:49.277 >> open model we can identify we can now 00:35:49.360 --> 00:35:51.597 identify uh vulnerabilities at like 10 00:35:51.680 --> 00:35:54.957 100x yeah but then who patches them 00:35:55.040 --> 00:35:58.877 >> how do you patch them on prem how do you patch them on prem so that your 00:35:58.960 --> 00:36:02.957 specialized knowledge stays on prem like if you're any business or Intel service 00:36:03.040 --> 00:36:06.717 a lot of these things are very similar like you're not putting your classified 00:36:06.800 --> 00:36:11.117 data in a public cloud. Same thing if you're like you have a special way of 00:36:11.200 --> 00:36:14.317 farming soybeans. >> Yeah. >> Well, you're not. So, it's like how do 00:36:14.400 --> 00:36:19.837 you have how do you so all these problems are exposed identified and then 00:36:19.920 --> 00:36:23.197 >> you always have a thing of where's the charisma which people really 00:36:23.280 --> 00:36:25.437 underestimate and it's it's not global. 00:36:25.520 --> 00:36:29.197 There's no global charisma now. Like so right now the large language models are 00:36:29.280 --> 00:36:33.517 very frontier companies are super charismatic with investors. Mhm. 00:36:33.600 --> 00:36:38.397 >> I'll give you some news. They're >> super not charismatic with enterprises 00:36:38.480 --> 00:36:41.277 and the people like in a way. >> Even with enterprise. 00:36:41.360 --> 00:36:43.197 >> No, no. I mean, it's >> because I understand what the people No, 00:36:43.280 --> 00:36:46.477 no. The enterprise people. I have a secret. I have a secret like every 00:36:46.560 --> 00:36:49.917 company has a secret way of selling. You know what my secret way of selling is? 00:36:50.000 --> 00:36:54.317 Don't even call. Don't Don't come talk to us. There's a frontier company. Go 00:36:54.400 --> 00:36:57.917 spend two days with them. And if you're lucky, after you're done, I'll let you 00:36:58.000 --> 00:36:59.917 in my door. They're like clamoring. 00:37:00.000 --> 00:37:03.357 They're like They're like, "Hey, I'll take your bad brand." We have a great 00:37:03.440 --> 00:37:06.797 brand in enterprise, but like it it's like it's like secret knowledge because 00:37:06.880 --> 00:37:09.517 the investors love this. They're like, "Hey, my stocks are all up. Everything's 00:37:09.600 --> 00:37:11.037 up. I mean, Palunteer's done very well. 00:37:11.120 --> 00:37:14.557 We're like, but it's like and you know, you guys are doing very well, I 00:37:14.640 --> 00:37:17.917 imagine." Right? And it's okay. We're we're you know, we can But I'll tell you 00:37:18.000 --> 00:37:21.997 what, you go down the street, you talk to a marine, you talk to a bus driver, 00:37:22.080 --> 00:37:25.997 you talk to the person who owns the bus driving company, 00:37:26.080 --> 00:37:29.837 >> they are not happy. They do not like these people. They're tired of people 00:37:29.920 --> 00:37:32.477 token maxing. Yeah. That it looks like 00:37:32.560 --> 00:37:37.677 masturbation at their that's cost them money. They they're like And honestly, 00:37:37.760 --> 00:37:39.837 then you have something we're not allowed to talk about in this country. 00:37:39.920 --> 00:37:43.037 Likability. >> Like a palunteer. We have I think we 00:37:43.120 --> 00:37:46.637 have like 50 100 million global fans. We have like 5 million people that wake up 00:37:46.720 --> 00:37:49.517 in the morning literally calling me Satan. 00:37:49.600 --> 00:37:53.997 >> I didn't know I had that kind of >> warm hand. But uh you know, it's like 00:37:54.080 --> 00:37:55.517 that's what they believe. >> Yeah. 00:37:55.600 --> 00:37:57.437 >> And like and they really believe it. 00:37:57.520 --> 00:38:02.157 Okay. What people are not allowed to really address is like we have fans and 00:38:02.240 --> 00:38:04.317 enemies. >> Yeah. >> Yeah. These people polarizing. 00:38:04.400 --> 00:38:07.597 >> Yeah. We're polarizing which means both sides. Yep. These people have one side. 00:38:07.680 --> 00:38:12.477 >> Yep. >> They're just It is. So it's like you and it's like it's a really big 00:38:12.560 --> 00:38:14.157 >> social media companies too have the same problem. 00:38:14.240 --> 00:38:16.237 >> Yeah. >> Everyone uses them but no one likes them. 00:38:16.320 --> 00:38:19.757 >> Yeah. But but then they also live in a circle and that circle's printing money. 00:38:19.840 --> 00:38:21.677 Yep. >> So it's like you know when you look in 00:38:21.760 --> 00:38:23.917 the mirror and you just printed a lot of money, 00:38:24.000 --> 00:38:26.237 >> you look pretty fresh. Is part of it is 00:38:26.320 --> 00:38:28.637 is part of it that that some element of 00:38:28.720 --> 00:38:32.797 the technology let's just say LLMs is so 00:38:32.880 --> 00:38:37.757 magical that the companies involved that the companies that are making and 00:38:37.840 --> 00:38:41.037 selling frontier intelligence can be bad at a bunch of other things and still 00:38:41.120 --> 00:38:44.877 grow >> well no no they are magical at a certain 00:38:44.960 --> 00:38:49.357 kind of thing allowing you to write for example code now that code doesn't can't 00:38:49.440 --> 00:38:52.637 be used as a knowledge store so if you look at code in like three different 00:38:52.720 --> 00:38:55.437 ways like just using palent as a model we have code That's basically 00:38:55.520 --> 00:38:58.077 infrastructure. So what what are the Ukrainians using? What is the department 00:38:58.160 --> 00:39:01.917 of war using? What do a lot of our enterprises? We call that primitives. 00:39:02.000 --> 00:39:05.597 It's basically hard-coded things that that understand the world low way do you 00:39:05.680 --> 00:39:10.157 do? It would take millions of technical hours and an understanding of all these 00:39:10.240 --> 00:39:13.997 enterprises to do it. So it's it's much more like how do you build a steel beam? 00:39:14.080 --> 00:39:18.317 Then you have like code that is written by FDAs. 00:39:18.400 --> 00:39:22.877 >> Okay. So that's kind of managed. It's the reason why FDAs work. The secret is 00:39:22.960 --> 00:39:25.997 it's actually managed on something that we as a product. So you're writing to a 00:39:26.080 --> 00:39:28.717 codebase. We're managing that. We're increasing our product. It's not just 00:39:28.800 --> 00:39:32.397 random people write. Then you have let's call it free code. 00:39:32.480 --> 00:39:36.797 >> That free code is that's magical. Like you can do it very quickly. It's almost 00:39:36.880 --> 00:39:38.317 right. It doesn't have to be exact. 00:39:38.400 --> 00:39:41.517 >> Dashboards, >> dashboards, financial stuff, 00:39:41.600 --> 00:39:44.637 >> probabilistic stuff where you just have to get one off analysis. 00:39:44.720 --> 00:39:47.917 >> It was magical. >> By the way, it's magical. It not only 00:39:48.000 --> 00:39:51.837 creates and it's magical in a way. I know people don't like the porn thing, 00:39:51.920 --> 00:39:55.037 but it's also addicting. It's like, you know, it's not good for you, but you 00:39:55.120 --> 00:39:57.197 know, it may lead to damage. >> One more dashboard, 00:39:57.280 --> 00:39:59.597 >> one more time. It can't hurt that much. 00:39:59.680 --> 00:40:03.677 I know my doctor says it I shouldn't do it, but it's like it's like that, right? 00:40:03.760 --> 00:40:07.277 And you just keep go and like and if you're involved in the that thing, 00:40:07.360 --> 00:40:09.117 you're also making money. >> Yeah. 00:40:09.200 --> 00:40:10.957 >> And then last not least in certain 00:40:11.040 --> 00:40:13.597 circles like if you have you want to be 00:40:13.680 --> 00:40:18.557 a Russ researcher or you believe essentially it's a religion. So like you 00:40:18.640 --> 00:40:21.517 know and like one of the things is very charismatic especially to people who've 00:40:21.600 --> 00:40:24.637 never had a religion because all of a sudden that hole in your heart that was 00:40:24.720 --> 00:40:26.877 yearning for I don't know I would say 00:40:26.960 --> 00:40:32.957 you know a a established religion Judaism, Christianity, Islam is like being filled. 00:40:33.040 --> 00:40:35.517 >> Yeah. >> And and all the answers are there but it 00:40:35.600 --> 00:40:38.397 it's very very successful at doing 00:40:38.480 --> 00:40:40.717 things that a company has to do. 00:40:40.800 --> 00:40:44.237 >> But it is not actually solving the 00:40:44.320 --> 00:40:47.837 problem that enterprises are. It is now it can solve them. That's the trick. 00:40:47.920 --> 00:40:51.197 It's not It's not binary. It's not like you can't say they're not valuing. 00:40:51.280 --> 00:40:53.037 They're totally putting our business on 00:40:53.120 --> 00:40:56.557 steroids. Like without LM, nobody would 00:40:56.640 --> 00:40:58.957 be talking about our ontology, about 00:40:59.040 --> 00:41:03.677 Apollo managing SEC exploits, about our ability to manage an enterprise 00:41:03.760 --> 00:41:07.917 essentially as turning all these companies into FDES. These deploy codes, 00:41:08.000 --> 00:41:10.557 we love them because now every company wants a deploy code. You know how you do 00:41:10.640 --> 00:41:13.597 that? >> You replplatform on Palunteer and like 00:41:13.680 --> 00:41:16.317 and it actually works. It's not somebody with no taste, who's never done 00:41:16.400 --> 00:41:19.357 enterprise, who has no earthly clue how these things work, who's done something 00:41:19.440 --> 00:41:22.237 else, and is like just imagining they know how to do it, right? 00:41:22.320 --> 00:41:25.117 >> Yeah. Is part is part of this moment 00:41:25.200 --> 00:41:28.157 quite entertaining for you because 00:41:28.240 --> 00:41:30.477 you guys have been working on 00:41:30.560 --> 00:41:32.717 understanding businesses at a deep 00:41:32.800 --> 00:41:35.917 fundamental level creating 00:41:36.000 --> 00:41:37.997 you guys have effectively been doing the 00:41:38.080 --> 00:41:40.637 work that it that people are promising 00:41:40.720 --> 00:41:43.997 AI can do for 20 years now but actually 00:41:44.080 --> 00:41:46.717 doing it finding all the really rough 00:41:46.800 --> 00:41:52.317 finding all the really rough edges and uh and not and and being at a point 00:41:52.400 --> 00:41:55.277 where you don't have to oversell the technology. You can sell both things. 00:41:55.360 --> 00:42:00.397 But now there's maybe >> Here we go. We got it together. Um, now 00:42:00.480 --> 00:42:03.677 there's maybe >> Oh, it's the wrong side. That's why. 00:42:03.760 --> 00:42:07.837 Flip it around. >> The dyslexic. 00:42:07.920 --> 00:42:10.317 >> There you go. There you go. Living the brand. Living the brand. 00:42:10.400 --> 00:42:12.637 >> We got that. I don't know. All that stuff wasn't recording. 00:42:12.720 --> 00:42:17.117 >> I I trailed off. But uh but is part of it entertaining to you that it feels 00:42:17.200 --> 00:42:18.957 like 00:42:19.040 --> 00:42:21.677 >> you know Palunteer has always been in 00:42:21.760 --> 00:42:24.637 some ways um not had competitors because 00:42:24.720 --> 00:42:27.437 there's nobody with Alex Karp running a 00:42:27.520 --> 00:42:29.517 company that is does what Palanteer does 00:42:29.600 --> 00:42:31.837 besides Palanteer. But at the same time, 00:42:31.920 --> 00:42:34.557 there's been tens of billions of dollars 00:42:34.640 --> 00:42:36.957 deployed now to effectively do what 00:42:37.040 --> 00:42:40.077 Palanteer does, but just selling the 00:42:40.160 --> 00:42:44.717 intelligence part, not selling all the underlying kind of infrastructure that you 00:42:44.800 --> 00:42:47.357 >> Well, they're doing two things. They're selling they're trying to sell the 00:42:47.440 --> 00:42:51.277 intelligence part and they're trying to pretend if you just hire a bunch of 00:42:51.360 --> 00:42:52.397 people and let them run around their 00:42:52.480 --> 00:42:55.517 FDS. Now, the the very cool thing is 00:42:55.600 --> 00:42:58.717 when you've been in your basement doing 00:42:58.800 --> 00:43:03.277 your thing and everyone kind of views it as the freak show, it it's really 00:43:03.360 --> 00:43:05.277 interesting and and and great to have 00:43:05.360 --> 00:43:08.637 adoption. The pretty ironic thing is 00:43:08.720 --> 00:43:10.637 half the people adopting now don't even 00:43:10.720 --> 00:43:12.797 know they're copying. But now the 00:43:12.880 --> 00:43:16.477 copying thing helps and hurts. 00:43:16.560 --> 00:43:19.757 Where it hurts is in the beginning it puts clutter in the market. 00:43:19.840 --> 00:43:21.757 >> Yeah. >> Um and there's there's no doubt about 00:43:21.840 --> 00:43:26.637 it. where it helps and then we saw this with defense tech honestly. So like in 00:43:26.720 --> 00:43:28.397 defense tech, we were the only people. 00:43:28.480 --> 00:43:32.797 We were the first people despite what it's I I love these honestly other 00:43:32.880 --> 00:43:36.157 podcasters. They're interviewing people who are paring things I said 20 years 00:43:36.240 --> 00:43:39.437 ago. They don't know it and it's like oh that's so insightful. It's like yeah of 00:43:39.520 --> 00:43:43.037 course it's insightful. Karp said it 25 years ago. Uh [laughter] and like but 00:43:43.120 --> 00:43:45.197 it's but so that kind of that part is 00:43:45.280 --> 00:43:48.957 super weird but uh and like but um but 00:43:49.040 --> 00:43:53.517 it's but what really happens when we see is like it expands the market. So like 00:43:53.600 --> 00:43:57.837 in defense tech we would not be doing this well in just purely in government 00:43:57.920 --> 00:44:02.557 unless there weren't 50 companies that were doing similar things because then 00:44:02.640 --> 00:44:06.477 the people are like okay first of all >> you view it as like offbalance sheet 00:44:06.560 --> 00:44:09.117 sales resources where other people are basically 00:44:09.200 --> 00:44:14.157 >> well it's off well no that that's the large it's they do two things they 00:44:14.240 --> 00:44:17.837 increase the size of the market because de facto nobody wants to find it 00:44:17.920 --> 00:44:20.237 underwrite a market where there's only one person >> sure 00:44:20.320 --> 00:44:24.157 >> so like if you're the one person the percentage of the defense budget you can 00:44:24.240 --> 00:44:26.797 get is much smaller and two they set up 00:44:26.880 --> 00:44:31.997 a comparator. It's like you know you may not like the freak show. Okay, if like 00:44:32.080 --> 00:44:36.077 but have you noticed the people who are serious buy it and then then it changed 00:44:36.160 --> 00:44:37.757 and then three it changes the standard. 00:44:37.840 --> 00:44:41.677 Now what you're seeing now is like that times 100x. 00:44:41.760 --> 00:44:43.917 >> Yeah. And it does change like recruiting 00:44:44.000 --> 00:44:47.997 retention and like how you build a company and we're we're always think you 00:44:48.080 --> 00:44:50.397 have to think about how to being dyslexic huge advantage there because 00:44:50.480 --> 00:44:53.517 like you don't have a playbook and now that need you need things to shift and 00:44:53.600 --> 00:44:55.757 we're doing that. Um [clears throat] the 00:44:55.840 --> 00:44:58.477 the the central thing though that is 00:44:58.560 --> 00:45:03.277 just cannot be developed even if you understood the playbook. A lot of these 00:45:03.360 --> 00:45:07.197 things are like appear like it's like you know LM code appears like pounder 00:45:07.280 --> 00:45:11.117 code but isn't for a deployed thing appears like pounder it isn't ontology. 00:45:11.200 --> 00:45:15.117 You could theoretically copy parts of it, but they're essentially structures 00:45:15.200 --> 00:45:19.357 that are built deep into organizations that we own. And by the way, take you 00:45:19.440 --> 00:45:23.277 three years. And in three years, we're in a completely different world. But 00:45:23.360 --> 00:45:27.757 there is this magical thing called taste. Like in the end of the day, the 00:45:27.840 --> 00:45:31.277 reason why you guys have done so well, it's of course there's aptitude and 00:45:31.360 --> 00:45:35.517 diligence and showing up and all those things. Yeah. But you have to be able to 00:45:35.600 --> 00:45:39.597 differentiate between two people who are in business, one of whom is saying 00:45:39.680 --> 00:45:43.517 something that sounds weird, that is insightful, one of whom is pariting 00:45:43.600 --> 00:45:47.357 something that sounds weird and that's all they're doing. And a lot of people, 00:45:47.440 --> 00:45:49.677 very few people can do that. And you 00:45:49.760 --> 00:45:51.917 have the same thing like the enterprises 00:45:52.000 --> 00:45:55.117 that succeed, there is a taste arbiter. 00:45:55.200 --> 00:45:59.757 And at Palunteer, we have taste art. We have taste in every product, taste in 00:45:59.840 --> 00:46:03.677 every deployment, taste in every casting. Who puts the people there? How 00:46:03.760 --> 00:46:06.717 do you put them there? How do you organize the thing? Our ontology then 00:46:06.800 --> 00:46:10.397 does that technically. How do you manage the whole org with taste? Who should be 00:46:10.480 --> 00:46:12.397 in charge? What data set should come in? 00:46:12.480 --> 00:46:15.677 What what are what are the ways in which you protect? What is what should you 00:46:15.760 --> 00:46:19.677 push into the public crowd? What should be on prem? What what should I mean 00:46:19.760 --> 00:46:24.477 leaving aside the law and like wars or ethics? What do you want to protect? 00:46:24.560 --> 00:46:27.917 What should you protect? What should you not protect? Because quite frankly, you 00:46:28.000 --> 00:46:31.597 want that to be out there so you can get more data. All those things are arbited 00:46:31.680 --> 00:46:36.237 by taste. And then you have to have the credibility of having taste. That's a 00:46:36.320 --> 00:46:39.197 real problem for a lot of these places because they don't have they they're 00:46:39.280 --> 00:46:41.837 popular with their friends. They don't they really don't understand how 00:46:41.920 --> 00:46:43.347 unpopular they are in enterprise. 00:46:43.430 --> 00:46:46.077 [laughter] Like they think it's like oh yeah like the way I think I have a 00:46:46.160 --> 00:46:50.077 problem with like professors at Colombia. It's like no it's a real 00:46:50.160 --> 00:46:52.637 problem. Like they think I'm Satan and >> [laughter] 00:46:52.720 --> 00:46:55.997 >> uh you know it's like I I I think you know we grew up in the same community. 00:46:56.080 --> 00:46:59.117 Let's talk about Haidiger. They're like they don't want to talk about Haidiger. 00:46:59.200 --> 00:47:03.117 >> So it's like it's like Yeah. And so that's just a it's a weird thing. It's 00:47:03.200 --> 00:47:05.117 going to be a super The one thing I 00:47:05.200 --> 00:47:08.317 would say for anyone listening, if 00:47:08.400 --> 00:47:12.317 you're listening to this and you're chillaxing and not active, I'm not 00:47:12.400 --> 00:47:14.237 saying you have to agree with me politically or anything. 00:47:14.320 --> 00:47:17.037 >> Yeah. >> There the like partly because of this 00:47:17.120 --> 00:47:21.997 dynamic and very self-inflicted because I I I tell you I can't name names. I 00:47:22.080 --> 00:47:26.157 called many of the titans of this world and and 00:47:26.240 --> 00:47:29.037 like started this six months ago like every couple days 00:47:29.120 --> 00:47:31.517 >> were going to be national every couple days. 00:47:31.600 --> 00:47:33.197 >> Like some of them I'm like, "Yeah, we're 00:47:33.280 --> 00:47:35.517 going to be I mean, you know, it's like honestly they're like the bat." They 00:47:35.600 --> 00:47:38.957 they find me very entertaining. Like I'm not sure. Like so they call cuz Yeah. 00:47:39.040 --> 00:47:41.277 It's like it's like, "Oh yeah, this going to be entertaining. You're going 00:47:41.360 --> 00:47:44.077 to pick up." >> Yeah. So any case, so I've been telling 00:47:44.160 --> 00:47:47.517 them for 6 months, six. We're going to be nationalized. 00:47:47.600 --> 00:47:50.477 >> Yeah. You talked about >> we're going to be nationalized. And 00:47:50.560 --> 00:47:53.597 they're like, "Why would anyone nationalize? Never happened in America. 00:47:53.680 --> 00:47:56.957 It's never Why would anyone nationalize us? We're so likable. [laughter] We're 00:47:57.040 --> 00:48:00.557 creating so much value." like okay I'm not going to debate that I know how 00:48:00.640 --> 00:48:02.397 likable I am I'm not going to tell you 00:48:02.480 --> 00:48:04.877 how likable you are but I am telling you 00:48:04.960 --> 00:48:09.437 and you know the momentum on this is on the side of people in national we don't 00:48:09.520 --> 00:48:12.877 get our act together and figure out ways we can say hey look there are problems 00:48:12.960 --> 00:48:15.997 here we're going to deal with these things are not going to yes they are 00:48:16.080 --> 00:48:19.357 going to create opportunities you have to talk openly about how these things 00:48:19.440 --> 00:48:23.037 are valuable because we have adversaries you can't just say these all that stuff 00:48:23.120 --> 00:48:26.637 so the primary risk honestly to Palenteer and a lot of these other 00:48:26.720 --> 00:48:30.877 countries is and And it's going to be nationalized before national it's going 00:48:30.960 --> 00:48:33.997 to be regulated by people who don't understand this. And now they'll tell 00:48:34.080 --> 00:48:37.997 you in private I'm working on this. I'm d this lobbyist. It's like not going to 00:48:38.080 --> 00:48:41.837 work. So like that's something like if you're listening to this and you're like 00:48:41.920 --> 00:48:45.677 look you know you don't have to agree with me on all my proclamations. I got a 00:48:45.760 --> 00:48:48.557 lot of by the way there's some people who think I'm saying we should have a 00:48:48.640 --> 00:48:51.997 draft. Too lazy to read. I'm just saying we should like in a world where 00:48:52.080 --> 00:48:56.637 everything is changing. Everything is changing. Don't we have to find some 00:48:56.720 --> 00:48:59.837 communal structure to remember we're American? You don't like my you don't 00:48:59.920 --> 00:49:04.077 like my idea of like we all do a week in the park? Great. Come up with some other 00:49:04.160 --> 00:49:08.237 idea. We can have no idea, you know, and like and then they're like, "Well, I'm 00:49:08.320 --> 00:49:11.677 saying I I do not want a draft, just to be explicit." They're like, "Oh, that's 00:49:11.760 --> 00:49:13.117 pro-war." No, honestly, you know what? 00:49:13.200 --> 00:49:16.237 Most of our wars are fought because no working-class person is making a 00:49:16.320 --> 00:49:19.677 decision. You start making sure everyone is involved in everything. I'll see you 00:49:19.760 --> 00:49:23.437 how few wars we fight. It's actually the anti-war position. But any case, 00:49:23.520 --> 00:49:25.677 disagree with everything. If you're we 00:49:25.760 --> 00:49:30.877 have on the right and on the left people uh people who have no earthly clue what 00:49:30.960 --> 00:49:32.477 they're talking about right and left. 00:49:32.560 --> 00:49:35.837 >> All they're talking about is how much they hate us and those of us who are 00:49:35.920 --> 00:49:39.437 sensible in the middle. >> You know too many of us are chill waxing 00:49:39.520 --> 00:49:44.237 like it oh like nationalization it can't happen. America would never do that. 00:49:44.320 --> 00:49:47.997 >> Sleepwalking into and you guys have tendies to protect now. You guys should 00:49:48.080 --> 00:49:51.917 be on the front line of this. Like you got full Oh, I'm sorry. Hey, I have a I 00:49:52.000 --> 00:49:56.877 have a full on very impressive uh corporate leader coming on. So, I got I 00:49:56.960 --> 00:50:00.557 got to turn it down. >> Last question. Last question if we have time. Uh 00:50:00.640 --> 00:50:03.357 >> how are your conversations going with 00:50:03.440 --> 00:50:06.637 Fortune 500 CEOs around headcount 00:50:06.720 --> 00:50:10.397 planning? There's been so many layoffs this last year that people were saying, 00:50:10.480 --> 00:50:12.077 hey, we're getting so much out of AI. 00:50:12.160 --> 00:50:13.997 We're able to, you know, cut back here 00:50:14.080 --> 00:50:18.157 or there. uh people inside tech often 00:50:18.240 --> 00:50:21.037 know like these maybe there's just a 00:50:21.120 --> 00:50:25.757 reduction because there needs to be a reduction or got bloated maybe they do 00:50:25.840 --> 00:50:30.637 need to fund some AI >> declining business model yeah the business just doesn't have 00:50:30.720 --> 00:50:33.677 momentum but how are those conversations going what does it look like 00:50:33.760 --> 00:50:35.597 >> like the the like I by the way I talk to 00:50:35.680 --> 00:50:37.837 fortune 500 companies I talk to unions I 00:50:37.920 --> 00:50:40.477 talk to soldiers I talk to fire it 00:50:40.560 --> 00:50:45.757 >> if you upscale somebody >> they're more valuable And like all these 00:50:45.840 --> 00:50:49.277 whether it's people working on batteries, people driving trucks, people 00:50:49.360 --> 00:50:53.117 corporate leaders and again this is where I think we have to be very careful 00:50:53.200 --> 00:50:57.037 to be more disciplined on the corporate side. Like if you run around saying AI 00:50:57.120 --> 00:51:00.797 allowed you to fire twothirds of your workforce and you did it because maybe 00:51:00.880 --> 00:51:02.477 your competitor's kicking your ass. 00:51:02.560 --> 00:51:05.197 Yeah, >> that could that is a really like you 00:51:05.280 --> 00:51:09.357 might as well just go sign up for Bernie Bernie Sanders manifest. And part of the 00:51:09.440 --> 00:51:13.037 thing is they really believe that can't happen. So they're free riding on the 00:51:13.120 --> 00:51:16.237 fact that it could >> like we have and it just cannot work 00:51:16.320 --> 00:51:20.157 anymore. These things are very very explosive. The American people sense 00:51:20.240 --> 00:51:22.077 that there is something dangerous here. 00:51:22.160 --> 00:51:25.917 And when people are playing with that fire, it's like it's a they assume the 00:51:26.000 --> 00:51:29.117 fire won't burn their hands. That's not the world we're in. That fire is going 00:51:29.200 --> 00:51:33.917 to consume us. And what we see, again, the war fighting example is just the 00:51:34.000 --> 00:51:38.477 most neutral, not for everybody, but like the soldiers at the bottom have 00:51:38.560 --> 00:51:41.357 gotten much more valuable. And and I don't even just mean the special 00:51:41.440 --> 00:51:45.197 operators, which obviously they're in a different league, but like every the 00:51:45.280 --> 00:51:48.237 people doing a lot of the operations now are doing our product. They're high 00:51:48.320 --> 00:51:52.797 school vocationally trained. You see this everywhere. The the the modern 00:51:52.880 --> 00:51:54.877 enterprise is going to have like we have 00:51:54.960 --> 00:51:59.677 a a true like very very very smart uh 00:51:59.760 --> 00:52:03.437 person coming on and it's like you're going to have a very smart executive. 00:52:03.520 --> 00:52:05.597 He's much better at hiding it than I 00:52:05.680 --> 00:52:07.197 would be if I were him. But that's you 00:52:07.280 --> 00:52:11.117 can talk to him about that. But um uh um 00:52:11.200 --> 00:52:15.197 and uh and then very talented, creative 00:52:15.280 --> 00:52:19.117 people with taste all up and down the stack. Any case, I think this is time 00:52:19.200 --> 00:52:20.477 for me to >> I think this is time. 00:52:20.560 --> 00:52:22.157 >> We uh >> Thank you so much. 00:52:22.240 --> 00:52:24.397 >> Great to catch up. [clears throat] >> Always fun first. 00:52:24.480 --> 00:52:26.957 >> Oh, they want me to stay for two minutes or what? 00:52:27.040 --> 00:52:30.477 >> Oh, I'm only going to stay. Look, but just a minute. He's got to be the star. 00:52:30.560 --> 00:52:33.357 >> The other headset. >> Put it Put him Put you go in here. 00:52:33.440 --> 00:52:36.077 >> Yeah. And I'm just going to I'm going to take off after a minute. We're going to 00:52:36.160 --> 00:52:39.917 put here here and why don't you here put that headset on 00:52:40.000 --> 00:52:43.277 >> car. Why don't you introduce our guest >> microphone on the left? 00:52:43.360 --> 00:52:46.557 >> Well, I you know let he's he's 00:52:46.640 --> 00:52:49.757 >> one of the smarter people in in business 00:52:49.840 --> 00:52:52.717 um has developed 00:52:52.800 --> 00:52:55.517 um unique ways to underwrite that did 00:52:55.600 --> 00:53:00.557 not involve firing people >> and someone and someone I admire. 00:53:00.640 --> 00:53:03.437 >> Thanks, Alex. >> With that, I'm going to let you guys go. 00:53:03.520 --> 00:53:07.597 Make sure to tell them that the anttology powers are [laughter] >> it's everything. 00:53:07.680 --> 00:53:09.277 >> Always selling. Hey, >> fantastic. 00:53:09.360 --> 00:53:11.197 >> Thanks for coming on the show. It's great to meet you. Pleasure. 00:53:11.280 --> 00:53:15.517 >> Uh yeah, [clears throat] please uh kick us off with like a bit of a more formal 00:53:15.600 --> 00:53:19.357 introduction. >> Yes. So, I'm Peter Zapino. Um I'm the 00:53:19.440 --> 00:53:23.757 executive chairman as effective on Monday of of AIG. I used to be the 00:53:23.840 --> 00:53:27.917 >> chairman and CEO. Um and have you know worked with the company for nine years 00:53:28.000 --> 00:53:31.757 to help transform it. It was in a place where underwriting profitability was 00:53:31.840 --> 00:53:35.917 challenging, operations were challenging, data was challenging. Um, 00:53:36.000 --> 00:53:40.157 capital was challenging. Uh, so you know had a great team of people 00:53:40.240 --> 00:53:41.757 >> with me to transform the company. 00:53:41.840 --> 00:53:45.117 >> So give us the give us the shape of the business in terms of the different 00:53:45.200 --> 00:53:49.517 business lines, the different products, the international footprint, the 00:53:49.600 --> 00:53:52.877 workforce like what give us the scope and the scale here. 00:53:52.960 --> 00:53:55.917 >> Global company with a little bit of a unique footprint. We're 50% 00:53:56.000 --> 00:54:01.357 international, 50% North America, but our second largest country after US is Japan. 00:54:01.440 --> 00:54:03.437 >> Oh. >> Um, we have a big business in India. 00:54:03.520 --> 00:54:06.397 >> Okay. >> Um, and then we have a very big business 00:54:06.480 --> 00:54:10.157 in in the UK. Uh, we do complicated risks. 00:54:10.240 --> 00:54:13.677 >> So, you could think about what's happening uh in the Middle East now with 00:54:13.760 --> 00:54:16.957 shipping, marine, energy. We're heavily involved in that. 00:54:17.040 --> 00:54:20.317 >> So, something where there's not an existing futures contract that a company 00:54:20.400 --> 00:54:24.557 can just go and hedge. It's not, oh, I'm going to buy some oil futures because I 00:54:24.640 --> 00:54:28.397 I I fly planes around and I know I'm going to need diesel fuel in a couple 00:54:28.480 --> 00:54:31.917 months and so I'm going to hedge that out. This is for more complex risks. 00:54:32.000 --> 00:54:35.437 >> It's for more complex risks and you know you think about the largest you know 00:54:35.520 --> 00:54:37.677 sort of customers uh in the world and 00:54:37.760 --> 00:54:42.477 you know big oil companies um you know Fortune 500 companies but we also have a 00:54:42.560 --> 00:54:46.557 >> personal insurance business which will cover things like accident health 00:54:46.640 --> 00:54:48.957 >> uh that are distribution to consumers. 00:54:49.040 --> 00:54:52.557 Uh so we have a real balance. Part of that feels like if you're talking about 00:54:52.640 --> 00:54:57.037 insuring a Fortune 500 company against a geopolitical risk, uh that feels like a 00:54:57.120 --> 00:54:58.557 meeting that takes place in a boardroom. 00:54:58.640 --> 00:55:02.877 It feels like there's a lot of folks with a lot of trust built up over years 00:55:02.960 --> 00:55:05.357 to understand each other's businesses. 00:55:05.440 --> 00:55:09.037 Uh but then there's probably a lot of other underwriting happening and teams 00:55:09.120 --> 00:55:13.197 uh putting together comps and spreadsheets and data. And I want to 00:55:13.280 --> 00:55:15.197 know about the the intersection there. 00:55:15.280 --> 00:55:19.277 It feels like the business is and I don't know if it ever will be just 00:55:19.360 --> 00:55:21.997 oneclick checkout for for insurance 00:55:22.080 --> 00:55:26.557 products for Fortune 500 companies. Uh but what what is the interface between 00:55:26.640 --> 00:55:30.237 the quantitative the qualitative the relationship and the data and then how 00:55:30.320 --> 00:55:32.717 is that changing? >> So the quantitative you have to start at 00:55:32.800 --> 00:55:36.077 the portfolio level. Okay. >> Um and you want as much data as you 00:55:36.160 --> 00:55:40.077 possibly can to look at deterministic y modeling probabilistic and then 00:55:40.160 --> 00:55:44.317 stochastic. I think once you understand like your mean and you understand the 00:55:44.400 --> 00:55:47.277 standard deviation around that then you have to apply it to you know sort of the 00:55:47.360 --> 00:55:51.597 widgets which is each policy >> throughout you know the globe as well as 00:55:51.680 --> 00:55:55.117 um ways in which you structure yeah insurance. So for us 00:55:55.200 --> 00:56:00.237 >> you can't look at you can't look at an individual policy uh in in isolation you 00:56:00.320 --> 00:56:04.557 you're you're managing portfolio risk risk to the entire firm and and that's 00:56:04.640 --> 00:56:07.357 something that's happening probably 247 I imagine 00:56:07.440 --> 00:56:09.517 >> it's hard and that's what led me to Alex 00:56:09.600 --> 00:56:13.437 Karp. Um, you know, it's hard to get the aggregation done in anything that looks 00:56:13.520 --> 00:56:17.917 like real time. It's usually static. It can be 30, 60, 90 days. 00:56:18.000 --> 00:56:19.277 >> And your portfolio could change. I mean, 00:56:19.360 --> 00:56:23.117 it's not going to change dramatically, >> but having the ability to, you know, 00:56:23.200 --> 00:56:26.317 sort of assess risk and use the quantitative data to make better 00:56:26.400 --> 00:56:29.677 decisions on a daily basis is the aspiration of the way the company's going. 00:56:29.760 --> 00:56:32.397 >> Yeah. >> Take us back to your first meeting with 00:56:32.480 --> 00:56:35.917 Karp. Curious uh what the experience was like. 00:56:36.000 --> 00:56:37.757 >> It's a unique individual. >> Call you. 00:56:37.840 --> 00:56:40.237 >> Yeah. No. Um I was actually introduced 00:56:40.320 --> 00:56:43.517 by a board member many years ago and it 00:56:43.600 --> 00:56:46.557 was really in this pursuit of um not 00:56:46.640 --> 00:56:51.757 necessarily foundry or AIP or ontology that's where it led us but it was more 00:56:51.840 --> 00:56:55.837 on sort of the quantitative ways in which I was looking at the portfolio and 00:56:55.920 --> 00:56:59.357 could he help me think through computing >> and could he help me think through sort 00:56:59.440 --> 00:57:04.877 of portfolio optimization and I just got more and more uh intrigued. I mean, you 00:57:04.960 --> 00:57:07.917 see the brain. I mean, he just thinks about things. Um, yeah, 00:57:08.000 --> 00:57:12.077 >> he doesn't hold back. I mean, so he's so I always knew where he stood with uh 00:57:12.160 --> 00:57:14.317 with me and with AIG, but just developed 00:57:14.400 --> 00:57:17.037 a very strong trusting relationship 00:57:17.120 --> 00:57:19.597 >> and there's such a tremendous partner 00:57:19.680 --> 00:57:24.477 that we're able to iterate with them almost like no other company because we 00:57:24.560 --> 00:57:28.157 do things in 90-day increments >> because going out like a year or two 00:57:28.240 --> 00:57:31.677 years is is too static. And so, we actually build 00:57:31.760 --> 00:57:33.837 >> our relationship on 90-day goals. Okay. 00:57:33.920 --> 00:57:35.597 And that's been incredibly effective. 00:57:35.680 --> 00:57:39.997 >> What is uh you know a lot of the AI companies talking about scaling laws, 00:57:40.080 --> 00:57:42.397 exponential growth in token production 00:57:42.480 --> 00:57:47.117 or even revenue in many cases, but what's growing exponentially in your 00:57:47.200 --> 00:57:49.277 business? Are you bringing exponentially 00:57:49.360 --> 00:57:52.317 more data into the platform every year? 00:57:52.400 --> 00:57:55.837 Exponentially more compute resources, 00:57:55.920 --> 00:57:58.317 teams, number of policies like what what 00:57:58.400 --> 00:58:02.877 is the what is the thing that's experiencing a boom right now? most 00:58:02.960 --> 00:58:06.637 important part I believe in terms of business is that you have to have a 00:58:06.720 --> 00:58:10.557 business solution you're trying to solve for. So for us it was >> more data. 00:58:10.640 --> 00:58:12.877 >> Yeah. >> Um better data. >> Yeah. 00:58:12.960 --> 00:58:16.237 >> And then reduce cycle time. So in other words like when we get the data that 00:58:16.320 --> 00:58:21.117 comes in from our distribution partners, how fast can we get it with higher 00:58:21.200 --> 00:58:24.477 quality data and more data to the underwriter to make decisions? 00:58:24.560 --> 00:58:27.117 >> Got it. >> Um and then how do we actually make 00:58:27.200 --> 00:58:29.917 >> what's an example of distribution partner in this context? So it would be 00:58:30.000 --> 00:58:32.157 like a insurance broker or insurance 00:58:32.240 --> 00:58:35.757 agent um or you know someone who has 00:58:35.840 --> 00:58:37.837 >> their client is the product effectively. 00:58:37.920 --> 00:58:42.077 Exactly. Okay. Yes. Yeah, that makes sense. Um what else? Jordy, do you have 00:58:42.160 --> 00:58:47.277 something? >> Uh where was I going to go? The 00:58:47.360 --> 00:58:51.037 >> Alex ontologies. We'll get there. So So 00:58:51.120 --> 00:58:53.677 there's been uh we we primarily I mean 00:58:53.760 --> 00:58:56.077 we at least started covering early stage 00:58:56.160 --> 00:58:58.397 startups. There's been a debate uh in 00:58:58.480 --> 00:59:01.677 our kind of little subindustry right now 00:59:01.760 --> 00:59:07.037 uh around a bunch of new uh insurance focused startups that are growing 00:59:07.120 --> 00:59:08.397 incredibly quickly. 00:59:08.480 --> 00:59:10.957 >> Uh and there's a debate going on is one 00:59:11.040 --> 00:59:14.397 uh maybe AI makes it more possible to 00:59:14.480 --> 00:59:17.197 underwrite risk and if you can do that 00:59:17.280 --> 00:59:20.157 well grow very quickly. Uh the other 00:59:20.240 --> 00:59:23.037 side, you know, says uh hey, you know, 00:59:23.120 --> 00:59:28.077 if you're hypers scaling an insurance company, uh maybe that's not maybe you 00:59:28.160 --> 00:59:31.117 don't want to work with a company that is, you know, going through that hyper 00:59:31.200 --> 00:59:34.397 >> the iron law of the universe goes up fast. 00:59:34.480 --> 00:59:36.877 >> But yeah, talk talk about um talk about 00:59:36.960 --> 00:59:41.757 what AI has actually enabled, where you're excited about it, where it's 00:59:41.840 --> 00:59:45.837 failing broadly, maybe where it's overhyped, and you can I guess tie that 00:59:45.920 --> 00:59:50.157 into uh everything you built with Palunteer. 00:59:50.240 --> 00:59:52.317 There's never been a time in my opinion 00:59:52.400 --> 00:59:55.117 whether it was you know introduction to 00:59:55.200 --> 00:59:58.557 fintech and sh how to use algorithms how 00:59:58.640 --> 01:00:01.197 to build data lakes and repositories for 01:00:01.280 --> 01:00:06.797 data there's never been a time in in my professional career so it's 35 years in 01:00:06.880 --> 01:00:11.197 big companies yeah >> that I've seen the ability to change how 01:00:11.280 --> 01:00:13.597 an an organization actually runs itself 01:00:13.680 --> 01:00:15.837 and that can come from big companies 01:00:15.920 --> 01:00:19.277 like Palunteer or Google or it could 01:00:19.360 --> 01:00:21.437 come from uh you know companies that are 01:00:21.520 --> 01:00:24.557 being funded by venture and have a very 01:00:24.640 --> 01:00:28.957 specific niche that can be you know additive to the organization. And what 01:00:29.040 --> 01:00:30.957 what I think is happening we talked 01:00:31.040 --> 01:00:33.677 about the sort of data ingestion portion 01:00:33.760 --> 01:00:36.077 getting that into a digital workflow 01:00:36.160 --> 01:00:38.317 using large language models to extract 01:00:38.400 --> 01:00:42.077 more data from what comes in but also uh 01:00:42.160 --> 01:00:43.997 helping underwriters make decisions that 01:00:44.080 --> 01:00:47.117 are you know more comprehensive. You 01:00:47.200 --> 01:00:49.917 also have the ability in the way in 01:00:50.000 --> 01:00:52.717 which you service customers to be much 01:00:52.800 --> 01:00:57.517 better through the use of AI. I think companies generally uh my observations 01:00:57.600 --> 01:01:01.277 are struggling with the orchestration of how you actually drive agents, people 01:01:01.360 --> 01:01:03.837 and data into an organization and once 01:01:03.920 --> 01:01:09.037 that is solved and it's certainly on it on its way capabilities are there um 01:01:09.120 --> 01:01:12.877 then you start to think about the entire endtoend chain being very different. 01:01:12.960 --> 01:01:15.357 >> Yeah. What I think about Palunteer while 01:01:15.440 --> 01:01:20.637 they've been such a critical partner is one as we evolved together but in that 01:01:20.720 --> 01:01:25.197 data ingestion to be able to take structured unstructured text all sorts 01:01:25.280 --> 01:01:28.397 of data and get into a workflow in a fraction of the time helps us on the 01:01:28.480 --> 01:01:31.837 things I try to achieve is like we have now data that we probably wouldn't have 01:01:31.920 --> 01:01:35.437 used before because it wasn't good or we couldn't translate it couldn't get it 01:01:35.520 --> 01:01:38.237 into the digital workflow. Um, and then 01:01:38.320 --> 01:01:43.117 we start to build out an ontology. And I, and I really do think it's incredibly 01:01:43.200 --> 01:01:47.117 important. If there's one thing I look at for our organization, certainly the 01:01:47.200 --> 01:01:49.837 advancements of LLM's, their ability to 01:01:49.920 --> 01:01:54.237 do things more autonomously now where we started with the binary gen AI, now 01:01:54.320 --> 01:01:57.437 we're into a Gentic AI where we can just do things autonomously for so much 01:01:57.520 --> 01:02:00.397 longer. without the ontology of actually 01:02:00.480 --> 01:02:05.117 building like what the sort of digital twin of your business looks like where 01:02:05.200 --> 01:02:08.317 you take it and how you evolve it becomes very challenging. So we've been 01:02:08.400 --> 01:02:11.277 able to do things with Palunteer. I'll 01:02:11.360 --> 01:02:15.997 use the ontology example again. We did the full ontology of AIG and then we 01:02:16.080 --> 01:02:20.317 went to look at an acquisition um called Everest which had about $2 billion of 01:02:20.400 --> 01:02:22.637 premium. >> We got Palunteering to work with our 01:02:22.720 --> 01:02:24.237 team. We could build an ontology of 01:02:24.320 --> 01:02:29.837 Everest's portfolio on top of ours in 4 days. Um and quite frankly what we 01:02:29.920 --> 01:02:33.917 started to learn again about that evolution is that you always relied on 01:02:34.000 --> 01:02:36.637 data lakes or global data repositories. 01:02:36.720 --> 01:02:39.357 What we found is that we could get you 01:02:39.440 --> 01:02:44.557 know sort of foundry and start to build out this ontology with going to the 01:02:44.640 --> 01:02:47.917 admin platforms. All of a sudden these repositories and the central places of 01:02:48.000 --> 01:02:51.197 getting data and make sure it's scrubbed wasn't as relevant. So I think we 01:02:51.280 --> 01:02:55.357 continue to advance that >> in in the way in which we are looking at 01:02:55.440 --> 01:02:59.597 our business. I have a la I have one last question. Um just on the actual 01:02:59.680 --> 01:03:04.157 change management, the organization like how the office feels. What how did you 01:03:04.240 --> 01:03:06.157 go about actually working with 01:03:06.240 --> 01:03:08.797 Palunteer? Do you set up your own 01:03:08.880 --> 01:03:12.717 internal Palunteer workforce who sits 01:03:12.800 --> 01:03:15.357 alongside FDES? Do you let Palunteer 01:03:15.440 --> 01:03:18.477 come in and plug in like one person per 01:03:18.560 --> 01:03:22.717 team that you have set up? Like was there a best practice? Did you go with 01:03:22.800 --> 01:03:27.437 the best practice? like what was the actual like experience of deploying the 01:03:27.520 --> 01:03:30.237 forward deployed engineers they get deployed into the organization that's 01:03:30.320 --> 01:03:33.917 got to be uh a unique situation 01:03:34.000 --> 01:03:37.757 >> first is making sure Alex and then you know two of the senior executives Ryan 01:03:37.840 --> 01:03:40.797 and Ted that everybody knows what we're trying to do together so we start there 01:03:40.880 --> 01:03:43.997 >> then we wanted to embed the engineers with our team so if we had a business 01:03:44.080 --> 01:03:47.837 leader that was trying to drive the underwriting output you'd have you know 01:03:47.920 --> 01:03:51.837 technology from AIG you would have some of the change management but you'd have 01:03:51.920 --> 01:03:56.237 the engineers sitting there with our teams throughout the entire process because 01:03:56.320 --> 01:03:59.597 >> the iteration is really important in terms of translating what you're trying 01:03:59.680 --> 01:04:03.437 to achieve from the business side and the engineers actually helping us think 01:04:03.520 --> 01:04:06.397 through the application of some of the LLMs or ways in which we could 01:04:06.480 --> 01:04:08.157 circumvent some of the things that we were doing. 01:04:08.240 --> 01:04:10.717 >> Yeah, that makes a ton of sense. Jordan, anything else? 01:04:10.800 --> 01:04:14.797 >> Uh no. Well, insurance has to be the most important topic. 01:04:14.880 --> 01:04:17.517 >> No, the last if we do have a second I I 01:04:17.600 --> 01:04:21.997 was uh uh >> not sure on timing. what how are you how 01:04:22.080 --> 01:04:26.877 are you thinking about you know workforce planning uh asked uh Karp 01:04:26.960 --> 01:04:30.477 about this and he said to ask >> token budgets um you know we we've 01:04:30.560 --> 01:04:32.957 stayed uh you know as as you've had this 01:04:33.040 --> 01:04:37.277 wave of AI layoffs we've been uh over 01:04:37.360 --> 01:04:42.237 and over and over reminded people that uh if you have a an individual and you 01:04:42.320 --> 01:04:45.437 give them more capability you make them more productive you make them more 01:04:45.520 --> 01:04:49.357 efficient a a thriving business will want to hire more people right because 01:04:49.440 --> 01:04:53.197 you get more individual and so we've tried to remind 01:04:53.280 --> 01:04:57.197 people of that over and over and over as you know companies that often times are 01:04:57.280 --> 01:04:59.117 you know underperforming or bloated for 01:04:59.200 --> 01:05:01.437 whatever reason but what's your kind of 01:05:01.520 --> 01:05:04.557 philosophy around uh hiring headcount 01:05:04.640 --> 01:05:07.117 planning uh riffs all that stuff in this 01:05:07.200 --> 01:05:11.677 kind of uh new era we've been focusing on I heard Alex at the tail end and I 01:05:11.760 --> 01:05:13.837 agree with him so we're focusing on 01:05:13.920 --> 01:05:16.557 growth uh we're focusing on reskilling 01:05:16.640 --> 01:05:21.357 and actually training um our employees to be in different part of the workflow. 01:05:21.440 --> 01:05:25.837 Now >> you would do this I believe in all of this you have to still have great 01:05:25.920 --> 01:05:29.997 endto-end process and so things that have been the humans been an LLM trained 01:05:30.080 --> 01:05:34.317 how to do things like outside of the normal workflow has to you have to get 01:05:34.400 --> 01:05:36.877 rid of that. I mean so I think that's just normal business. Yeah. 01:05:36.960 --> 01:05:41.677 >> Um but you know our aspiration is not to implement you know AI or anything that 01:05:41.760 --> 01:05:44.877 we're doing with our partners to eliminate jobs. I mean, it's about 01:05:44.960 --> 01:05:49.277 growth, reskilling, and finding ways in different markets to have exponential 01:05:49.360 --> 01:05:52.637 growth and opportunity and having a lot more insight in the business that we run. 01:05:52.720 --> 01:05:54.317 >> That's a great optimistic vision. I love 01:05:54.400 --> 01:05:56.877 it. Thank you so much for taking time to come with us. Thanks for coming. Great 01:05:56.960 --> 01:05:58.317 to be with you. >> Have a great rest of your time. 01:05:58.400 --> 01:06:02.237 >> Thanks. >> And up next, next >> we have Chad Walquist. First, I'm going 01:06:02.320 --> 01:06:04.957 to tell you about Crowd Strike. Your business is AI. Their business is 01:06:05.040 --> 01:06:09.677 securing it. Crowd Strike secures AI and stops breaches. Welcome to the show. How 01:06:09.760 --> 01:06:12.077 are you doing, Chad? >> Great overcoat. 01:06:12.160 --> 01:06:15.357 That's a new one. That's an Eleaniano special. It is. 01:06:15.440 --> 01:06:18.637 >> Oh, yeah. He is the master. 01:06:18.720 --> 01:06:19.757 >> Giving us a run for our money. 01:06:19.840 --> 01:06:22.957 >> Yeah, it's fantastic. Uh, anyway, kick us off with an introduction on yourself, 01:06:23.040 --> 01:06:25.917 how you fit into Palunteer, a little bit of backstory. I'm sure we have a ton of 01:06:26.000 --> 01:06:29.117 questions to run through. >> Uh, first, how often do you guys do 01:06:29.200 --> 01:06:32.477 these things? Because it feels like this feels like an annual It feels like an 01:06:32.560 --> 01:06:36.157 annual event. >> Yeah, >> but we're getting the call every month. 01:06:36.240 --> 01:06:40.157 Karp carp karp talks about you know manipulating time [laughter] working you 01:06:40.240 --> 01:06:45.357 know a quarter at palunteer is like a day or a year at an exact so that it 01:06:45.440 --> 01:06:49.037 kind of makes sense. >> Yeah I'm like actually 23. [laughter] 01:06:49.120 --> 01:06:53.117 >> Yeah time the time warp is real. So we do these quarterly. So I'm I'm a ford 01:06:53.200 --> 01:06:54.717 deployed architect technically. Yeah. 01:06:54.800 --> 01:06:58.477 >> I do what is needed and so doing the needful is kind of the palunteer way is 01:06:58.560 --> 01:07:01.677 like there's no job below me. And so no matter if I'm out on the edge with 01:07:01.760 --> 01:07:05.597 customers, I'm talking to executives, explaining the ontology, doing YouTube 01:07:05.680 --> 01:07:08.157 videos. >> That's all what I'm doing. So really the 01:07:08.240 --> 01:07:10.957 goal is how do we help people decomp problems differently and apply the 01:07:11.040 --> 01:07:14.237 technology new ways. >> Can AI do decomp? 01:07:14.320 --> 01:07:17.037 >> Yes. >> Okay. Unpack that because that feels 01:07:17.120 --> 01:07:19.997 like the secret sauce. That feels like the special thing about Palunteer is 01:07:20.080 --> 01:07:23.517 actually being able to bring someone in who understands an organization. I think 01:07:23.600 --> 01:07:27.757 a lot of people see AI tools. A lot of people see AI tools. No, a lot of people 01:07:27.840 --> 01:07:30.157 see AI tools and they and they think uh 01:07:30.240 --> 01:07:33.277 okay very defined workflow input output 01:07:33.360 --> 01:07:35.997 but now instead of just math that Python 01:07:36.080 --> 01:07:37.357 can deal with you can deal with some 01:07:37.440 --> 01:07:41.357 text and that's great but demp to me has 01:07:41.440 --> 01:07:44.477 always felt less like let's go into your 01:07:44.560 --> 01:07:47.037 HR system and understand the basic job 01:07:47.120 --> 01:07:49.197 description and like oh someone uploaded 01:07:49.280 --> 01:07:52.717 this resume versus oh Steve actually 01:07:52.800 --> 01:07:54.957 does this completely outside of that 01:07:55.040 --> 01:07:57.277 system and marketing has two two 01:07:57.360 --> 01:07:59.997 platforms for this thing and engineering 01:08:00.080 --> 01:08:04.557 has three systems for CAD files and the all the clues that have built up over 01:08:04.640 --> 01:08:08.797 decades sometimes hundreds of years for some of these organizations like you 01:08:08.880 --> 01:08:13.517 that's what was so special about the forward deployed engineer program the 01:08:13.600 --> 01:08:17.357 Palunteer model y >> I'm surprised to hear you say I AI can 01:08:17.440 --> 01:08:21.597 do it at all it feels like the final boss >> well this is where the the really the 01:08:21.680 --> 01:08:24.957 palunteer thesis is humans and AI working together 01:08:25.040 --> 01:08:27.997 >> and so the way we thing about this is modeling our business process and we 01:08:28.080 --> 01:08:30.717 heard some other people talking about this of modeling my business process and 01:08:30.800 --> 01:08:35.837 the ontology um because the LMS don't necessarily have an a world view or 01:08:35.920 --> 01:08:39.597 world model of your business and your operations the ontology provides that okay 01:08:39.680 --> 01:08:42.797 >> and so when we talk about demp this is really about actually now I make more 01:08:42.880 --> 01:08:45.837 data computable as well so we think about LM on the agents and I'm 01:08:45.920 --> 01:08:50.077 interacting also we use LM to make more data computable and then model that in 01:08:50.160 --> 01:08:53.357 the ontology of how things are really working and so what we're actually doing 01:08:53.440 --> 01:08:55.197 a lot of times now is is building out 01:08:55.280 --> 01:08:57.197 that world view and then running 01:08:57.280 --> 01:09:02.237 multiple agents over this actually um being combative towards each other right 01:09:02.320 --> 01:09:06.157 and so actually working against each other and having critiques and so after 01:09:06.240 --> 01:09:09.437 you do that you can also then give the human human in the loop feedback about 01:09:09.520 --> 01:09:12.157 this and iterate on this and so what we find is that's really a scaling 01:09:12.240 --> 01:09:15.197 mechanism it's like a new power tool right I think you guys were just talking 01:09:15.280 --> 01:09:18.957 about this the kind of the perspective around jobs and all this stuff it's like 01:09:19.040 --> 01:09:22.157 when you gave carpenters power tools there weren't less carpenters there were 01:09:22.240 --> 01:09:24.797 more I could do more with it right it's an empowering thing 01:09:24.880 --> 01:09:28.957 >> yeah So, uh, how often like I I I I'm 01:09:29.040 --> 01:09:33.037 interested in the, uh, like the pie [clears throat] in the sky, Palunteer 01:09:33.120 --> 01:09:35.837 pitch, understand your entire business, 01:09:35.920 --> 01:09:38.797 run your entire business on Palunteer. 01:09:38.880 --> 01:09:43.037 And then some of the nitty-gritty where sometimes like the lowhanging fruit is 01:09:43.120 --> 01:09:47.677 like, wait, there's a like there's someone's job to just like take a form 01:09:47.760 --> 01:09:51.757 and type it into a sheet. Like, we have we've had image recognition for a long 01:09:51.840 --> 01:09:56.077 time. let's actually go and implement that and get that into a database, get 01:09:56.160 --> 01:09:59.197 that into the ontology, get that into Palunteer so then we can start building 01:09:59.280 --> 01:10:03.357 on top of it. And it feels like there might be a tension there. Obviously, 01:10:03.440 --> 01:10:05.437 both processes are speeding up, but h 01:10:05.520 --> 01:10:10.477 how do you how do you sort of like keep the project centered around the big goal 01:10:10.560 --> 01:10:13.437 while still chopping wood on all the things that actually need to happen? 01:10:13.520 --> 01:10:14.557 >> Yeah. And I think this comes back to the 01:10:14.640 --> 01:10:17.597 forward deployed piece and like what do we deliver outcomes and and we work 01:10:17.680 --> 01:10:19.837 backwards from that rather than hey I have this data I'm going to build a data 01:10:19.920 --> 01:10:22.557 warehouse and then I'll build reports because all my data is in one place 01:10:22.640 --> 01:10:25.357 that's the that's the field of dreams and no one shows up 01:10:25.440 --> 01:10:28.957 >> right and so really when we demp things and work backwards from that you know 01:10:29.040 --> 01:10:32.717 the simple things like the form filling out there's a lot of that now the one 01:10:32.800 --> 01:10:35.757 approach that we see a lot is you know enterprise software is going to force 01:10:35.840 --> 01:10:39.757 you into their box. >> Sure. >> Right. You go fit you go fit into this box. 01:10:39.840 --> 01:10:43.517 >> Yeah. Well, then you know, okay, did I take away the special sauce which was my 01:10:43.600 --> 01:10:47.037 company because people were doing these all these kind of amalgamations. Hey, 40 01:10:47.120 --> 01:10:50.637 ways to do a PO. >> Yep. >> Well, maybe it is okay to do 40 ways, 01:10:50.720 --> 01:10:53.437 but my software can't handle it and it's fragmented, right? And so there's 01:10:53.520 --> 01:10:56.477 there's actually a middle ground because, you know, for a long time 01:10:56.560 --> 01:11:00.077 >> customization was kind of a four-letter word, right? No, no one wanted to do 01:11:00.160 --> 01:11:02.797 that. And I think that's where we think about malleable software actually. How 01:11:02.880 --> 01:11:04.877 do we help you be more different, not more similar? 01:11:04.960 --> 01:11:07.997 >> Interesting. And that's so that when we decomp problems thinking about not only 01:11:08.080 --> 01:11:12.637 the the kind of the uh quantitative piece but the qualitative piece and the 01:11:12.720 --> 01:11:16.477 people and process around this how do we enable those people to do the things 01:11:16.560 --> 01:11:19.437 that made them special? >> Is is software getting more malleable 01:11:19.520 --> 01:11:21.757 because I I I can look at it two ways. I 01:11:21.840 --> 01:11:25.437 can look at uh one you know obviously AI 01:11:25.520 --> 01:11:30.157 agents are incredible at coding. They can run they they can make changes very 01:11:30.240 --> 01:11:33.037 very quickly that would take you a day in just a few minutes. Y 01:11:33.120 --> 01:11:36.717 >> at the same time I see you know so many screenshots of people saying I 01:11:36.800 --> 01:11:41.037 implemented this feature and the GitHub is plus a million lines of code and at a 01:11:41.120 --> 01:11:43.517 certain point like the context window is 01:11:43.600 --> 01:11:45.757 growing as fast as the code generation's 01:11:45.840 --> 01:11:48.637 growing like there's a I I'm a believer 01:11:48.720 --> 01:11:51.037 in the answer to bad slop is good slop 01:11:51.120 --> 01:11:55.917 and more slop maybe but what are you actually seeing on the malleability of 01:11:56.000 --> 01:11:59.677 software because sometimes the most malleable software in the past has been 01:11:59.760 --> 01:12:01.677 oh well there was a really incredible 01:12:01.760 --> 01:12:03.677 engineer who figured out this problem 01:12:03.760 --> 01:12:07.037 and baked it down to a 2,000line repo 01:12:07.120 --> 01:12:10.717 and you can actually just put it in your own context window so it becomes more 01:12:10.800 --> 01:12:13.997 malleable and you can use it as a building block. Yep. And [snorts] that 01:12:14.080 --> 01:12:18.157 feels like that's going away and I want to make sure that we've that we're ready 01:12:18.240 --> 01:12:20.557 for when it goes away and it remains malleable. 01:12:20.640 --> 01:12:24.077 >> Well, I think what what's missing is the the malleable enterprise scaffolding 01:12:24.160 --> 01:12:27.117 that you and that's what we think about the ontology and foundry and the 01:12:27.200 --> 01:12:30.957 platform and then Apollo that allows us to go deploy these changes. So it gives 01:12:31.040 --> 01:12:33.517 us the right amount of structure but the right amount of freedom. So I think 01:12:33.600 --> 01:12:37.357 that's the balance we try to find is that malle malleability in the middle 01:12:37.440 --> 01:12:41.197 where we can actually scale. We can enable people to do things differently 01:12:41.280 --> 01:12:46.237 while still creating enterprise you know grade robust secure scalable software. 01:12:46.320 --> 01:12:48.797 And so it's actually a balance there about how 01:12:48.880 --> 01:12:50.797 >> I can enable that engineer that you know 01:12:50.880 --> 01:12:54.877 has been doing that now they can write code much faster. they can oversee 01:12:54.960 --> 01:12:58.477 things and that enterprise scaffolding in the middle allows us to actually 01:12:58.560 --> 01:13:01.757 create the right guard rails create a safe system of work for them to go 01:13:01.840 --> 01:13:05.037 develop things in. Um and then it's also the feedback loop. So the other thing 01:13:05.120 --> 01:13:08.397 that we do with our ontology and our platforms is implicit and explicit 01:13:08.480 --> 01:13:12.637 feedback from users using it. So the udal loop that I create and really that 01:13:12.720 --> 01:13:15.517 udal loop allows our customers as they're doing workflows they're giving 01:13:15.600 --> 01:13:19.197 feedback to agents. Now can agents help them do more based on the feedback. So 01:13:19.280 --> 01:13:22.077 both explicitly saying hey that was wrong that sucked or I chose this 01:13:22.160 --> 01:13:25.997 option. Now if you do that enough agents can start to learn from that. So we 01:13:26.080 --> 01:13:29.197 actually store that in our ontology to allow it to scale. So it's really that 01:13:29.280 --> 01:13:32.637 human centric process around AI. AI is not like we shouldn't be thinking about 01:13:32.720 --> 01:13:37.837 AI from the sake of AI for AI. It's AI to enable humans to do more. Yeah, 01:13:37.920 --> 01:13:41.757 >> that's the frame. >> Udaloo, observe, orient, decide, act, 01:13:41.840 --> 01:13:45.037 right? Uh I have a different question, but you can you can go. 01:13:45.120 --> 01:13:48.157 >> Uh if you were giving if you had 30 01:13:48.240 --> 01:13:49.037 minutes 01:13:49.120 --> 01:13:53.357 >> to uh give feedback to the AI labs, what 01:13:53.440 --> 01:13:55.517 are the kind of key areas? Let's say the 01:13:55.600 --> 01:13:58.477 frontier labs, right? Uh leading models. 01:13:58.560 --> 01:14:02.557 What uh what are the kind of key areas that you would be focused on? 01:14:02.640 --> 01:14:05.117 >> Yeah, I mean I think when we think about the enterprise space, you know, we 01:14:05.200 --> 01:14:06.797 >> one you're like don't compete with us. 01:14:06.880 --> 01:14:10.397 No, I actually like I I think optionality is a good thing. Like I am 01:14:10.480 --> 01:14:13.757 agnostic to where you store your data, where you store, how what model you 01:14:13.840 --> 01:14:17.197 choose, what compute you use. So like we we can allow you to use any of that 01:14:17.280 --> 01:14:20.637 because the last thing that actually drives an outcome is replatforming, 01:14:20.720 --> 01:14:23.357 moving to another >> and that goes back to the onrem culture, 01:14:23.440 --> 01:14:28.077 the secure cloud culture, ITAR compliance, like this is in the DNA of 01:14:28.160 --> 01:14:30.477 the company. >> Yeah. And so how do we actually enable 01:14:30.560 --> 01:14:33.917 people where they are instead of the focus on oh if you replplatform 01:14:34.000 --> 01:14:37.197 everything to Palanteer everything will be great. like well actually you've 01:14:37.280 --> 01:14:39.677 probably been replplatforming for years. 01:14:39.760 --> 01:14:42.637 Can we enable what you have to go do these new things? So when we think about 01:14:42.720 --> 01:14:45.597 like the model companies and it's you know how do we ensure that we can give 01:14:45.680 --> 01:14:49.277 the feedback loops around you know tool usage and um you know 01:14:49.360 --> 01:14:53.117 >> yeah that's the kind of that's the kind of stuff I was uh wanting to get your 01:14:53.200 --> 01:14:57.117 point of view on is like are I'm sure you're getting into the nitty-gritty 01:14:57.200 --> 01:15:01.597 >> with individual models where where they're spiky where there's you know 01:15:01.680 --> 01:15:03.197 where there's shortcomings etc. 01:15:03.280 --> 01:15:06.077 >> Yeah. So we we actually just launched I just put a YouTube video out last week 01:15:06.160 --> 01:15:08.477 on this new tool called Evolve. We talked about it in the kind of the 01:15:08.560 --> 01:15:12.717 halftime show where customers are using actually AI to help them understand 01:15:12.800 --> 01:15:17.757 which model. So like maybe you know the the the the meme around hey make it 01:15:17.840 --> 01:15:19.277 exist first and then make it good. 01:15:19.360 --> 01:15:21.597 >> Yeah. >> Most of the time I see people a building 01:15:21.680 --> 01:15:25.037 with agents they're using the latest frontier model. I just got it working 01:15:25.120 --> 01:15:27.757 and then then all of a sudden the token maxing and everyone everything else and 01:15:27.840 --> 01:15:30.397 you're like oh my gosh I just blew through my whole budget. Yep. So we 01:15:30.480 --> 01:15:33.917 built a tool called evolve that will actually go analyze the logs in 01:15:34.000 --> 01:15:37.037 production about how these models are operating, what people are doing with 01:15:37.120 --> 01:15:41.357 them, the architecture over it, um, and actually be able to swap out different 01:15:41.440 --> 01:15:45.117 models from different providers or hey actually for most of this workflow you 01:15:45.200 --> 01:15:48.717 can use this model that's older and actually without thinking and and test 01:15:48.800 --> 01:15:51.917 time compute it. It's more deterministic >> or even cached models 01:15:52.000 --> 01:15:54.877 >> cache models and then or hey if you actually just have this piece of data in 01:15:54.960 --> 01:15:58.717 the ontology then you would eliminate all this and 50% of your cost. Yep. And 01:15:58.800 --> 01:16:01.117 so you know some of the customers McCarthy talked about this at our 01:16:01.200 --> 01:16:05.437 halftime you know they they were able to in two days eliminate 60% of their token 01:16:05.520 --> 01:16:09.997 cost by rearchitecting picking a different model and in prompt tuning. So 01:16:10.080 --> 01:16:12.877 it's the combination of all those the permutations get really hard especially 01:16:12.960 --> 01:16:17.197 when it's in this probabistic models we've have tools to do this in the 01:16:17.280 --> 01:16:19.757 deterministic world >> prompt tuning it's a instead of don't 01:16:19.840 --> 01:16:23.517 make mistakes it's okay to make some mistakes 01:16:23.600 --> 01:16:27.357 if the mistake is going to cost just a little bit I'm fine because don't make 01:16:27.440 --> 01:16:30.237 mistakes that's going to cost me a fortune. Well, there there were there's 01:16:30.320 --> 01:16:32.477 there was some chatter yesterday around 01:16:32.560 --> 01:16:38.317 uh uh something a model was doing to be more efficient was uh talking in in like 01:16:38.400 --> 01:16:42.637 this bad. >> Oh. Oh. Kan caveman caveman prompting. 01:16:42.720 --> 01:16:44.397 Um >> the caveman prompt method actually 01:16:44.480 --> 01:16:47.597 works. What how how often are you 01:16:47.680 --> 01:16:52.397 working with a company that is having call it like a mini chat GBT moment 01:16:52.480 --> 01:16:56.157 within their enterprise and then they're just like let's not tell anyone about 01:16:56.240 --> 01:17:01.117 this because I imagine like there's all these there's clearly places where 01:17:01.200 --> 01:17:03.997 >> what does that mean their product is taking off like Chad GBT or 01:17:04.080 --> 01:17:08.397 >> well so they've found a way to apply AI in a way that is highly highly effective 01:17:08.480 --> 01:17:11.117 and gives them an edge >> oh interesting >> uh but 01:17:11.200 --> 01:17:14.717 >> yeah like like the theoretical like technology 01:17:14.800 --> 01:17:18.717 transform. So, so X people are very loud, right? Figure uh they they're 01:17:18.800 --> 01:17:20.717 like, I just had this >> done using everything. 01:17:20.800 --> 01:17:24.957 >> Yeah, I just had a product work for 30 hours on this thing. They'll talk about 01:17:25.040 --> 01:17:29.037 it. But if you're a Fortune 500 and you figure out how to do something, it's not 01:17:29.120 --> 01:17:32.237 like you want to like put put your hand up and say like, I figured something 01:17:32.320 --> 01:17:37.277 out, right? Like secrets are valuable and >> these advancements and kind of 01:17:37.360 --> 01:17:40.077 breakthroughs are not going to be uniform. the airline industry will never 01:17:40.160 --> 01:17:43.277 be the same and then your direct competitor copies [laughter] you and 01:17:43.360 --> 01:17:47.277 you're like, >> "Yeah." >> And so, and so part of part of why, you 01:17:47.360 --> 01:17:49.117 know, right now the meme is token 01:17:49.200 --> 01:17:52.237 maxing. Um, and that's an obvious going 01:17:52.320 --> 01:17:54.477 to be an obvious era area of debate. 01:17:54.560 --> 01:17:58.077 People are happy to go talk about it, say, you know, CEOs might say, "Hey, 01:17:58.160 --> 01:18:02.077 let's stop doing this." Um, but there has to be all these other kind of 01:18:02.160 --> 01:18:07.037 pockets of interesting moments where we won't hear about them until they become 01:18:07.120 --> 01:18:09.117 kind of like standard operating procedures 01:18:09.200 --> 01:18:11.997 >> or you see it in the the the uh earnings 01:18:12.080 --> 01:18:13.277 and the economics piece, right? Yeah. 01:18:13.360 --> 01:18:18.397 Yeah. So, I Yes. Unfortunately, X is not the real world, [laughter] 01:18:18.480 --> 01:18:20.957 >> you know, and there there's a lot of grift and noise and, you know, 01:18:21.040 --> 01:18:23.917 podcasting, PMing and, you know, that kind of stuff that goes on, 01:18:24.000 --> 01:18:27.357 >> but I I I think in the real world, yes, there is the halves and have nots. I 01:18:27.440 --> 01:18:31.677 mean, we were just talking about AIG, like when you can start to actually do 01:18:31.760 --> 01:18:35.917 the underwriting and, you know, have quotes back in hours or days instead of 01:18:36.000 --> 01:18:41.037 months on these highly complex enterprise, you know, kind of insurance agreements. 01:18:41.120 --> 01:18:43.677 >> If you don't have that, how are you ever going to compete? Yeah. 01:18:43.760 --> 01:18:46.957 >> And so when we think about this of the N of one, right, you know, that those are 01:18:47.040 --> 01:18:49.277 the companies that we're going after and we see where there are those moments 01:18:49.360 --> 01:18:51.517 that are not public. It's the competitive advantage. interesting 01:18:51.600 --> 01:18:54.397 category because you can imagine AIG, 01:18:54.480 --> 01:18:57.597 you know, is um, you know, working with 01:18:57.680 --> 01:19:01.757 a potential customer or renewing a policy and that customer is going and 01:19:01.840 --> 01:19:03.837 talking to all of AIG's competitors. 01:19:03.920 --> 01:19:04.317 Yep. 01:19:04.400 --> 01:19:08.077 >> And uh, if AIG is able to turn around, 01:19:08.160 --> 01:19:11.437 >> you know, a quote or a policy in 24 01:19:11.520 --> 01:19:14.237 hours and then it takes another player, 01:19:14.320 --> 01:19:18.637 >> you know, two weeks because it's, you know, complicated. So many so many teams 01:19:18.720 --> 01:19:22.157 will just say like hey we you know you know especially once you have two bids 01:19:22.240 --> 01:19:25.837 you can basically say like okay this that third fourth fifth we'll kind of 01:19:25.920 --> 01:19:28.637 wait on those because we have a good option here 01:19:28.720 --> 01:19:31.197 >> well it builds trust the other piece here so when you think when you see 01:19:31.280 --> 01:19:35.277 people operating that with that level of efficiency what else can you do so I see 01:19:35.360 --> 01:19:38.717 this whether I'm doing you know SAP migrations the least sexy thing you 01:19:38.800 --> 01:19:43.597 could talk about but hey if I can cut your SAP migration >> let's give it up for 01:19:43.680 --> 01:19:47.357 >> yeah it's like the the least you know exciting thing on on paper but Actually, 01:19:47.440 --> 01:19:50.797 if if you're spending hundreds of millions of Yeah, you guys get it. But 01:19:50.880 --> 01:19:54.077 hundreds of millions of dollars on a migration and we can cut it in half. 01:19:54.160 --> 01:19:56.237 >> Yeah, >> that's a massive deal. 01:19:56.320 --> 01:19:59.197 >> Uh, back on the udaloop, observe, 01:19:59.280 --> 01:20:02.477 orient, decide, act on the observation 01:20:02.560 --> 01:20:05.197 side, what is the supply and demand 01:20:05.280 --> 01:20:08.157 imbalance for dashboards? 01:20:08.240 --> 01:20:12.397 Like, and what I mean by that is is when you're working with a company, is there 01:20:12.480 --> 01:20:14.477 is there more demand for dashboards? 01:20:14.560 --> 01:20:16.877 more people asking, hey, we need a dashboard for this, we need a dashboard 01:20:16.960 --> 01:20:19.917 for that. And you have to back people off and say, I don't know if the 01:20:20.000 --> 01:20:21.357 dashboard's right for this. Like, you 01:20:21.440 --> 01:20:23.277 might just want to do an ad hoc analysis 01:20:23.360 --> 01:20:27.357 or actually go and see. Uh versus you're 01:20:27.440 --> 01:20:30.557 seeing so much opportunity that you're like, okay, we want to push dashboards 01:20:30.640 --> 01:20:34.317 out everywhere. Like what? Walk me through dashboarding right now because 01:20:34.400 --> 01:20:37.517 I've always been like sort of like, oh, the too many dashboards. You build them 01:20:37.600 --> 01:20:38.637 and then no one looks at them. 01:20:38.720 --> 01:20:39.997 >> Yeah, I want to kill all dashboards. 01:20:40.080 --> 01:20:44.957 >> Okay, that's my perspective. [laughter] dash I mean KPIs and dashboards should 01:20:45.040 --> 01:20:48.317 be a byproduct of operational applications where I'm making decisions. 01:20:48.400 --> 01:20:51.117 So we talk about the udaloop I have to actually act for things to hit the 01:20:51.200 --> 01:20:52.877 bottom line be valuable >> in the actual application 01:20:52.960 --> 01:20:56.557 >> in the application. So as I need those things and it's going to inform a better 01:20:56.640 --> 01:20:58.877 decision. Yeah, >> that's where I want those metrics. That 01:20:58.960 --> 01:21:01.997 should be a byproduct. Not if I go out with the goal of building a dashboard, 01:21:02.080 --> 01:21:04.877 it's going to be the field of dreams again. No one shows up. 01:21:04.960 --> 01:21:07.917 >> And so yes, it should be you you're going to have to build some of those 01:21:08.000 --> 01:21:11.197 things. The other side of this also is when you think about a data warehouse 01:21:11.280 --> 01:21:14.717 like literally I won't go too deep into this technical riff but like you know 01:21:14.800 --> 01:21:19.037 Kimble and dimensional modeling was built in 96 for scaling databases 01:21:19.120 --> 01:21:22.477 and you're still modeling in the same way in 2026 01:21:22.560 --> 01:21:24.717 >> for your dashboard your Tableau whatever 01:21:24.800 --> 01:21:28.637 those things are and like that's not actually how the world works in rows and 01:21:28.720 --> 01:21:31.677 columns you need complex things to model how the world really works and that's 01:21:31.760 --> 01:21:35.117 what we think about the ontology which means I can reuse it for an operational 01:21:35.200 --> 01:21:37.997 application KPIs agents all on one 01:21:38.080 --> 01:21:42.477 single ontology which it makes it the compound effect where as I add things in 01:21:42.560 --> 01:21:47.357 I'm now compounding with each individual decision I'm design working with gets 01:21:47.440 --> 01:21:50.717 better and better and better for the next use cases I connect across my business. 01:21:50.800 --> 01:21:53.277 >> Yeah. Is there an analogy there to just 01:21:53.360 --> 01:21:55.437 the deployment of AI tools currently? 01:21:55.520 --> 01:22:00.877 I'm I'm just reflecting on the the NoSQL boom and I don't know how strong this 01:22:00.960 --> 01:22:03.677 was. This is probably just like an online take but this idea of like why 01:22:03.760 --> 01:22:06.477 would you ever want a relational database? Why would you ever want a 01:22:06.560 --> 01:22:08.317 schema? Don't never do a migration ever 01:22:08.400 --> 01:22:11.837 again. Uh and the future looked like a 01:22:11.920 --> 01:22:15.997 win-win almost like I I think Postgress installations probably grew and so did 01:22:16.080 --> 01:22:19.997 MongoDB and other non-reational databases. uh and people use Reddus for 01:22:20.080 --> 01:22:22.957 things and they use all sorts of different tools and we built and we 01:22:23.040 --> 01:22:25.997 stood on the shoulders of giants and we got more giants and then you know that 01:22:26.080 --> 01:22:27.357 means full employment for you obviously 01:22:27.440 --> 01:22:30.717 but uh but I'm wondering like as like 01:22:30.800 --> 01:22:34.717 are you seeing uh glimmers of of the AI 01:22:34.800 --> 01:22:37.197 tools eating into different pieces of 01:22:37.280 --> 01:22:39.197 the technical stacks or is it all like 01:22:39.280 --> 01:22:41.917 yes and across the enterprises? Um I 01:22:42.000 --> 01:22:46.237 think it's yes and and in in a couple different things there is when you think 01:22:46.320 --> 01:22:49.517 about the real world it is not just rows and columns you can't describe 01:22:49.600 --> 01:22:53.357 everything with measures and attributes >> and so it's actually multimodal and so 01:22:53.440 --> 01:22:55.997 like we think about this in our ontology where you can have one semantic object 01:22:56.080 --> 01:23:00.717 that actually has a CAD file and an image a CB model and tabular stuff in 01:23:00.800 --> 01:23:02.477 one semantic thing of a plant. 01:23:02.560 --> 01:23:04.877 >> Yeah. >> Which means I'm starting to talk in the 01:23:04.960 --> 01:23:08.957 language of my business. So being able to have the multimodal representation 01:23:09.040 --> 01:23:12.637 where in other places oh I have to have MongoDB and I have to have a SQL 01:23:12.720 --> 01:23:15.837 database here and I have to have an S3 bucket here to put all of these 01:23:15.920 --> 01:23:20.157 different things to store them in ways well we can do that all in the ontology 01:23:20.240 --> 01:23:23.437 vectors everything else so that that's really the the goal around how do I 01:23:23.520 --> 01:23:27.357 model the real world how it actually works and make that transparent so 01:23:27.440 --> 01:23:30.077 you're not having to figure out which technology to put in a time series thing 01:23:30.160 --> 01:23:34.797 for sensors on a you know oil platform >> don't care right and that that's where 01:23:34.880 --> 01:23:38.317 we want to have the non- differentiated heavy lifting like truly in the platform 01:23:38.400 --> 01:23:40.957 to remove the friction about getting stuff done. 01:23:41.040 --> 01:23:42.717 >> How common is it for a business with 01:23:42.800 --> 01:23:45.437 more than a hund00 million of revenue to 01:23:45.520 --> 01:23:50.397 have very little understanding of how their business actually works? Like 01:23:50.480 --> 01:23:53.517 maybe they own maybe they own maybe they know like the main thing which is like 01:23:53.600 --> 01:23:56.397 you know >> we make a product and try to sell it for 01:23:56.480 --> 01:23:59.037 more than it costs to deliver. Yeah. Um 01:23:59.120 --> 01:24:05.037 but uh but but is is some element of uh 01:24:05.120 --> 01:24:08.477 how how much can chaos and mystery be 01:24:08.560 --> 01:24:12.317 reduced effectively today? Because it feels like we're entering an era like 01:24:12.400 --> 01:24:16.637 >> you go back um >> you know 50 years and 01:24:16.720 --> 01:24:20.157 >> uh >> the level of like mystery in a large 01:24:20.240 --> 01:24:23.997 company would have been like is almost unconceivable today, right? because you 01:24:24.080 --> 01:24:26.317 have different time zones, different 01:24:26.400 --> 01:24:31.997 offices, you know, no email, all that stuff. And now like mystery and chaos is 01:24:32.080 --> 01:24:33.037 probably 01:24:33.120 --> 01:24:36.557 >> uh reduced dramatically, but uh still 01:24:36.640 --> 01:24:39.197 there's companies that that uh maybe 01:24:39.280 --> 01:24:43.597 maybe before you start working with them, I'm curious what those look like. 01:24:43.680 --> 01:24:45.517 >> Yeah, I mean we work with a lot of a lot 01:24:45.600 --> 01:24:49.357 of different varieties of companies. Um you know, I joke that a lot of times, 01:24:49.440 --> 01:24:52.797 you know, companies make money by accident. like they don't actually know 01:24:52.880 --> 01:24:55.597 what their most profitable product is and often they're trying to sell the 01:24:55.680 --> 01:24:58.477 thing that isn't isn't actually the most profitable and actually not selling the 01:24:58.560 --> 01:25:01.837 thing that actually is profitable and it comes back to how they've modeled their 01:25:01.920 --> 01:25:04.877 data is to aggregate it up to KPIs and other metrics when you actually need to 01:25:04.960 --> 01:25:07.597 model at the finest grade how your business operates to get a true cost of 01:25:07.680 --> 01:25:12.557 goods sold for example or true cost to serve like you that's very complicated 01:25:12.640 --> 01:25:15.757 it's very complex so like we really think about how do I embrace that 01:25:15.840 --> 01:25:19.757 complexity so that I can truly understand tactically at the edge how do 01:25:19.840 --> 01:25:23.197 I do more of the things that are good and less of the bad. It's that simple. 01:25:23.280 --> 01:25:26.797 And those get peanut buttered across with KPIs and metrics. And people don't 01:25:26.880 --> 01:25:29.757 actually know how their businesses are operating. I can't tell you whether it's 01:25:29.840 --> 01:25:33.597 a hundred million dollar company or a $50 billion company how many times I see 01:25:33.680 --> 01:25:37.677 this that they don't actually understand how they're making money at a fine grain. 01:25:37.760 --> 01:25:42.077 >> Yeah. >> Uh last question. Uh, 01:25:42.160 --> 01:25:46.157 is there a world in the future where a company gets created, let's say on 01:25:46.240 --> 01:25:49.597 Stripe Atlas, and the first account they sign up for 01:25:49.680 --> 01:25:52.317 >> other than that is, let's say, a Palunteer. 01:25:52.400 --> 01:25:54.957 >> Yes, that's interesting. >> I would love that. And so, we do have a 01:25:55.040 --> 01:25:57.757 Palanteer for builders program. We have small companies that there's people here 01:25:57.840 --> 01:26:01.037 that are two person startups, you know, 01:26:01.120 --> 01:26:04.797 that are working in their attict in Canada. I mean, like, so it it is 01:26:04.880 --> 01:26:08.557 literally um any size company. Come come work. There's a free dev tier people can 01:26:08.640 --> 01:26:11.757 come build. You actually there's actually a Shopify integration in 01:26:11.840 --> 01:26:14.557 Palunteer. You can go hook up to your Shopify and pull in Palanteer. Yeah, 01:26:14.640 --> 01:26:17.677 there are people doing this. Now, are we always great at selling it or telling 01:26:17.760 --> 01:26:20.877 the story? Sure. No. >> But but there are companies doing this 01:26:20.960 --> 01:26:23.917 and I do think there's a day where it's going to be ubiquitous because I also 01:26:24.000 --> 01:26:27.037 think, you know, there's some some guys here that have, you know, they hey, my 01:26:27.120 --> 01:26:29.357 my business is dying. I was, you know, I 01:26:29.440 --> 01:26:34.157 was down 10% negative margin on on what I was selling and through using 01:26:34.240 --> 01:26:37.917 Palanteer. I they watched our YouTube videos and they built it themselves and 01:26:38.000 --> 01:26:40.797 increase to a 9 or 10% positive margin in three months. 01:26:40.880 --> 01:26:43.517 >> That's great. >> And so like people can go do it. I think 01:26:43.600 --> 01:26:46.797 that's the great American story is like how do we enable that and I think we'll 01:26:46.880 --> 01:26:48.477 get there. Um it might take a little time. 01:26:48.560 --> 01:26:52.397 >> I love it. Well, thank you so much for taking the time. Great to see you. We 01:26:52.480 --> 01:26:57.197 will talk soon. Uh our next guest is joining in just 15 minutes. We're going 01:26:57.280 --> 01:27:01.037 to go back to the timeline. First I'm going to tell you about Figma agents 01:27:01.120 --> 01:27:03.677 meet the canvas. So your AI agents can 01:27:03.760 --> 01:27:07.037 now create and modify Figma files with 01:27:07.120 --> 01:27:11.997 design system context. >> It's so crazy how many companies 01:27:12.080 --> 01:27:14.477 >> Yeah. >> are their whole strategy is like we're 01:27:14.560 --> 01:27:16.157 going to hire guys like Chad. Yeah. 01:27:16.240 --> 01:27:18.317 >> And they're gonna they're going to do stuff. 01:27:18.400 --> 01:27:20.397 >> He is he is he is the final boss of 01:27:20.480 --> 01:27:25.597 >> FD meme. What what what drove the FTE meme? Was it was it Palunteer going 01:27:25.680 --> 01:27:29.037 public or was it >> I think it was Palunteer going parabolic? 01:27:29.120 --> 01:27:31.037 >> Maybe. Maybe. Yeah. Once once. 01:27:31.120 --> 01:27:35.517 >> Yeah. because it just it just because before it was like okay yeah successful 01:27:35.600 --> 01:27:38.317 company but like no one really knows where the valuation's going. Now it's 01:27:38.400 --> 01:27:43.917 like my my uncle just told me that he made a bunch of money and so I got pay 01:27:44.000 --> 01:27:47.837 attention >> well that but also they had been banging 01:27:47.920 --> 01:27:52.317 the FTE drum >> or the and getting the consulting 01:27:52.400 --> 01:27:55.277 >> but people had earplugs in to the banging of the drum and the and the 01:27:55.360 --> 01:27:58.477 earplugs came out. >> Yeah. But when you're when they were a 01:27:58.560 --> 01:28:01.597 10 to20 billion dollar company, a lot of 01:28:01.680 --> 01:28:05.757 people could still convince themselves that they were right. It's just a 01:28:05.840 --> 01:28:07.837 consulting business. >> Yeah. Yeah. Yeah. Exactly. But 01:28:07.920 --> 01:28:10.397 >> and that gets harder harder to ignore. 01:28:10.480 --> 01:28:13.357 We covered this very briefly. Um but uh 01:28:13.440 --> 01:28:17.597 but yeah, very excited for Joe Rogan to be hosting his uh you know, 01:28:17.680 --> 01:28:19.437 >> this is rumored. This is a rumored leak. 01:28:19.520 --> 01:28:22.877 It is not confirmed by any means yet, but >> but I like the sound of it. 01:28:22.960 --> 01:28:25.917 >> It would be it's a very different direction. Uh 01:28:26.000 --> 01:28:30.557 >> um this was a good post. I wanted to bring it up. Buco Capital says, "It's 01:28:30.640 --> 01:28:33.997 really incredible the absolute AI garbage in all caps that people are 01:28:34.080 --> 01:28:37.197 comfortable sending to their co-workers and bosses. There's a good chance 01:28:37.280 --> 01:28:40.877 productivity will actually decrease as AI adoption increases because everyone 01:28:40.960 --> 01:28:44.157 is busy waiting through uh AI slop. I 01:28:44.240 --> 01:28:46.477 don't think I don't think it'll actually 01:28:46.560 --> 01:28:48.477 I don't think it'll actually get there. 01:28:48.560 --> 01:28:53.357 But I have had uh I have had moments over the last month where somebody has 01:28:53.440 --> 01:28:55.437 sent me, you know, a deck for their 01:28:55.520 --> 01:28:59.117 company or materials and I can tell that 01:28:59.200 --> 01:29:02.557 >> uh 90% of the work that went into it was 01:29:02.640 --> 01:29:07.037 on prompting. >> Yeah. And uh I have a very like visceral 01:29:07.120 --> 01:29:08.717 reaction toward it especially for like 01:29:08.800 --> 01:29:12.717 early stage companies where uh ideas 01:29:12.800 --> 01:29:14.477 and the way in which you go about doing 01:29:14.560 --> 01:29:19.357 things matter so much that uh it's 01:29:19.440 --> 01:29:23.117 almost like you know painting this initial vision and things like your your 01:29:23.200 --> 01:29:26.237 go to market um product differentiation 01:29:26.320 --> 01:29:31.197 why you'll actually win like use AI to make your team slide that's great right 01:29:31.280 --> 01:29:35.037 just taking like a set of facts and making it look good, right? You're 01:29:35.120 --> 01:29:39.117 giving somebody a bio, something like that. Um, 01:29:39.200 --> 01:29:41.757 >> but I just remember uh I I I got this 01:29:41.840 --> 01:29:44.957 deck. I was clicking through it. Um, and 01:29:45.040 --> 01:29:49.517 I very uh respectfully said like go and 01:29:49.600 --> 01:29:55.277 like do this yourself because uh just because you've made something that that 01:29:55.360 --> 01:29:59.037 looks like a deck. >> Yeah. But you didn't do the sort of like 01:29:59.120 --> 01:30:04.477 fundamental work to actually present this in a way uh if you looked at each 01:30:04.560 --> 01:30:07.997 slide individually. >> Yeah. Your eyes kind of glaze over. 01:30:08.080 --> 01:30:10.877 >> Yeah. >> And and you just sort of like lose focus, stop paying attention. 01:30:10.960 --> 01:30:13.277 >> Yeah. It's like it would have been more it would have been more compelling to 01:30:13.360 --> 01:30:16.557 actually just have a bulleted list of like problem. 01:30:16.640 --> 01:30:20.077 >> I mean a lot of times you can just send me the prompt because I can instantiate 01:30:20.160 --> 01:30:23.917 in my head. I can imagine the rest of the paragraphs. I I have the context 01:30:24.000 --> 01:30:26.557 window preloaded. Yeah. uh for for 01:30:26.640 --> 01:30:28.877 myself. Yeah. Uh we should talk about 01:30:28.960 --> 01:30:33.757 the new Audi, the Nuvo Lari. Is this real? Motor one. This seems real. It's 01:30:33.840 --> 01:30:36.317 real. >> Uh it's big deal. It's the It's the 01:30:36.400 --> 01:30:38.717 brand's first supercar since the R8. 01:30:38.800 --> 01:30:43.517 Twinturbocharged 4 liter V8 hybrid. 217 01:30:43.600 --> 01:30:47.517 mph top speed. That is 10% faster than a 01:30:47.600 --> 01:30:51.597 Cayenne Turbo GT. What is the Cayenne Turbo GT market doing right now? Is it 01:30:51.680 --> 01:30:55.357 tanking? Depreciation must be just through the roof on this news because 01:30:55.440 --> 01:30:59.997 you have a car that's 10% faster and so why every everyone is gonna be rotating 01:31:00.080 --> 01:31:05.357 out. >> I mean I think they did I I think the new Ari design. It's a really cool 01:31:05.440 --> 01:31:07.597 design. >> Feels like somewhat cybert truck. 01:31:07.680 --> 01:31:10.317 >> Cyber punky futuristic. I don't know. It 01:31:10.400 --> 01:31:15.197 just checks the box for like the next supercar for me. And uh in a way that the 01:31:15.280 --> 01:31:17.037 >> Ben says it can't touch the R8. 01:31:17.120 --> 01:31:19.677 >> Oh, can't touch the R8. Okay. Okay. 01:31:19.760 --> 01:31:22.237 Well, it goes 0 to 60 in 2.6 seconds. Uh well 01:31:22.320 --> 01:31:24.477 >> almost 1,000 horsepower. Let me tell you 01:31:24.560 --> 01:31:28.797 about Cisco. Critical infrastructure for the AI era. Unlock seamless realtime 01:31:28.880 --> 01:31:31.037 experiences and new value with Cisco. 01:31:31.120 --> 01:31:34.717 And our next guest Sam Barry is here from the USDA. Welcome to the show. 01:31:34.800 --> 01:31:35.997 >> How are you doing? >> Very good. 01:31:36.080 --> 01:31:39.837 >> Good to meet you. Thank you so much for coming on down. Let's throw this on. 01:31:39.920 --> 01:31:42.397 >> And just like that. >> On the left side. 01:31:42.480 --> 01:31:45.597 >> So >> good. >> Introduce yourself a little. Tell us about yourself. 01:31:45.680 --> 01:31:50.477 >> All right. Yeah. My name is Sam Barry. I am uh proud to be working at the USDA. 01:31:50.560 --> 01:31:52.477 Would you do that right now? I'm the chief. 01:31:52.560 --> 01:31:56.717 >> Nominative determinism. [laughter] Do you know about nominative determinism? 01:31:56.800 --> 01:31:59.517 >> No. >> It's the idea that you you know a 01:31:59.600 --> 01:32:01.837 person's name could possibly influence 01:32:01.920 --> 01:32:05.677 or or uh the but but Barry and working 01:32:05.760 --> 01:32:09.197 at the Department of Agriculture is like pretty perfect. No, it's incredible. 01:32:09.280 --> 01:32:14.317 Actually, my uh the berries came over here from France in um >> like 1640. 01:32:14.400 --> 01:32:16.317 >> Wow. >> So, we've been here for a long time. 01:32:16.400 --> 01:32:18.157 >> That's crazy. >> And uh it was all farmers. 01:32:18.240 --> 01:32:21.597 >> Yeah. >> Yeah. I was like all farmers until my 01:32:21.680 --> 01:32:26.797 grandpa then he became a materials engineer actually worked on jet engines. 01:32:26.880 --> 01:32:30.077 >> Okay. >> And so then his sons became engineers. 01:32:30.160 --> 01:32:32.797 My dad became an engineer. Then I was an engineer. So we're kind of >> okay 01:32:32.880 --> 01:32:33.837 >> trying to bring the two together. 01:32:33.920 --> 01:32:35.197 >> Now you're going back to the USDA. 01:32:35.280 --> 01:32:38.717 >> Yeah. >> What what is the shape of the USDA? Like what what is the shape of the 01:32:38.800 --> 01:32:43.597 organization headquarters? Do you go to the office? Is this you know US? You 01:32:43.680 --> 01:32:47.197 think just America international footprint? Like do you travel for work? 01:32:47.280 --> 01:32:49.917 What's it like working there? Well, actually, it'd be kind of interesting to 01:32:50.000 --> 01:32:53.437 ask you what you think, like what are the things that you think USDA does? 01:32:53.520 --> 01:32:55.997 >> Uh, they grade the milk in the states. 01:32:56.080 --> 01:32:57.357 That's what I think about it. So, I 01:32:57.440 --> 01:33:00.157 imagine that at some point farmers send 01:33:00.240 --> 01:33:04.877 the cows to you and you kind of inspect them and say, "This is a good cow." Is 01:33:04.960 --> 01:33:06.477 that what happens? >> I don't know. 01:33:06.560 --> 01:33:10.397 >> Um, there's like inspectors. There's a whole area that does that. 01:33:10.480 --> 01:33:12.077 >> There's like a series of certifications, 01:33:12.160 --> 01:33:15.437 but but but is but what else is happening? 01:33:15.520 --> 01:33:17.997 >> So, all kinds of stuff. Do you know that like food stamps? 01:33:18.080 --> 01:33:19.917 >> Yeah, >> SNAP is inside of USDA. 01:33:20.000 --> 01:33:21.437 >> Oh, I didn't know that. >> I didn't know that either. I thought I 01:33:21.520 --> 01:33:25.517 figured it was in like HHS or something, but uh yeah, it's in USDA. 01:33:25.600 --> 01:33:28.157 >> So, that's hundred billion dollars a year. It's kind of a big deal. 01:33:28.240 --> 01:33:31.437 >> Yeah. >> Um so, we do we have SNAP that's in the 01:33:31.520 --> 01:33:35.437 food nutrition service. >> Uh Forest Service is inside of USDA. 01:33:35.520 --> 01:33:38.717 Like crazy. >> Yeah. Yeah. Uh, and then FPAC is like 01:33:38.800 --> 01:33:43.037 what you would really think that USDA it's like the farmer facing like okay 01:33:43.120 --> 01:33:46.077 >> where farm programs are where they do acreage reporting like the stuff I 01:33:46.160 --> 01:33:47.197 talked about today. >> Got it. 01:33:47.280 --> 01:33:49.037 >> Um, then there's rural development. 01:33:49.120 --> 01:33:53.757 >> Okay. >> Which is like loans like a bank basically do loans for all kinds of 01:33:53.840 --> 01:33:59.117 things. >> Okay. >> Um, actually in some of the reviews I came in on Doge and uh 01:33:59.200 --> 01:34:02.077 >> uh there's like beachfront hotels that are being funded out of RD. So there's 01:34:02.160 --> 01:34:03.517 like a lot of things that need to be cleaned up. 01:34:03.600 --> 01:34:06.317 >> Okay. >> Um, [laughter] >> yeah. And then there's like food 01:34:06.400 --> 01:34:10.957 inspection service. And then there's actually a huge scientific arm. 01:34:11.040 --> 01:34:14.797 >> Yeah. >> That's inside of >> That makes sense. Testing things and 01:34:14.880 --> 01:34:17.597 >> Yeah. Like labs, >> advancing different pesticides. 01:34:17.680 --> 01:34:21.917 >> So things that I mean actually become very passionate about it because I 01:34:22.000 --> 01:34:24.717 certainly didn't have an appreciation for I thought the same thing. It's like 01:34:24.800 --> 01:34:28.157 grading meat. Yeah. You know, >> um 01:34:28.240 --> 01:34:32.877 >> but like our we are so uniquely positioned as a country because of the 01:34:32.960 --> 01:34:34.157 fact that we can feed ourselves. 01:34:34.240 --> 01:34:38.237 >> Yeah. Yeah. >> And like that is not the case for a lot of a lot of countries. 01:34:38.320 --> 01:34:41.677 >> Isn't America basically a net exporter of food too? You hear about this in the 01:34:41.760 --> 01:34:46.717 China debate all the time. Will they buy XYZ product from us as a retaliation? 01:34:46.800 --> 01:34:49.757 And uh yeah, you just don't think about it. But uh 01:34:49.840 --> 01:34:53.437 >> yeah. So like China can like minimally feed itself like bare minimum it could 01:34:53.520 --> 01:34:57.357 like keep itself alive. >> Um but you know they're getting like 01:34:57.440 --> 01:35:01.037 like we just did a big deal with them to move a bunch of beef over there. 01:35:01.120 --> 01:35:03.437 >> Yeah. >> And kind of got some negative press 01:35:03.520 --> 01:35:06.397 >> on that. So, it's important to know it's uh I forget exactly what it's called, 01:35:06.480 --> 01:35:09.837 but it's like the parts of the cow that we don't eat here. So, it's a little 01:35:09.920 --> 01:35:11.917 misleading to say like the amount that we're sending over there. 01:35:12.000 --> 01:35:14.717 >> Also, all these trade deals are like very complex and there's like six 01:35:14.800 --> 01:35:19.757 different moving parts. We get batteries or they get the chips and like these are 01:35:19.840 --> 01:35:22.237 always like, you know, seven part negotiations. It's hard to look at 01:35:22.320 --> 01:35:24.397 anyone in isolation. >> But, I mean, I think it's a little 01:35:24.480 --> 01:35:27.957 surprising that aggro like food is actually part of that. I mean and then 01:35:28.040 --> 01:35:33.357 [clears throat] in warfare uh like agriculture the food supply is usually 01:35:33.440 --> 01:35:36.637 hit before anything like kinetic even happens you know and then before even 01:35:36.720 --> 01:35:39.277 the world knows that it's warfare >> you know okay 01:35:39.360 --> 01:35:44.157 >> uh because you can do that and you can do things to you know impact a nation's 01:35:44.240 --> 01:35:47.197 food supply in the future and so agriculture is like a really big deal 01:35:47.280 --> 01:35:49.757 >> sure >> um really important so all this to tie 01:35:49.840 --> 01:35:53.357 back to I wanted to talk about the labs because this is like a whole area inside 01:35:53.440 --> 01:35:58.237 of uh USDA but uh we do all of these these things like invest in figuring 01:35:58.320 --> 01:36:01.517 out. So like personally I try and avoid like GMOs and we eat, you know, like we 01:36:01.600 --> 01:36:04.717 drink raw milk and we get our meat from a local farm. 01:36:04.800 --> 01:36:06.877 >> Um but GMOs are actually really important. Yeah. 01:36:06.960 --> 01:36:09.277 >> Because if we were hit with some kind of 01:36:09.360 --> 01:36:13.277 adverse event or something and we needed to create corn that would could survive 01:36:13.360 --> 01:36:17.037 a drought better like we have >> the science and the research to be able 01:36:17.120 --> 01:36:21.197 to do that. Got it. >> And it's a huge edge that we have >> like geopolitically. 01:36:21.280 --> 01:36:25.437 >> Interesting. Yeah. >> Yeah. talk about uh over the years I've 01:36:25.520 --> 01:36:27.997 read so many stories of you know this uh 01:36:28.080 --> 01:36:32.557 this insect has been detected in you know some region of the US and there's 01:36:32.640 --> 01:36:34.957 speculation on is it um you know kind of 01:36:35.040 --> 01:36:37.037 foreign interference things like that is 01:36:37.120 --> 01:36:42.157 that is that in USDA domain is trying to help monitor and track and make sure 01:36:42.240 --> 01:36:46.157 that >> um pests yeah pests pests like pests are 01:36:46.240 --> 01:36:48.397 obviously naturally occurring right they 01:36:48.480 --> 01:36:50.877 flourish for their own reasons or there 01:36:50.960 --> 01:36:54.877 can be um some uh some sort of malicious 01:36:54.960 --> 01:36:58.477 intent as well. Is that your guys' >> Yeah, cuz they're not ne necessarily 01:36:58.560 --> 01:37:01.357 naturally occurring, right? >> And uh so one that we have going on 01:37:01.440 --> 01:37:04.637 right now, and I'm not saying this one's not naturally occurring, but the uh new 01:37:04.720 --> 01:37:09.437 world screw screw worm that's coming up through Mexico. So our secretary, which 01:37:09.520 --> 01:37:12.317 by the way, I couldn't say enough good things about Secretary Rollins. I mean, 01:37:12.400 --> 01:37:14.237 she's incredible. Just an actual like 01:37:14.320 --> 01:37:19.517 genuine good and like it's unbelievable what she's able to accomplish. Uh, but 01:37:19.600 --> 01:37:23.757 New World's screworm is something that's falling in USDA's uh, you know, 01:37:23.840 --> 01:37:27.117 responsibilities. And this is like a parasite basically that's coming up 01:37:27.200 --> 01:37:29.757 through Mexico and it's like a flesh eating parasite. So it's like really 01:37:29.840 --> 01:37:34.157 hardcore. Yeah. >> So we're developing a lab >> flesh. 01:37:34.240 --> 01:37:37.517 No sorts. No, but you know, I don't think you want to be around it. But no, 01:37:37.600 --> 01:37:42.077 it's for like cattle mostly is what it impacts. Um, and so we're developing a 01:37:42.160 --> 01:37:45.677 lab and like sterilizing flies, which again, like personally, I don't really 01:37:45.760 --> 01:37:49.677 like any of this stuff, but it's better to be doing this and be able to protect 01:37:49.760 --> 01:37:53.117 our nation than like if we let this just come and flourish in our country, I 01:37:53.200 --> 01:37:54.397 mean, it' be very detrimental. 01:37:54.480 --> 01:37:57.437 >> Yeah. >> So, I'll go back and and if and if it's 01:37:57.520 --> 01:38:01.677 a necessary, you know, technique that needs to be harnessed, it needs to be 01:38:01.760 --> 01:38:05.037 harnessed securely and it needs to be harnessed with uh, you know, the right 01:38:05.120 --> 01:38:07.837 teams in place to make sure that whatever's rolled out is rolled out 01:38:07.920 --> 01:38:09.437 effectively and safely, right? 01:38:09.520 --> 01:38:13.277 >> Yeah. I mean, I think it's just so important. You know, there's uh like 01:38:13.360 --> 01:38:17.437 techn there's so much farther we can go with technology, but we have so much 01:38:17.520 --> 01:38:20.717 right now. >> And uh so many people are just black 01:38:20.800 --> 01:38:22.797 pill, right? And I think it's important. 01:38:22.880 --> 01:38:25.117 I think you should be like black pill on certain things, but you should probably 01:38:25.200 --> 01:38:27.117 take a lot of pills. Like you should be red pill and black pill [laughter] and 01:38:27.200 --> 01:38:30.237 white pill at the same time. >> Uh because take 01:38:30.320 --> 01:38:34.077 >> like we have a long way to go and when we're just like 01:38:34.160 --> 01:38:37.117 >> sitting feeling sorry for ourselves, like it's not a good position to be in. 01:38:37.200 --> 01:38:40.237 This is the most incredible country on earth 01:38:40.320 --> 01:38:44.317 >> and other countries are advancing though. You know, our edge is like our 01:38:44.400 --> 01:38:46.237 edge doesn't come for free. >> No, we got to work at it. 01:38:46.320 --> 01:38:47.837 >> We got to keep pushing at these things. 01:38:47.920 --> 01:38:51.437 But when we do this, like when there's a parasite that's, you know, coming into 01:38:51.520 --> 01:38:54.637 our country and we're able to just like use biology 01:38:54.720 --> 01:38:56.877 >> to combat it. Like that's incredible 01:38:56.960 --> 01:38:58.077 that our country can do that. 01:38:58.160 --> 01:39:02.317 >> Talk about uh these more SMB scale 01:39:02.400 --> 01:39:04.397 farmers and their approach to 01:39:04.480 --> 01:39:06.957 technology. Uh I think a lot of people 01:39:07.040 --> 01:39:11.517 would be surprised at how much uh how much these individuals at least from 01:39:11.600 --> 01:39:13.357 from what I've experienced are happy to 01:39:13.440 --> 01:39:16.077 lean into technology. I met uh a group 01:39:16.160 --> 01:39:19.117 in Texas that had developed this was 01:39:19.200 --> 01:39:21.837 years ago so pre-ai boom developed their 01:39:21.920 --> 01:39:24.077 own SAS product to help manage their 01:39:24.160 --> 01:39:26.957 operations like a a a tool that that 01:39:27.040 --> 01:39:29.197 they had built by discovering problems 01:39:29.280 --> 01:39:33.037 that they had uh on their property. And 01:39:33.120 --> 01:39:37.117 I just thought that was um that was really fascinating and and cool at the 01:39:37.200 --> 01:39:38.397 time because I think Silicon Valley 01:39:38.480 --> 01:39:41.517 would have maybe some expectation uh 01:39:41.600 --> 01:39:46.717 that uh that there might be an aversion to that until you get into the more like 01:39:46.800 --> 01:39:51.037 enterprisegrade uh scale. >> Yeah, I think it's a really important 01:39:51.120 --> 01:39:54.077 topic because you're essentially talking about like democratizing access to 01:39:54.160 --> 01:39:56.797 technology, right? And certainly with uh 01:39:56.880 --> 01:40:01.357 with like AI becoming so much more widely available, uh that was a big step 01:40:01.440 --> 01:40:04.797 forward. But I mean, and this is a big point that's being hit on at this 01:40:04.880 --> 01:40:06.717 conference and what Palanteer is really 01:40:06.800 --> 01:40:12.237 focusing on is those LLMs become useless if they're not uh if you're not 01:40:12.320 --> 01:40:16.077 deploying them in the right way with the right like data boundaries, right? So, 01:40:16.160 --> 01:40:18.477 um you know, I think that's something that we're seeing even in our 01:40:18.560 --> 01:40:22.877 universities. We do a lot of university research and uh like all the you know 01:40:22.960 --> 01:40:26.157 kids or whatever the the university students like they're wanting to do 01:40:26.240 --> 01:40:29.117 experiments with LLMs and do like meat grading like better meat grading because 01:40:29.200 --> 01:40:32.317 that's something that can happen at the farms and uh if you can make that 01:40:32.400 --> 01:40:34.477 automated then you know our ability to 01:40:34.560 --> 01:40:36.957 produce beef uh you know is is greatly 01:40:37.040 --> 01:40:41.597 impacted. Um, but there's a major issue in secession planning right now for 01:40:41.680 --> 01:40:44.877 farms, right? Like this is a big thing that's happening. Like the farmer 01:40:44.960 --> 01:40:48.477 generation is getting very old and kids don't want to go and and run the farm. 01:40:48.560 --> 01:40:51.997 >> A lot of them went to big cities and jobs and white color work and stuff. 01:40:52.080 --> 01:40:56.237 >> So, uh, you know, this is a big thing that is a H2A. Yeah. You know, these H2A 01:40:56.320 --> 01:40:58.717 visas where like a lot of the farmers are actually still saying like we need 01:40:58.800 --> 01:41:01.597 the help from, you know, we need immigrants to come and help us 01:41:01.680 --> 01:41:04.797 >> and, uh, you know, the best way that we can solve that is through automation. 01:41:04.880 --> 01:41:07.997 So, I think that >> that's something I would love to see 01:41:08.080 --> 01:41:12.157 USDA do more of or, you know, uh >> it's something that needs to be 01:41:12.240 --> 01:41:14.957 answered. I don't have an answer for you right now. Yeah. But in order for us to 01:41:15.040 --> 01:41:18.397 continue to, you know, remain self-sufficient and providing food. 01:41:18.480 --> 01:41:21.117 >> Yeah. Whenever you have a dwindling workforce, increasing the leverage and 01:41:21.200 --> 01:41:25.677 productivity of the existing workforce allows you to maintain overall aggregate 01:41:25.760 --> 01:41:29.117 productivity. This is Yeah. general technological leverage. So, it makes a 01:41:29.200 --> 01:41:31.517 ton of sense. >> Do you know anybody that's becoming a farmer? 01:41:31.600 --> 01:41:33.357 >> Uh, well, we know some folks. We've had 01:41:33.440 --> 01:41:35.277 a number of uh entrepreneurs on the show 01:41:35.360 --> 01:41:38.797 who are getting into a tech building. Uh 01:41:38.880 --> 01:41:43.037 we've had the founder of the laser weeder that uh uses uh a lot of people 01:41:43.120 --> 01:41:46.877 don't like pesticides, but they don't mind if uh if a pest is zapped with a 01:41:46.960 --> 01:41:51.117 laser because that's just heat that's being transferred to the particular 01:41:51.200 --> 01:41:54.477 plant right there and the tomato plant continues flourishing. So it uses just 01:41:54.560 --> 01:41:59.117 cameras and lasers. Very cool sort of modern solution to something that people 01:41:59.200 --> 01:42:02.797 have had a lot of beer around about around different pesticides. fruit 01:42:02.880 --> 01:42:07.277 picking robotics company. >> Yeah, Orchard as well. Uh but but mostly 01:42:07.360 --> 01:42:11.677 from tech side, usually with some family lineage uh sort of returning to the 01:42:11.760 --> 01:42:16.237 roots or or tapping into their networks uh to go back, but uh I mean truthfully, 01:42:16.320 --> 01:42:19.757 I don't know that many people that I grew up with. I mean, I grew up in LA, 01:42:19.840 --> 01:42:23.517 so not much farming activity. I know one family that had an avocado farm. 01:42:23.600 --> 01:42:26.557 >> I mean, it actually it would be super based to be a large scale farmer. Like 01:42:26.640 --> 01:42:29.437 more people should do it. And um >> we need a maybe you could be the Alex 01:42:29.520 --> 01:42:32.637 Hormosi of farming. >> Yeah. No, for real. I mean, you can. So, 01:42:32.720 --> 01:42:35.437 USDA, one of the great things that USDA does is you can get financial 01:42:35.520 --> 01:42:39.037 assistance. Like, you get big time like big time loans from USDA. You have to go 01:42:39.120 --> 01:42:41.917 through the process and they're actually doing a loan modernization effort right 01:42:42.000 --> 01:42:45.277 now to try and make that better. But >> like USDA, we'll fund it for you. You 01:42:45.360 --> 01:42:49.517 got to pay it back, but you can like get the interest rates low rate subsidized. 01:42:49.600 --> 01:42:53.517 Yeah. >> Yeah. I mean, one of our uh administrators at USDA, he like pulled 01:42:53.600 --> 01:42:56.157 up his phone one day and he's like, "Look, it's a planting day for me." And 01:42:56.240 --> 01:42:59.517 it was his John Deere app. It's like the most advanced, you know, like 01:42:59.600 --> 01:43:02.637 >> uh he had all these tractors going and there's still people sitting in the 01:43:02.720 --> 01:43:05.837 tractors, but it's to the point where it basically could be fully automated. 01:43:05.920 --> 01:43:08.957 >> So, I mean, you can get yourself a couple thousand acres and just start, 01:43:09.040 --> 01:43:13.197 you know, growing corn or wheat or cotton like cotton and then, you know, whatever. 01:43:13.280 --> 01:43:17.677 >> Uh talk about data collection. I feel like data is the lifeblood of, uh, you 01:43:17.760 --> 01:43:21.597 know, any decision- making, any udaloop, anything related to Palunteer, USDA. And 01:43:21.680 --> 01:43:23.037 I'm wondering about like you mentioned 01:43:23.120 --> 01:43:25.757 that screw worm. You got to track that 01:43:25.840 --> 01:43:28.637 thing. it shows up on some cattle 01:43:28.720 --> 01:43:32.397 rancher's farm and they're detecting it 01:43:32.480 --> 01:43:35.037 or they're seeing symptoms, maybe they 01:43:35.120 --> 01:43:39.437 know roughly what percentage of the herd is affected, but how do they actually 01:43:39.520 --> 01:43:40.797 get that information to you? Are they 01:43:40.880 --> 01:43:43.757 going to usda.govreport 01:43:43.840 --> 01:43:46.637 incident or are you pulling things from 01:43:46.720 --> 01:43:49.197 their filings? Like h how how do you 01:43:49.280 --> 01:43:51.997 want that to evolve? I imagine that that 01:43:52.080 --> 01:43:56.877 with more AI and technology we're it's only as good as the data that we can 01:43:56.960 --> 01:44:00.477 actually put into the system. So just broadly data collection where is that 01:44:00.560 --> 01:44:02.637 going these days? >> Well if you don't mind instead of screw 01:44:02.720 --> 01:44:04.797 I'd like to focus on SNAP for that question. 01:44:04.880 --> 01:44:08.957 >> So SNAP is funded by the federal government but it's administered by the 01:44:09.040 --> 01:44:12.797 states. >> Okay. So uh when it comes to so something that we're doing right now and 01:44:12.880 --> 01:44:15.597 it was one of the first things that our secretary did like on our first day was 01:44:15.680 --> 01:44:18.557 she did a data call to all the states that you know we want all of your snap 01:44:18.640 --> 01:44:21.917 data to understand how because it's our responsibility as the funer of this 01:44:22.000 --> 01:44:25.357 program >> to understand [snorts] the integ like to verify the integrity 01:44:25.440 --> 01:44:27.997 of the program. >> So we put a request out there but it has 01:44:28.080 --> 01:44:29.117 to come from every single state. 01:44:29.200 --> 01:44:31.277 >> Yeah. >> And a lot of the state programs they're 01:44:31.360 --> 01:44:34.157 not technical or they've got contractors that you know it's just a difficult 01:44:34.240 --> 01:44:37.037 thing to get us the data. Uh but then there's also a bunch of states that are 01:44:37.120 --> 01:44:39.597 just not complying, you know, for whatever reason, which it shouldn't be a 01:44:39.680 --> 01:44:41.277 problem. I don't understand what the problem is. 01:44:41.360 --> 01:44:46.557 >> Um >> but the importance of so that program that's a hundred billion taxpayer 01:44:46.640 --> 01:44:50.237 dollars a year. Like that's pretty substantial. Um that's an area where we 01:44:50.320 --> 01:44:52.717 really want to have all angles of the 01:44:52.800 --> 01:44:57.277 data available so that we can deploy AI and become really smart in detecting 01:44:57.360 --> 01:45:00.957 fraud. Like we want to get it to the point where um if somebody's committing 01:45:01.040 --> 01:45:04.477 snap fraud, we should be able it's like your card, right? If you if if somebody 01:45:04.560 --> 01:45:07.837 stole your card and did a transaction that wasn't recognized, like your card's 01:45:07.920 --> 01:45:10.957 shut off. Yep. Right. Yep. >> So, we want to get to the point where 01:45:11.040 --> 01:45:14.317 we're very intelligent and we're confident enough in the system that we 01:45:14.400 --> 01:45:17.837 can do that. When there's fraud detected, it's off immediately. 01:45:17.920 --> 01:45:19.357 >> Uh because it's an important program. 01:45:19.440 --> 01:45:22.637 You know, we want to be able to support people that can't support themselves, 01:45:22.720 --> 01:45:26.477 but it's uh it's not arguable that there's a massive amount of fraud in 01:45:26.560 --> 01:45:30.637 there. I mean, even the uh the organization itself does like an audit 01:45:30.720 --> 01:45:35.357 every year and they're at like there's 12% improper payments. Um improper 01:45:35.440 --> 01:45:36.477 payments is kind of a bad word. 01:45:36.560 --> 01:45:40.397 >> So 12 12 billion a year. >> Yeah. Right. And that's just like kind 01:45:40.480 --> 01:45:42.957 of based on >> that's money that that's that could 01:45:43.040 --> 01:45:45.277 actually be going towards the intent of 01:45:45.360 --> 01:45:50.877 the program, which is to provide >> food to people that otherwise wouldn't 01:45:50.960 --> 01:45:53.837 be able to get it. And there's other, you know, you could like grock how SNAP 01:45:53.920 --> 01:45:58.157 has been used to fund uh like international crime organizations and 01:45:58.240 --> 01:46:02.397 like terrorist groups and everything. So it's it's being exploited at a at a huge 01:46:02.480 --> 01:46:05.677 level. And uh I mean it's something that our secretary has prioritized, but 01:46:05.760 --> 01:46:08.157 that's probably our biggest fpack. What 01:46:08.240 --> 01:46:10.557 I talked about today is like our most 01:46:10.640 --> 01:46:14.637 complex system of data, but the SNAP 01:46:14.720 --> 01:46:19.357 challenge is uh like the biggest or like this the snap environment is probably 01:46:19.440 --> 01:46:20.957 the biggest challenge on the data front. 01:46:21.040 --> 01:46:26.637 Mhm. >> What's next for you? You making a career out of this or you going to go be a 01:46:26.720 --> 01:46:31.037 farmer? >> Hopefully both. Okay. Yeah. Yeah. I mean, um 01:46:31.120 --> 01:46:35.037 >> yeah, I've got some farmland. Um trying to convert it in It's like woods right 01:46:35.120 --> 01:46:37.917 now, but um >> uh where's that? 01:46:38.000 --> 01:46:39.997 >> Uh in Virginia, so actually when I lived 01:46:40.080 --> 01:46:43.837 in Michigan, we had like a little bit of a farm. We had some goats and sheep and 01:46:43.920 --> 01:46:47.597 a bunch of chickens and and ducks. Um you don't ever want to get you don't 01:46:47.680 --> 01:46:51.437 want to get ducks. You don't want to get goats. Um, ducks are like really savage 01:46:51.520 --> 01:46:55.437 actually. Yeah, like a chicken sleeps, you know? So, like it's got a normal 01:46:55.520 --> 01:46:58.157 cycle like at night time it goes into the coupe and it like sleeps. 01:46:58.240 --> 01:47:01.677 >> Ducks don't sleep. >> No, ducks do not sleep. No, they like in 01:47:01.760 --> 01:47:05.117 our house was kind of this like really unique house. So, the windows were like 01:47:05.200 --> 01:47:08.817 on the ground and the ducks would come and just stare at us in the window. 01:47:08.900 --> 01:47:11.837 [laughter] No, they're savage. They just like they sleep for like 10 minutes at a 01:47:11.920 --> 01:47:14.717 time. So, they'll just like waddle around and then sleep for 10 minutes. 01:47:14.800 --> 01:47:17.997 You have to have the right balance of female and male ducks. Otherwise, it's 01:47:18.080 --> 01:47:21.117 like really ugly. >> Yeah, chickens are a lot. I grew I grew 01:47:21.200 --> 01:47:25.677 up with chickens and uh most of the time they're they're cool. My dad would build 01:47:25.760 --> 01:47:27.997 >> these sort of like complex contraptions 01:47:28.080 --> 01:47:30.957 to automate the opening enclosure. So, 01:47:31.040 --> 01:47:33.837 he would use like irrigation to 01:47:33.920 --> 01:47:36.237 >> uh on a timer to fill a bucket which 01:47:36.320 --> 01:47:38.557 would lift there would lift it up. Yeah. 01:47:38.640 --> 01:47:41.117 >> Okay. Interesting. Um but but then I 01:47:41.200 --> 01:47:44.317 still core memories as a kid was waking 01:47:44.400 --> 01:47:48.717 up my dad would yell like there's a fox in the coupe and then we'd be like 01:47:48.800 --> 01:47:52.637 running out. We'd be it would be like game on. Yeah. Yeah. Or you get like 01:47:52.720 --> 01:47:54.957 skunks [clears throat] in there and coyotes maybe. 01:47:55.040 --> 01:47:58.717 >> Yeah. We would just everybody would get up and [laughter] try to go 01:47:58.800 --> 01:48:01.197 >> deal with be satisfying. That's way more 01:48:01.280 --> 01:48:02.877 satisfying than some software bug. 01:48:02.960 --> 01:48:06.157 there's so much uh fear and doom and and 01:48:06.240 --> 01:48:09.837 blackpilling uh around data centers. Uh 01:48:09.920 --> 01:48:13.997 I wanted to hear from you how you're I 01:48:14.080 --> 01:48:16.077 imagine your your role is to be an 01:48:16.160 --> 01:48:19.597 advocate for for farmers as well on on 01:48:19.680 --> 01:48:22.477 on water supplies, things like that. 01:48:22.560 --> 01:48:25.757 California went through, you know, probably 01:48:25.840 --> 01:48:28.157 >> many many really rough years from a from 01:48:28.240 --> 01:48:32.957 a water supply um and a water scarcity standpoint. Thankfully, you know, had a 01:48:33.040 --> 01:48:34.957 lot of rains over the last few years, 01:48:35.040 --> 01:48:38.157 but how how are you working with farmers 01:48:38.240 --> 01:48:41.357 or or what is the situation around um 01:48:41.440 --> 01:48:43.997 the the the kind of like tension between 01:48:44.080 --> 01:48:46.397 a lot of farmland could also be great 01:48:46.480 --> 01:48:51.597 land for data centers, right? And there's been some um uh pretty 01:48:51.680 --> 01:48:54.957 high-profile stories where farmers either sold their land, but from your 01:48:55.040 --> 01:48:58.237 side, you're trying to make sure that we have, you know, can produce an abundance 01:48:58.320 --> 01:49:01.197 of food, you know, from a from a national security standpoint. or how are 01:49:01.280 --> 01:49:02.557 you guys thinking about that balance? 01:49:02.640 --> 01:49:05.357 >> Yeah, I mean I think the best solution is putting the data centers in space, 01:49:05.440 --> 01:49:09.117 you know, like which is totally led by Elon and people are jumping on that 01:49:09.200 --> 01:49:11.837 train, but it's going to be a couple years it sounds like before they're to 01:49:11.920 --> 01:49:16.077 that point. We're actually uh USDA is pursuing a partnership with SpaceX. 01:49:16.160 --> 01:49:20.077 >> Um and that that part isn't isn't ready yet. We don't really have a need for 01:49:20.160 --> 01:49:23.357 that, but it's there's a partnership on the technical side, but there's also 01:49:23.440 --> 01:49:26.797 just on the like conceptual side of the fact that like we're align because we do 01:49:26.880 --> 01:49:30.557 care about conservation. um you know there was a lot of green stuff that was 01:49:30.640 --> 01:49:33.437 like you know sure not stuff that we care about but we do care about 01:49:33.520 --> 01:49:36.957 conserving our land and uh putting data centers in space just makes a ton of 01:49:37.040 --> 01:49:40.237 sense but that being a couple years out so for today you know I'm actually 01:49:40.320 --> 01:49:44.077 pretty passionate about this because in my hometown of Selen Michigan it's like 01:49:44.160 --> 01:49:46.797 small town mostly farmland they're 01:49:46.880 --> 01:49:48.717 putting a data center in there 01:49:48.800 --> 01:49:51.437 >> and it's like you know 30 miles from 01:49:51.520 --> 01:49:54.157 Detroit and Flint and like all these 01:49:54.240 --> 01:49:59.037 very industrialized areas and so it's very confusing to me why we wouldn't be 01:49:59.120 --> 01:50:02.397 putting these data center and they're like struggling areas. Detroit's doing 01:50:02.480 --> 01:50:04.637 all right, but like Flint struggling big time. 01:50:04.720 --> 01:50:08.317 >> Uh like why not put a data center there where there's already the infrastructure 01:50:08.400 --> 01:50:11.357 there's like >> it's already developed land, but instead 01:50:11.440 --> 01:50:13.757 it's like taking these small townships 01:50:13.840 --> 01:50:18.397 and and plopping them in the middle. Um and the people don't really like it. Now 01:50:18.480 --> 01:50:22.237 the boards seem to like it for some reason, the councils. So I don't know 01:50:22.320 --> 01:50:24.077 what's up with that. Um but it doesn't 01:50:24.160 --> 01:50:28.237 align. It creates it creates a massive 01:50:28.320 --> 01:50:32.077 amount of tax revenue that can be used to fund a bunch of other programs that 01:50:32.160 --> 01:50:36.157 >> but it's got to actually flow back to the people who are in the town and I 01:50:36.240 --> 01:50:39.837 think that there's like a disconnect there sometimes. I actually this is kind 01:50:39.920 --> 01:50:43.597 of outside but something that I do think is uh probably going to happen is you 01:50:43.680 --> 01:50:47.117 know there was this big shift to go to the cloud right it's like everybody kind 01:50:47.200 --> 01:50:50.157 of had their own servers you know it's on prem and now we're in the cloud and 01:50:50.240 --> 01:50:53.997 it's like really you just took you like moved it across the street right 01:50:54.080 --> 01:50:56.077 >> um and now that people are becoming more 01:50:56.160 --> 01:51:00.317 aware of like what that means and when it's like oh my data is in AWS or you 01:51:00.400 --> 01:51:03.357 know it's like and maybe this is a global company and how much can I really 01:51:03.440 --> 01:51:07.917 trust this company that there's going to be a shift back to caring actually 01:51:08.000 --> 01:51:11.037 actually caring about where your data is living. Interesting. So I think a good a 01:51:11.120 --> 01:51:14.877 b business opportunity would be I I think there's a world where there's a 01:51:14.960 --> 01:51:18.397 culture that comes up around data centers because like me personally like 01:51:18.480 --> 01:51:22.877 I want to build like my house is like uh like I'll have a kill switch for my 01:51:22.960 --> 01:51:25.517 Wi-Fi and then like we've got the data in the basement. 01:51:25.600 --> 01:51:27.037 >> You got your raw milk supply. 01:51:27.120 --> 01:51:30.157 >> Got raw milk. No, like we're ready to go. I mean we're I I was ready to go off 01:51:30.240 --> 01:51:32.317 the grid before I came and joined the government. [laughter] This is a much 01:51:32.400 --> 01:51:35.597 better option. But um still like I care about my data. I don't really want to 01:51:35.680 --> 01:51:38.797 use YouTube music anymore for my music cuz now my recommendations are getting 01:51:38.880 --> 01:51:42.157 worse and you're like very beholden to that. It's like I could very easily just 01:51:42.240 --> 01:51:45.197 >> have the music >> buy my music and write a simple program 01:51:45.280 --> 01:51:48.317 to like make my recommendations and it would be way better because there's 01:51:48.400 --> 01:51:51.357 certain artists that are not getting recommended because they're not, you 01:51:51.440 --> 01:51:53.597 know, prioritized behind the scenes >> or something. 01:51:53.680 --> 01:51:56.797 >> So, but not everybody's going to want to manage their own servers, right? 01:51:56.880 --> 01:51:59.837 >> Jensen just announced a data center that bolts onto the side of your house. 01:51:59.920 --> 01:52:02.477 >> Oh, that's sweet. >> And uh I mean and there's more stuff 01:52:02.560 --> 01:52:05.437 that's coming that way. I mean, people are doing with the Mac minis. Can't 01:52:05.520 --> 01:52:07.277 really do the Frontier AI on the Mac 01:52:07.360 --> 01:52:09.517 Mini just now, but in a few years, you 01:52:09.600 --> 01:52:14.477 know, the DJX uh desktops, like it's all coming and I think it will be more of an 01:52:14.560 --> 01:52:16.637 option. >> So, just to like kind of wrap this up, 01:52:16.720 --> 01:52:18.957 so there's this um uh or this like 01:52:19.040 --> 01:52:23.917 topic, the uh one of the things that USDA does is we pay 600,000 federal 01:52:24.000 --> 01:52:26.957 employees. So, like we pay Secret Service, we pay DHS, we pay it's like 01:52:27.040 --> 01:52:28.557 it's like a thing inside of USDA. 01:52:28.640 --> 01:52:31.357 >> Interesting. >> And so the the payroll system that does 01:52:31.440 --> 01:52:34.797 that is a mainframe. Okay. And people literally explained it to me like this 01:52:34.880 --> 01:52:37.597 thing has a personality. Like you have [clears throat] to like you can't touch 01:52:37.680 --> 01:52:40.477 it the wrong way. You have to has to have like the right environment to work 01:52:40.560 --> 01:52:43.757 and it like all these things. I mean like a dozen people came to me and told 01:52:43.840 --> 01:52:47.277 me all these things. So then I went and visited it. I was like really excited to 01:52:47.360 --> 01:52:49.597 you know encounter this being mainframe 01:52:49.680 --> 01:52:51.677 >> and uh it's like a 01:52:51.760 --> 01:52:57.197 >> uh five-year-old brand new like IBM server you know it's just like it's not 01:52:57.280 --> 01:53:00.477 there's no tapes. There's not like a team of people underwhelmed. Yeah. It's 01:53:00.560 --> 01:53:02.477 like you know it's like this big. Okay. 01:53:02.560 --> 01:53:06.157 >> Um, but I was expecting like a [clears throat] like a small micro data 01:53:06.240 --> 01:53:10.317 center or something. Exactly. Yeah. So totally modern and it like I like formed 01:53:10.400 --> 01:53:14.237 this connection with it and I was like we have had so many conversations about you 01:53:14.320 --> 01:53:17.917 >> and I just thought that like this is potentially a future where it's like a 01:53:18.000 --> 01:53:22.077 data center a coffee shop you know like people might want their data to be 01:53:22.160 --> 01:53:24.477 hosted in a place that's like aligned with their views. 01:53:24.560 --> 01:53:25.997 >> Sure. Sure. Yeah. You know, because it's 01:53:26.080 --> 01:53:29.757 like I can trust like I don't want this in my house, but I can like trust this 01:53:29.840 --> 01:53:33.197 like cool company, local company that my my data lives there because I don't need 01:53:33.280 --> 01:53:35.837 it distributed across the globe. It's like >> I'm here. 01:53:35.920 --> 01:53:37.197 >> No, that makes sense. That's interesting. 01:53:37.280 --> 01:53:39.437 >> Country intelligence. >> Yeah. Yeah. Yeah. 01:53:39.520 --> 01:53:41.597 >> Intelligence. This is the future. I love 01:53:41.680 --> 01:53:43.517 it. Anyway, thank you so much for coming on the show. 01:53:43.600 --> 01:53:45.837 >> Great to meet you for Thanks for doing this work. 01:53:45.920 --> 01:53:48.477 >> Have a great rest of your day. Talk to you soon. 01:53:48.560 --> 01:53:50.717 >> We will wrap up the show. Yeah. 01:53:50.800 --> 01:53:55.757 >> Thank you for tuning in with us today, folks. We will be back on Monday. 01:53:55.840 --> 01:53:57.757 >> Yes. >> And we look forward to it. 01:53:57.840 --> 01:54:01.037 >> Some business to do tomorrow, but see you Monday. Leave us five stars on Apple 01:54:01.120 --> 01:54:04.477 Podcast and Spotify. Sign up for the newsletpn.com 01:54:04.560 --> 01:54:10.120 and have a wonderful weekend. We'll see you. Goodbye. Goodbye.