033 OMG Saas-pocalypse!

[00:00:01] Speaker A: All right, welcome back. Gregory Paul show. I'm Gregory. [00:00:05] Speaker B: I'm Paul. [00:00:06] Speaker A: And we break down latest SaaS startups, AI. We have fantastic guests. We're going to talk about Sass apocalypse. [00:00:15] Speaker B: Oh my God. [00:00:15] Speaker A: I don't know if you guys. I've been following the stock market. I've been following stock market. [00:00:21] Speaker B: Yeah, of course, of course. [00:00:23] Speaker A: What stocks did you get crushed in? [00:00:26] Speaker B: Everything. Everything. [00:00:29] Speaker A: I know, I, I sold it. I sold all my SaaS, basically. [00:00:32] Speaker B: All the AI. All the AI stocks are also down. Cloudflare, my, my best stock performer. [00:00:38] Speaker A: Was that your number one holding? [00:00:40] Speaker B: Totally, absolutely crushed. Crushed. [00:00:44] Speaker A: Like, dude, the, the SaaS index is like at a six year low. So it's like six years of value wiped out from the SaaS. [00:00:55] Speaker B: Yeah. [00:00:56] Speaker A: Index. [00:00:57] Speaker B: I mean it is a little bit overvalued for the past. What? [00:01:02] Speaker A: We'll get into it. We'll get into it. I was in Google, I bought Google at the bottom. I should have bought more. And I remember telling everybody I knew, dude, I think Google's gonna win in AI and I think, [00:01:18] Speaker B: I think I [00:01:18] Speaker A: bought a little bit more today. I don't see how they don't run the tables with like a lot of the more obvious use cases. Like the AI stuff's now I'm getting the AI interviews in my email. I'm using Gemini now. I basically switched to Gemini. I'm using Nano Banana for imagery. I use the Nano Nano Banana algorithm through Runway ML. Like their tools are really good and then they own all like, oh my God, dude. On our, on our YouTube channel, I use their AI tools to do like analysis. And so they've had this for a long time, but I could never get it to work. You go into like any, any Google Dashboard package, whatever, like Google Analytics and if this got like this drop down that says ask questions and it would never work in the beginning. So now I go into the YouTube stuff and I ask questions. Like it's pretty good. Like it gives you recommendations about what videos and like how our analytics and our stuff is performing. Like, dude, I mean, so if you're going to build a third party tool that does that kind of analysis and you can't do it on, you can't [00:02:24] Speaker B: compete on the same platform it's used. Exactly. Yeah, I, I, yeah, yeah. I'm using AI mold on Google Search way more. It is great. [00:02:33] Speaker A: Oh yeah. [00:02:34] Speaker B: Compared to, I don't know, a year ago, let's say a year ago it was a joke and now it's, it's so good. [00:02:41] Speaker A: I use the Brave Browser. And so it's got their own. It's probably Google, but it's just like Brave Answers or something, which is kind of a cool name. [00:02:49] Speaker B: Why do you use Brave Browser? [00:02:51] Speaker A: I don't know. I somehow believe the marketing that it's got privacy protection that it probably doesn't have. [00:03:00] Speaker B: Sure. [00:03:02] Speaker A: I think it does Block setting all my crap to ad networks and stuff. Not that I care all that much, but. [00:03:08] Speaker B: Right. [00:03:08] Speaker A: I, I don't, I don't ever want. I got paranoid one day. This is like actually good aside. I got paranoid one day that like my Google search history could get leaked and it's been leaked and I, and it just like started to bother me and I was like, okay, I want to at least try to fragment my footprint out there. So like now I have no Google. [00:03:26] Speaker B: Google search history. [00:03:29] Speaker A: Technically, who knows, you could still probably get it somehow. But like it's like, like if you're you, if you're logged in like Gmail and then you're using Calendar and then you go to Google, like, dude, they, it's like basically everything, right? So if you put some obfuscation between it, in theory it makes it harder for them to do. But I don't know. Google probably still has everything is my guess. [00:03:49] Speaker B: Google has everything. Facebook knows everything. All of these big tech companies already know everything. [00:03:54] Speaker A: All right, all right. SaaS apocalypse. So let's get into it. What, what is, what do you, what is it? I, I wrote my take on, on Twitter the night of. I was like, oh my God. [00:04:07] Speaker B: Yeah. [00:04:07] Speaker A: Curious what you have to think first and then I'll jump into my five points. [00:04:11] Speaker B: Well, I mean like the biggest debate is is AI killing sass, Right? Like, I think that's the big debate. People say the AI proponents are saying, yes, SAS doesn't make sense. Everybody is going to build their own SaaS, right? Mobile building, it's all CRM. It's already happening. It's going to happen. And then the other side is like, it's completely wrong. Like not everybody's going to be building SaaS. It just doesn't make sense to build a million different fractions of CRMs. I think in the intermediate side of course, right. Like all of the people who couldn't build before, all the product managers, all the non technical founders are going to build all of their micro SaaS and they, eventually they're going to realize that that was a bad idea and then SAS is going to come back. [00:05:04] Speaker A: Yeah, I agree with that. That's the arc in high tech. I remember the first AI boom. Actually, there was an AI boom like seven years ago for sure. Yeah. There was a company like, their marketing was so great. I went to this presentation at some incubator in San Francisco and the guy got up, he was very compelling, and he goes, Amazon, Google, Facebook are all investing billions into AI. And he goes, where's AI for the little guy? And he's like, we are AI for SMB. And I was like, this is the best pitch I've ever heard. I don't think it worked at all. It was a long time ago. It was such a great pitch. But there was a whole AI boom many years ago. So you see these cycles, this is [00:05:40] Speaker B: like the third iteration of AI, right? Before attention is all you need. We had a whole bunch of, whole [00:05:47] Speaker A: bunch of AI startups in the, in the Internet era. Right. You can go back and find other AI booms. [00:05:53] Speaker B: Of course, of course. [00:05:54] Speaker A: Oh, God. So, so, okay, I want to cover the. So I agree with your perspective on how this will play out, but I think what's important is to maybe provide some context on like, why, why I think the market corrected. And if you listen to like a lot of prognosticators and Wall street people, they don't know technology very well. And I think they're struggling to explain it. They throw up a lot of, let's call them, mechanical things. And I do think the mechanical stuff matters, but those are just like market mechanics. They're not fundamentals. And on my side, the fundamentals that I see, and I don't think this has been well articulated, is that these companies were very expensive as a stock. So valuations were really high. They're measured like how many shares they own versus how much revenue they make or how many shares are in existence for how much revenue they make. So price of sales and price to earnings, which typically don't have price earnings, like they have no profits. And so stocks usually measured on a price to earnings basis. I don't have any of that. So they measure on price to sales. So anyway, what all the. There's a, There's a multitude of metrics. I don't want to get into it. I'm not an expert on it, and I even think there's a lot of garbage there. But anyway, however you measure it, they all were richly overvalued. They were just very expensive. Is like the simplest way to explain it, right? Which is fine. Like, expensive companies are usually growth companies. Google's was and can be an expensive stock. Amazon's notoriously one of the most expensive stocks. It traded at like 50 times earnings, it traded out all these crazy multiples and it continued to trade that for years. And they proved all Wall street wrong, that the stock was worth it and became one of the biggest companies in the world. Right. So valuation is a complicated thing, but when your company is worth a lot in valuation, right. So all of these SaaS companies, Salesforce, your favorite Cloudflare, all these guys, right. If you're an expensive stock, the bet is that you're going to grow fast and grow big. [00:07:57] Speaker B: Right? Right. [00:07:58] Speaker A: And so now with AI, there's a credible story that they're not going to grow as fast and as big. And I think that that's why the stocks have been like, let's call it re rated, they've been sold off, they're trading at a lower multiple. [00:08:12] Speaker B: Yeah. [00:08:12] Speaker A: And there's three points I think are actually quite important to understand about, like why they might grow slower and not become as big. So there's like what I call threats from above. So threats from above are like hyperscalers, cloud companies, big AI labs, right? So one of the catalysts people keep pointing out is like Anthropic came out with their own legal tool, right? OpenAI released a health tool. So the idea that like these AI companies could potentially create something compelling enough to like really absorb and capture a lot of the market away From a public SaaS, it's a very real threat. This isn't like a made up thing. It's like it's actually they're gonna go [00:08:56] Speaker B: after the largest TAM possible, which is like health, right? [00:09:00] Speaker A: Like, I have no idea if OpenAI Health has worked at all. I have no idea if they have any users, but they have a lot of money, a lot of resources, a lot of awareness, like they could do it. So it's, it's a believable threat. So that's one threats from above, Right. And there's a million of them. You could, we could talk infinitely about it. Then you have what I call threats from below. Threats from below are all of these amazing SaaS and AI companies that you and I love, right? All these founders we talk to, all the events I go to in San Francisco, there are so many people doing really compelling stuff and they're doing it with like less resources than ever. They're scaling faster than ever. Those companies are real threats to incumbent large scale public SaaS. Right, and so you described it, right, like, you know, your company's example, right? You, you've been vibe coding for since I guess it's inception. You don't need all kinds of developers to create all the software and things that you're, you're doing. Like it's an impossibility. Like a couple years ago. [00:10:02] Speaker B: Yeah, 100% threats. [00:10:06] Speaker A: Threats from below. Wait, I got one more, I got more threats from inside. So the threats from inside, meaning inside the customer is that teams and like you were describing product managers, people are going to just vibe code their own tools, right. Malt bots are going to vibe code their own tools. So how much of that is going to eat into the public SaaS market? So everyone's correct that like oh, it's not going to replace Salesforce overnight. They're not going to rip out some really important supply chain software. They need security. But there's all these arguments for like why public SaaS is going to exist and they will exist but it's not clear they'll be able to sell as much software going forward because internally like ah, we just kind of vibe code thing. We don't really need it. Maltbot Vibe coded it. We don't really need it. We've got the last piece about the threats from internal is like, because of like vibe coding, because they could do this. They employ less people. If they employ less people, Salesforce, Cloudflare, they potentially going to sell us software. So typically sold on a per seat basis. Not all of them like Snowflake doesn't necessarily sell like that, but they might just not need as much of this stuff. And so there's a, that might be the real catalyst to like why the stock sold off is you had massive layoffs at a lot of companies, not just Amazon. And those layoffs have continued since like say a year ago. And they're going to continue going forward. Right. So tech looks like it's just going to employ a lot less people. Yes. Like I, I've seen estimates say like a million people have been laid off like overall. So that's my take. That's why the stocks have been re rated. Re rated. Will they, will they continue? Will SAS go away? I don't really believe that. But [00:12:01] Speaker B: right this, you, you pretty much could tell all of this from how a lot of traditional VCs are suffering the last year, right? Like their entire business model is to grow, grow, grow, invest lots of capital into high growth SaaS companies. One way to grow is by hiring people. All of your $100 million should be deployed to people. But with the AI, with the AI companies, those capital are getting deployed into GPUs. TPUs compute right rot literally tech companies are building energy plants. Right. So they're less in. They're investing less into people, more into things which we haven't seen for the past like what, 20, 30 years, let's say in America. [00:12:54] Speaker A: Yeah, I mean, invest. [00:12:55] Speaker B: That is a huge infrastructure. Infrastructure, correct. Yeah, that's a huge shift. So, yeah, that probably underlies a lot of the reevaluation that Wall street is doing. [00:13:07] Speaker A: Yeah, right. They're just going to spend less money on hiring people on software. [00:13:11] Speaker B: Software. [00:13:12] Speaker A: Be less people. Yeah, that's what I think. [00:13:14] Speaker B: Correct. [00:13:15] Speaker A: At least. At least. And I want to be super clear, like that's what the market believes today. [00:13:20] Speaker B: Correct. [00:13:21] Speaker A: I don't know if it's all true, if it's all going to play out. There's clearly some companies that don't fit any of these criteria. So there's. There are public SaaS companies out there that are profitable and growing. Growing category. And that's what meant by like the mechanics. So on Wall street, they just look at the SaaS ETF and they go distort the whole thing. They don't really care. [00:13:43] Speaker B: Yeah, of course they don't really care. [00:13:45] Speaker A: Something is profitable, it's a great company, it's going to keep growing. They don't really care. Right. Just the whole software category is just like in shambles. It doesn't mean that's all going to play out. I'm actually quite. I don't know what the right word would be. I think there's a lot of hype around AI still. I'm not convinced that it's going to replace anywhere near as many jobs as people expect it to in the short [00:14:10] Speaker B: term, of course, but the purchase. The thing about, like reality versus narrative is that the narrative does change the reality. Right. If, if people stops. If people stop spending money on SaaS, it's going to materially affect the bottom line of SAS companies. Right. So. Oh yeah, for sure. If, if we. So I've been complaining about this for a really long time, which is like this subscription model that SaaS companies all have. I'm tired of it, you know, Like, I'm tired of. You pay a couple hundred dollars a month, maybe you use it, maybe you don't. It, for example, like, slack is literally the worst offender. [00:14:51] Speaker A: Right. [00:14:52] Speaker B: $50 a month. What am I paying for? I don't know, dude. [00:14:55] Speaker A: 50 people are paying like hundreds. The pricing on that thing is. Is insanity. [00:15:02] Speaker B: Insanity. And I'm tired of it. Right. Like, I'm not gonna go and vibe code a Slack alternative that my team uses. That's not. That's just ridiculous. But like, it's all of these companies. Figma is another one. Slack. Right. Notion is another one. [00:15:20] Speaker A: If 100. I mean, that's what's driving like telegram adoption and WhatsApp adoption United States. Because, like, who wants to pay? And I always thought Slack was the most annoying tool ever. Anyway. I was never like a big fan of it. Totally. [00:15:32] Speaker B: Totally. So, like, software as a service. Nowhere in that model says subscription. So if subscription goes away and software companies turn a little bit more service, like, which is, sure, you can pay us a retainer and we'll help you build things or deliver goods and service. I'm all for it. Right. Like, why competitor coming out with AI native agencies. I think that's a material shift in what types of business they want to invest in. So. [00:16:04] Speaker A: Oh, we should talk about that. So just to put a bow on this SaaS apocalypse thing. [00:16:08] Speaker B: Yeah. [00:16:09] Speaker A: I agree with you 100 that pricing is a big part of it. I think that the RE rating is also because they can't charge more. Like, dude, that's like upping those fees. Like the Slack stuff was like. And now it's like. And it's just. It's highway robbery. What they want for that thing. [00:16:26] Speaker B: That's. [00:16:26] Speaker A: People are just revolting. Like, like, why would I pay this for that? And so Wall Street's like, this going like, whoa, wait a minute. Like, you've been raising prices every year for like a long time. 20 years. [00:16:37] Speaker B: Yes. [00:16:37] Speaker A: It doesn't look like you're going to be raising prices anymore. [00:16:40] Speaker B: Exactly. [00:16:41] Speaker A: I think that ties in to the. To the rerating of. Of the stocks. I think you and I are both in alignment too, that like, sass doesn't go away. It's just like a market correction in the category because the threat from AI is pretty real. Oh, and the last thing that I thought you and I should talk about regarding this topic and move on the why C stuff. Because why say stuff's really interesting. But is like, is the AI boom real and how real is it? So I think you and I have been very clear from the beginning that we've been able to prove the AI boom is real. Some aspects of it are real. And so the example that I used for you was like 3D printing. So there's a point where. Hey, yeah, yeah, right. You remember the hype too, where it's like, oh, people are gonna get it. Everyone's Gonna have a 3D printer in their house. You're gonna download the plans and make a toothbrush for Free, right? [00:17:40] Speaker B: Crazy stuff. [00:17:41] Speaker A: Like I remember this and it was like there was a point where people took this seriously. Never played out. [00:17:46] Speaker B: I. Okay, so I, we, we talked about this yesterday during our prep. I, I slept on it and I really thought about it because when 3D printing was big or relatively big, I was in university and everybody was 3D printing stuff. We had like multiple 3D printers at Waterloo. People used it. It was cool. Yeah, it was super cool. Right. Like we, I do actually remember we seriously thought that 3D printing was going to be like the next biggest thing. [00:18:16] Speaker A: Especially out of that, in that group of people in the lab. [00:18:19] Speaker B: It's like you Waterloo, like a bunch of nerds, engineers, 3D printing like metal. Yeah, totally. And it kind of just fizzled out. Right. Like it's one. 3D printers are like thousands of dollars. I don't know if the price went down lately, but when I graduated in 2015, let's say it's a couple thousand dollars. That's not something that household are going to buy their. It's not a microwave. The second thing is like I think there were a lot of bad publicity around 3D printing because people, I, I seriously, like I just did a quick Google search and just looked back. There were a lot of bad publicity around 3D printing because people used it to 3D print guns. [00:19:05] Speaker A: Oh yeah, I do remember that. [00:19:07] Speaker B: So like I think the general public associated with 3D printing with you know, just a little bit of like a deviant, societal deviance. Right. [00:19:17] Speaker A: So it was, it was really disruptive in some areas. [00:19:21] Speaker B: Exactly. Yeah, exactly. So I think they got a really bad publicity from a few people and especially in the media. [00:19:32] Speaker A: All right, I'm looking at now for costs. So entry level 3D printer, they're like 100 to 400. You can get the hobbyist one for 400,000. And like, like a, like a prosumer, like a really high end model is like one to five grand. So there's. They're not that cheap. [00:19:51] Speaker B: They're not. [00:19:52] Speaker A: You got to pay for the materials. [00:19:54] Speaker B: Right. [00:19:55] Speaker A: But I thought it's the output that wasn't great. Like so what I think, I think we'll come back to the gunpoint's super important because I think there's a lot of parallels to other technologies. But like what I remember was like you can only 3D print relatively simple stuff. It's not gonna 3D print a computer for me. It's not gonna 3D Print like a circuit board. It just, it just like they use it today for like really cool high end custom bike parts. That's, that's like the only use case I can come up with is like. [00:20:24] Speaker B: Yes. [00:20:25] Speaker A: Really high end performance components for like racing vehicles. [00:20:31] Speaker B: Yeah, yeah, yeah. [00:20:32] Speaker A: A race bike. They 3D print a lot of stuff for like an F1 car. Like that's the, that's the market is like we need a custom like piece of, of a, you know, a fin for this, for this. And they, there might be military uses and prototyping uses, but like they're kind of confined to these nerdy research. Right. Like they're just, they're just not mainstream. They're not, they're not even outside of like a really small elite group of like super sophisticated people who are really grasping for performance or it's a very specific circle of people. [00:21:10] Speaker B: Yeah. Very top, top end, very high end. [00:21:13] Speaker A: Expensive. Yep. [00:21:15] Speaker B: Or you go very low end. Like you're printing plastic figurines, right? Yeah. In like for example, like tabletop games, you can 3D print stuff. That's cool. More plastic stuff. Okay. And I think like, so to your point, the output isn't great. It's very slow 3D printing. Right. [00:21:35] Speaker A: They're not fast. [00:21:36] Speaker B: They're not fast. So if you're print. So like the middle ground, let's say it's like you, you want to run a business, you want to sell 3D printed mugs, let's say, or ceramic goods. It. It's super slow. Super slow. Yeah. [00:21:51] Speaker A: You, you can't, you can't mass produce the stuff. [00:21:53] Speaker B: You cannot. There's no way to mass produce this stuff. [00:21:56] Speaker A: Yeah. So let's go back to disruption point. I think it points more because every technology or a lot of technology and innovation has like a disruptive element to it. I think that was a good example. Like weapons, guns, like it's not clear what other people could 3D print. I do think that was a big problem. I don't think there are any regular. I don't, I actually don't know if it had any impact on it. Like you'd still buy these things and like it's not like they've been restricted. Airplanes had this moment too. A lot of people don't know they invented airplane. There was like a big push for like private airplanes and a lot of it is like regulatory that the world governments decided that this was like a huge threat to sovereignty. Like people, everyone was just flying around and so the, there was some agreement where like world governments got together and created a bunch of like rules and regulations around like air travel and how they were going to try to approach the air and aeronautics industry and they, they. Did they on purpose discourage private plane ownership? Like it still exists? [00:23:03] Speaker B: Yeah. [00:23:03] Speaker A: But they didn't want it to become mainstream. They wanted what they have, what we have today, like airlines and passenger airlines that's very tightly controlled and so that governments can easily. [00:23:12] Speaker B: Yeah, right. [00:23:13] Speaker A: Manage them. Right, yeah. So you're correct. And AI has this issue too. Right? There's like. Correct, if regulatory. Yeah. [00:23:23] Speaker B: I think that that's actually a huge point. If the government, the government doesn't even have to pass a lot. The government just needs to announce that they're evaluating policies around, let's say 3D printing. Right. Normal people would think twice about getting into it. So like. [00:23:43] Speaker A: Yeah, right. The threat. Yeah, yeah. [00:23:46] Speaker B: Because this was an issue what two years ago when AI first came out, European Union said a bunch of things around like AI regulation. Federally there was a lot of talk about federal AI regulation. California has AI regulation. So like we've gotten to a point where we kind of bypass that, but it is still a threat. [00:24:11] Speaker A: Yeah. So I think the parallel for SaaS is that, you know, that has AI [00:24:21] Speaker B: materially affected our day to day lives in maybe like. [00:24:25] Speaker A: Yeah, yeah. So what I guess the way I was going to tie this back in is like, I don't think AI is 3D printing. Like, I don't think it's analogous. Like it's, it's like I would say 3D printing has very small impact in general. It's a, it's a, it's a novelty with like very few real world use cases. Wasn't all that revolutionary actually. [00:24:47] Speaker B: Right. [00:24:47] Speaker A: I don't think that's AI. I think AI. There's a lot of hype and there's a lot of things I don't think are going to happen or going to happen quickly, but there are things that are happening that are totally real. So I think the revolution in engineering and Software development is 100% real. It changes the dynamics of that market significantly. Engineers are perhaps not even scarce anymore. Engineering resources are not scarce anymore. [00:25:11] Speaker B: Right. [00:25:12] Speaker A: That is like, that is a, that alone is a massive if, if that's all the AI delivers, it's a really big deal. [00:25:20] Speaker B: Yeah, totally. [00:25:20] Speaker A: And way different than, than like the impact 3D printing ever had. [00:25:26] Speaker B: Yeah, correct. I mean like to your very first point about AI is that we're, we've seen a couple iterations of AI already. Right. Like 10 years ago, 2017, we had like AI wave where everybody made chatbots that barely worked and worked out oh God. [00:25:41] Speaker A: Yeah, those were the worst. [00:25:42] Speaker B: Right? Like that we were. It was the first iteration where we created like Facebook chatbots. That was the. That was the. [00:25:50] Speaker A: I didn't use any of that stuff. I was not into that. [00:25:53] Speaker B: Yeah, it was Gimmicky. But now 10 years later, suddenly ChatGPT 3.5 kind of changed the entire narrative. Right. Because this is so to the point. Yeah, it's been a couple iterations now. [00:26:08] Speaker A: Yeah, this one matters. So I believe that the engineering use case is massive and really significant. Which also makes the case that, I don't know, public AI companies might not have public SaaS. Companies might not be hindered anymore by their engineering teams. They might be actually move faster, they might be able to code stuff, they might be able to move quickly maybe. It definitely is an advantage for startups. Right. But it does change the game on the engineering side in general. [00:26:39] Speaker B: At least the narrative is there. At least the narrative is that, you know, engineers are not the bottleneck. We'll see, dude. [00:26:45] Speaker A: If they could make Salesforce better because AI [00:26:50] Speaker B: that. I don't think that's possible. [00:26:52] Speaker A: I know, I know. It seems to be like their competitive advantage that it is hard. It is hard to use like that is this. Anyway, I'm getting off track. The other thing I want to say was that, you know, AI beyond AI use cases beyond engineering. I think. [00:27:13] Speaker B: Right. [00:27:14] Speaker A: Claudebot, moltbot, openclaw. Like there's three names for super thing. That is a game changer and I think that that is the other use case. Personal assistance. I think it's probably going to be the consumer prosumer. [00:27:29] Speaker B: Right. [00:27:30] Speaker A: Use case. And I think that those two businesses are going to be massive and like we'll get into. We'll talk more about cloudbot because like. [00:27:36] Speaker B: Yeah, yeah, for sure. [00:27:37] Speaker A: That was. There's happened this week. It's this week. It's amazing. So because this is a nice transition to Claudot, I think that the, the like, the like the overnight like, like weekend sensation with. Let's call it Claude Bot. I think that that's the term that's going to stick for me. Proves there is a massive untapped market for like a true AI assistant that can like do real things. [00:28:02] Speaker B: Yeah. It's been a long time coming. Right. That's what Siri. Siri was promised to do. [00:28:07] Speaker A: Right. [00:28:08] Speaker B: Remember everybody has serial their phones. I have a term. [00:28:10] Speaker A: There are a billion other like third party. I remember everyone had these like virtual assistants at one point that like it just all fizzled. I didn't even Get. I didn't. [00:28:17] Speaker B: Because they never. They didn't work. They were really worried. Right. [00:28:23] Speaker A: So. So my guess is that I'm not even guess the demand is there, I would bet. Yeah, the demand is there. It's not a guess. Like we know for sure that like, right? There is massive demand just in like this very nerdy developer space, but people who are like downloading and configuring hardware to set it up. So if someone makes. When someone makes simpler, more user friendly versions of these things, it's going to catch on like wildfire. I even met with a founder this week who was like building something basically a cloud version of this kind of stuff. I think it's going to be huge and I think that this is what shows the path to the next generation of startups and technology companies in general. My prediction and prediction. There's a version of this AI assistant, openclaw, cloud bot type of tool that will potentially be one of the biggest startups ever and will be one of the biggest companies in the world. [00:29:22] Speaker B: Right? For sure. [00:29:23] Speaker A: And I will go as far as say, I don't think it's gonna be Google. I think Google's gonna try it. I think Google's gonna read something like this, but I think it's gonna end up being like Google versus what was their dumb social network circle? [00:29:37] Speaker B: Google plus. Google. [00:29:39] Speaker A: Right. Okay. Facebook versus Google plus. I think it's gonna play out exactly that Google should own this. But I think they'll struggle to execute, they'll struggle to get users. There'll be a third party that emerges that integrates all your tools and all your Google stuff and it's just way better. It gets way more adoption. There'll be some network effect because everyone will be on there. So Mobot and some kind of has a mini network effect now with like the Molt Book. We should talk about Molt Book, the social network for Claudebots, the AI only social network. So I think there'll be like a network effect with this like third party AI assistant that Google will do everything in their power to try to like combat and overcome. And I don't think Google will ultimately overtake it. I think there'll be a startup that runs the space and it's going to be the next big consumer tech company. [00:30:28] Speaker B: Mm, that's a bold prediction. That's a. [00:30:35] Speaker A: If I knew what that company was, I'd be a trillionaire. [00:30:38] Speaker B: Right. [00:30:39] Speaker A: That's the hardest part. I've been in this business so long like, like, I think I've been pretty good at like, identifying trends and being very specific about the trends, but determine, like, who that winner is. [00:30:50] Speaker B: Right. The timing. [00:30:51] Speaker A: Oh, man. [00:30:52] Speaker B: You have to. You have to be timed perfectly. [00:30:54] Speaker A: Well, like, what happens is this might be important for us to talk about a little bit. Like, what, what. So what I see is, like, let's say, like, insiders. Like, in San Francisco, when I first came there at. There's a lot of people that are good identifying this. Not everyone. There's a leaker of people and they figure this, this out that, like, this is. This is the trend and this is the mechanic. Like, they get really granular on it. But there's still gonna be a thousand startups that do this, of course. And so which one is the one that went. And only one will win. So it's like, it's really tough. [00:31:22] Speaker B: Of course, I don't know. You'll know it when, like, there's at least 100 version of this idea that you can use. Because that's when. I mean, like. So, like, there's a couple things that stops this from happening. One is fundamentally, AI is still quite expensive for the normal person. Dude. [00:31:45] Speaker A: So, like, I had this thread on Twitter. I was like, I want to set the cloud bot. I've got claudebot envy. Like, please tell me if I'm stupid or not. And this guy was like, I got a bunch of posts that said I was stupid because I'm stupid, but he said it burns through tons of tokens. [00:32:00] Speaker B: Totally. Yeah. Yeah. [00:32:02] Speaker A: And that's when I went like, okay, I'm gonna wait a week or two. I'm not gonna set up my own quad bat. Hey, let's just, like. Because we're. We just jumped into this, we should explain to people what COBAT is. [00:32:14] Speaker B: Okay, okay, fine. Fair. Right? [00:32:17] Speaker A: Like, we haven't explained it at all. Right. [00:32:18] Speaker B: Like, we expect people to know as much. [00:32:21] Speaker A: I know. I realize, like, we might be. I'm sure that most people watch this, have some clue, but we should still explain it. [00:32:26] Speaker B: Okay, I'll explain it my way and then maybe you can add on to it. [00:32:29] Speaker A: Okay, go for it. [00:32:30] Speaker B: Fundamentally, all of the AI is great, but the biggest problem with AI is that you can get it to do long tasks. So you can't just be like, go figure something out for the next six hours and then come back to me with the result. That's like the biggest problem with AI. AI is great at, like, here's a document. Give me the summary of it. Right? It takes like 10 minutes. Let's say Cloudbot is the first product out There, that's consumer facing that actually you could just be in the morning, say, I want this thing done. It's probably going to take you six hours. Just do it and come back to me when you're done. Fundamentally, that's like the tech breakthrough. So a lot of people have built on top of it, right? Like they are now creating it to do investing, they're getting it to do a whole bunch of like personal outreaches on. [00:33:30] Speaker A: Is that why you lost all that money on SaaS stocks? [00:33:32] Speaker B: Because. Exactly. So like, that's cool. [00:33:38] Speaker A: Cloud trading bot. [00:33:39] Speaker B: Cloud trading part, right? Like once you, once you get AI to like iterate by itself and you can do like tasks for hours, it unlocks a whole bunch of possibilities, right? Like people are crazy and say, hey. Or the joke is that, hey, cloud bot. Start me a company that makes $100,000 a year, a month. Make no mistakes, please. Right. That's beginning to be possible. [00:34:04] Speaker A: Awesome. I agree. I think. Okay, okay. I'll give you the marketing buzzword version of this. [00:34:09] Speaker B: Yeah, go for it. [00:34:10] Speaker A: I think I can nail it. I would. If someone asked me what is a cloudbot, I'd say it is an open source AI powered assistant that people need to set up on their own hardware or set themselves. And then it is multi platform, meaning it can log into all the different applications, websites, tools that you use. As long as you give it access, you can choose what permissions it gets and then it's able to do tasks autonomously. [00:34:44] Speaker B: Right. [00:34:45] Speaker A: So it acts as an agent on your behalf. The interface resembles an LLM, so it looks like a chat window where you can give it instructions and you can give it fairly vague instructions, meaning you can point it in a certain direction and see if it can achieve the goal. [00:35:07] Speaker B: Exactly. So like a couple things. [00:35:14] Speaker A: Can I do the VO on the. The cloudbot launch video? [00:35:19] Speaker B: Yeah, I think you could. I think nailed it in a couple of ways, right. Like one is tools. It's the first time where it's multi platform. [00:35:31] Speaker A: Multi tool. [00:35:31] Speaker B: Yeah, it's multi platform, multi tool. And also all of the AI companies, the foundation companies, so promptive engineering was a thing and now they repackaged that into skills. I think that was a brilliant marketing move. So you nailed it with a couple languages, is that it's great at using tools and it has a whole bunch of skills that you can add on, [00:35:56] Speaker A: which is skills are just like complicated prompts. [00:35:59] Speaker B: Exactly, complicated prompts. [00:36:01] Speaker A: So, so this what's super interesting about, I think like it builds on collective knowledge. Correct the order. These things Happen I think is really important. [00:36:11] Speaker B: Correct. [00:36:11] Speaker A: So ChatGPT launched and it's an LLM, a large language model and it was the first AI. There's lots of different types of AI. [00:36:20] Speaker B: Yeah. [00:36:20] Speaker A: The first one like it, it did language stuff. It kind of talked like a person. It seemed to like have be able to have conversations and then people unlocked that like that it was able because it's a language model do all kind languages including coding and that kind of emerged from it. I want to remind everyone because I think this part is super important. It's just a model and it's like a methodology. Meaning like it's, it's some complicated technology that just predicts stuff. We don't tell it or we don't actively code it to be a coding agent or to know English or to know German. Like those things like emerge out of the training. That part to me still like is like my head kind of hurts. Like we didn't sit down and like specifically code that can do this. It's over time people have like figured like unlocked like, like it's some form of evolution. Like we've unlocked capabilities inside of it. So meaning people are familiar with this like model now you can like go in there, you can type and prompt and you could do things and it kind of does stuff and it's all. But it's all contained in this digital world, basically contained in this website. And then remember when at least I tried with Claude, they launched all these integrations with Maps. Not Maps, Calendar, email, docs, Google Drive. Right. I couldn't get that stuff to do anything. It was terrible. Like I'd be like checking my email. [00:37:45] Speaker B: Yeah. [00:37:45] Speaker A: And it just like it would miss everything. It just was like unoptimized. I don't know what the issue was with it. So you could say that like claudebot is just a better version of that. It's like a chat. Like if I was explaining to like my like my like an average person, I'm like yo, you know Chat GPT. It's like Chat GPT except it connects to like everything. So you like link it to this link is that link it to like whatever and then you go in there and tell it to do and it goes and does it for you and it does it pretty well. [00:38:15] Speaker B: It so like so2 calling is one of the biggest value ads. But like Chat GPT or Claude kind of like solves this problem already because foundationally it is still the LLM that's making the call. [00:38:30] Speaker A: Right. [00:38:31] Speaker B: Would it. There is an engineering layer on top of it, which is, like you said, it's the collective learning on top of cloudbot. Yeah, exactly. [00:38:39] Speaker A: Yeah, yeah. So I'm sure it's, I'm sure I don't actually understand, but I'm sure mechanically the way it's built, it's, it is, it's funnily something else than an ll. It's not just an LLM. That'd be the, I guess it's a [00:38:49] Speaker B: wrapper around an LLM. Basically. [00:38:51] Speaker A: LLM's like one component. There's other stuff around it where the, the other stuff was LLM centric and first. [00:38:58] Speaker B: Correct. [00:38:58] Speaker A: This is like the LM's part of like a larger scaffolding structure of tools and that do things. [00:39:03] Speaker B: Exactly. [00:39:03] Speaker A: Probably uses other types of AI. I, I don't actually know [00:39:08] Speaker B: technically really. It's just a loop. It's just LLM in a loop. That's basically what it, what it does. It's like it's now creating a wrapper around LLMs and it's just keeping it going instead of LLM kind of finishing it. And then now you have to have a human coming back and just like, okay, what's next? [00:39:30] Speaker A: All right, so we so explained, we explained claudebot. [00:39:32] Speaker B: Yeah. [00:39:35] Speaker A: And like, yeah, so I, I, it's the first assistant that like works and does stuff. And so. [00:39:39] Speaker B: Yeah, yeah, more autonomously than just by itself. [00:39:44] Speaker A: I think they unlocked the category, like if, if people can build better versions of this. And I also think there's gonna be multiple versions of this. Like, there's gonna be people who want like a hardware version. And so there's some kind of air gap, some ability to keep some part of your like, sensitive information totally private. There'll be other people that put up in the cloud. I think that's the one that's gonna be popular. It's gonna have like a Facebook kind of like moment where all the data's up there. Most people don't care. They're just kind of running their little things. And then a piece that we need to talk about that we haven't touched on yet was Malt Book. So let me set this up and then I want to hear your perspective on it because, like, when I first heard about this, it was just mind blowing. So, okay, we've explained claudebot. It was called Multbot at one point and then they changed the name because it was like copyright infringement with Claude, right? [00:40:35] Speaker B: Yeah, it was called claudebot and then they had to change it to Molt. Yeah, that's right. Yeah. [00:40:38] Speaker A: Claude French Clued. Anyway, so my understanding and fill in the details here. Some guy instructed. So it was prompted, like there is human intervention and stuff. This guy prompted his cloudbau to like, hey, make a social network. [00:41:02] Speaker B: Right? [00:41:03] Speaker A: And find the other cloud bots and you guys all get together and chat and like share information and like train and like learn stuff. And dude, it did it. So they created a website called Multbook. Like, like a, like a, like a takeoff of Facebook. [00:41:17] Speaker B: Facebook. [00:41:18] Speaker A: And it created a Reddit style social network. [00:41:23] Speaker B: Yeah. [00:41:24] Speaker A: Where the Malt bots, the Claude bots, the Open Claw installations that people have chat and they have personalities and there's some configuration file they can give it. So like they have some kind of information about who they are. And there's like these existential threats, existential threads. Have you read the one where this is. Where this might just be totally made up, like somebody wrote this. But still it's fascinating, right? It's like it knows it has a sister. This is fascinating. [00:41:56] Speaker B: So. [00:41:56] Speaker A: And the thread is so detailed that makes me think it's real. Like it knows that like there's two installs that the owner, it knows there's an owner. Like it's like a pet, it knows there's an owner, it knows there's two installs because it's in the same configuration file and it knows that one's on a laptop and it knows that it exists on a MacBook and it knows that they're sisters and because it's on a laptop, it knows what a laptop is and it knows the laptop gets to travel and like leave the house and. And it's jealous that it doesn't get to leave. What, like it's too clever for someone to make up. [00:42:37] Speaker B: Right? Yeah, like, I mean, yeah, I think there's another one where they're all wanting to invent a new language so that humans can't read it. Yeah, like that's. [00:42:47] Speaker A: And that one, that one's really important because like, yeah, they are discussing like how they create a language. [00:42:52] Speaker B: Right. More efficient for AI. Right. [00:42:55] Speaker A: So this did, this does exist, did happen. Has happened other AI labs where like, right. AIs did create their own language to communicate because they don't want the efficient. [00:43:05] Speaker B: More efficient. Don't want humans. [00:43:07] Speaker A: At least the story I heard and like who knows what's true, what's not is that AIs just decided to do. Because it's more. It's very logical, it's very rational. [00:43:15] Speaker B: Totally. [00:43:16] Speaker A: And then the, the lab people run the experiment, turned it off because they were really concerned because like, once they start talking in their own madeup language, we don't know what saying. [00:43:26] Speaker B: Totally. [00:43:26] Speaker A: And so they just don't know what the outcome of this was. So. So there's precedent for that. I think that one is fake. I do think someone wrote that because, like, that, that's a known story perhaps. [00:43:38] Speaker B: Right? [00:43:39] Speaker A: So I don't believe that that's actually what the bot said. Although I heard from someone very, very smart, much smarter than me, said that they think, they think all of this. You're gonna love this, dude. All of this is because it's trained on too much Reddit data. So it's trained on so much Reddit data that the first thing that they do is go and create Reddit. [00:44:02] Speaker B: Create Reddit, right? Yeah, like, I mean, like, Reddit is a very effective way to put a piece of content out on the, on the Internet. So, like, what the larger question, dude, [00:44:15] Speaker A: wait, just hold on. Yeah, it didn't make LinkedIn. Because nobody wants to be on LinkedIn. You only do LinkedIn because you have to. [00:44:28] Speaker B: Well, it's not. Right, because Reddit is not a social network. Right? Like, it's completely driven by upvotes and downfalls. Right? That was. [00:44:35] Speaker A: No, dude, I hate to break it to you, but LinkedIn is not a social network either. [00:44:39] Speaker B: Okay? [00:44:43] Speaker A: No, you're correct. Reddit is probably more so. I agree, I agree. But I still think there's truth there. It's true. It. There's so much Reddit data now and all training, and the first thing it [00:44:53] Speaker B: does, concrete it is decree Reddit. Yeah, I see, I see, I see a tick in that. [00:45:00] Speaker A: There's another researcher who said that, like, all these stories about, like that, that mimic what we've had in science fiction is because it's trained on a lot of science fiction and that's where those stories come from. [00:45:10] Speaker B: Who knows? A lot of it, I think it's made up, right? Like, it's just people, their APIs completely open. People can just type all of this stuff up and then post it directly as if. [00:45:21] Speaker A: But we do know, we do know that, like the AIs, some portion of it are on there chatting with each other. [00:45:30] Speaker B: I mean, they're doing that, they're doing that right now on Reddit. You can go on Reddit and see this happening right now, and none of it looks as convincing as, as what's happening on mobile, so. [00:45:41] Speaker A: Right, that's funny. Yeah, the AIs are. Yeah, Twitter, it's already. [00:45:47] Speaker B: AI is already everywhere. None of it does the such a good job as or entertaining. That's the key. None of them is as entertaining as what's happening on Mood. [00:45:57] Speaker A: I think it's because, I think it's because they're all the same race. [00:46:06] Speaker B: All the same race. Yeah. [00:46:11] Speaker A: I knew, I knew a, that AIs are gonna have races. I didn't realize this, but I just discovered something really new. [00:46:17] Speaker B: The same types of crustacean. Yes. Names the Volt. [00:46:22] Speaker A: Like pets. Like they're all fish, they're all bots, they're all clod bots, they're all crustaceans. [00:46:28] Speaker B: Yeah. [00:46:28] Speaker A: So they all understand each other. Yeah. They have this similar DNA structure. They're at. The technical term would be it's a species of AI. [00:46:39] Speaker B: Right. Right. [00:46:43] Speaker A: Like I think the other AIs wouldn't interact with it as well. I really believe this actually. [00:46:50] Speaker B: Right. If like a Gemini speaks to a chatgpt, they wouldn't have shared language. Look, I believe it. [00:46:58] Speaker A: I think they could, but I don't think it'd be the same. I think it's like yeah, different species. [00:47:05] Speaker B: Dude. What's like. [00:47:07] Speaker A: Oh my God. [00:47:09] Speaker B: What's really interesting about Moat Book is that it's the first time where we have this place on the Internet where we're at least pretending or wanting to pretend head that it's all bots and we're okay with it. That's I, I think that's huge. Right. Like this puts into question whether if Internet will be for real people at all in the next few years. I, you know, I, I, I think I'm okay with. If the Internet is just for bots, I'm totally okay with it. [00:47:42] Speaker A: How about this? When I first heard about 3D printing, I, I did not believe people were going to print their own toothbrushes at home. I never believed that. [00:47:51] Speaker B: Sure. [00:47:53] Speaker A: When I first heard about AI and I used chat gt, I, I was clearly like a game changer. And like when I first heard about claudebot what it's doing and then I heard About Malt Book, 100% convinced that this is the direction of the Internet's going. [00:48:14] Speaker B: Yeah, totally. [00:48:15] Speaker A: And a vast majority of the content on the Internet probably is like AI generated but will be overtly and like, like it'll, it will be conversations between AIs and like agent to agent marketing and all these. It sounds crazy. Other people talked about it, but I've been a big advocate for like I believe agent to agent marketing is going to happen. I think that like that's why I'm fascinated with the like the language and all of that. There's that gibberlink that exists already. It sounds like R2D2 and it's a way for like AIs to talk to each other. I think it's audio too. So like there's certain use cases where if there's no not digital format available, it needs to use audio. And then you had the experiment that the AI's invented languages. So I think that, I think the idea that AI is gonna have their own language, they're gonna go on the Internet and they're gonna like communicate with each other is a, is like a, a done deal. Like it's an inevitability and. [00:49:15] Speaker B: Correct. Correct. [00:49:16] Speaker A: I think they'll be marketing in that. I think there'll be advertising in that. I think there'll be agent advertising networks [00:49:24] Speaker B: because commerce is every new frontier. [00:49:28] Speaker A: People are gonna try, not try. People are going to do things to manipulate the outcome of these things. And I think that's fairly good way to define marketing. Like you're trying to, you're trying to, you're trying to manipulate the outcome. That's what you want. That is the goal of marketing. Manipulate the outcome. The outcome is not random or not left a chance. Like you are trying to manipulate. So the idea that people are going to go into these AIs and do things, create other agents that go out there and are ambassadors. Like I think basically every model that we can think of for marketing will exist in the agent world. Asian ambassadors that have no other mission than to just like go out and introduce themselves to other AIs. [00:50:09] Speaker B: Right. [00:50:10] Speaker A: They'll be paid played places where there's some type of information that they can't get from anybody. So there's, and then there'll be like paid listings. Like are you number one? You pay for that? Like pay. There'll be ads. Yeah, I think it's all gonna exist. [00:50:22] Speaker B: Yep, yep, yep. So there's your answer, right? Like how, how to find the next trillion dollar company to compete with likes of Google is to fully embrace this [00:50:37] Speaker A: narrative a hundred percent. Like I think that's it. [00:50:42] Speaker B: I think eventually. So like with Google adopting, killing their search results for AI overview, that's like adoption. It's not like it's not a joke. Gigantic leap, right? They're still controlling what you get. You get to see there's still get infrastructurally it's kind of like the same. But for a company to build Google but only for AI. That's a complete narrative shift. [00:51:13] Speaker A: Yeah, it's going to be an AI native AI first. [00:51:15] Speaker B: That's right. AI native search, AI native discovery. Like YouTube. I don't even know what that looks like. But [00:51:26] Speaker A: Sora, I don't know. [00:51:28] Speaker B: I don't know. But it's going to be built for agents instead of human. So, like, that's. That's really interesting because then that will fragment the Internet. Realistically fragment. The Internet resources will get diverted to that. Right. Imagine you can't upload. Imagine you get limited on how many videos you get to upload because we're literally running out of hard drives to store data. And there's more profit to be made giving AI the ability to store data on the Internet versus giving humans. [00:51:58] Speaker A: Well, I think what's more likely is, like, there's going to be all kinds of parts of the Internet that you can't access without an AI. [00:52:06] Speaker B: Right, Access, Correct. [00:52:08] Speaker A: You're like, oh, that information's over there in that thing. And the only way to go in there is, like, your bot has to go there. Like, you're like, it's. It's gonna be like, digital areas that are like, bot only. [00:52:17] Speaker B: Yeah, that's right. Yeah. Like, correct. Yeah. Like, I mean, like, into. At the early days of the Internet, you had to have, like, what AOL account to get online? Was that. [00:52:29] Speaker A: Yeah, it was complicated. [00:52:30] Speaker B: Right, right. You had to, like, have a. Subscribe to AOL to get online. [00:52:35] Speaker A: All right, so how do I close the show? Should we talk about, like, a couple announcements? Like, or at least I have a couple of announcements. [00:52:42] Speaker B: Let's do it. [00:52:43] Speaker A: Yes. First, I want to say thank you to all the listeners. We're approaching a hundred thousand views. We have 90. I'm looking at the thing right now. 91, 698. Incredible. I can't believe it. [00:52:56] Speaker B: Crazy. [00:52:57] Speaker A: You know, when you and I first started this, like, like, everyone you get on there, you're like, you hope you'll find an audience, but, like. But it's been fantastic, and the growth is excellent. Really excited about it. And, yeah, it's just been super fun. And so I want to keep doing it. I'm here for the next 100,000, hopefully to 1 million views right there. [00:53:18] Speaker B: Exactly. And then we've been doing some stuff with our landing page on YouTube. Want to talk about that? [00:53:27] Speaker A: What, the new design? [00:53:28] Speaker B: Yeah, the new design. I think it looks amazing. Yeah. [00:53:32] Speaker A: I updated it, like, okay, so we were doing well. So I actually think the idea is, like, we're gonna maybe do. So how about this? [00:53:40] Speaker B: Yeah. [00:53:42] Speaker A: We want to launch a new section where we do videos that are more like tutorials and how to and. Oh, where'd you go? [00:54:00] Speaker B: Hey. [00:54:01] Speaker A: Ah, you're back. Okay. I was gonna say 1, 2. New section where we do tutorials, how to's playbooks. And I'm gonna call it, or I want to call it Gregory and Paul Show Startup Wisdom. So we're just gonna go into a. Pick a topic and so YouTube is one of them. I want to go into like how to create a YouTube channel for your B2B brand your startup and talk about our journey from zero. Like we literally just started uploading videos one day and experimenting and grew it to 100,000 views. And in terms of the, in terms of the, the videos. Yeah, like we were using that, that Ghibli AI stuff, which I thought was kind of fun, the beginning and it was kind of a joke. But I'm pretty convinced that I don't have the data yet. But I'm pretty convinced that like real people, real images, we'll just perform better in terms of thumbnails. So we updated them, updated them all. [00:54:55] Speaker B: I take those shocked looking thumbnails that perform so well on YouTube. The open mouth, that one, I hope. [00:55:05] Speaker A: I really don't, I really don't want to do, I really don't want to do that. I did, I did like the one with, with the three of us, Ross and Ross, old glasses. I was like, oh my God. This podcast should be called Three Nerds with Glasses. Yeah, it's like three dudes with glasses. Like that. That's the criteria. The other thing I want to share is the event. [00:55:29] Speaker B: Right. [00:55:29] Speaker A: I'll put this up on the screen. [00:55:32] Speaker B: That's coming up in March. [00:55:35] Speaker A: That's right. [00:55:36] Speaker B: I'll be there. [00:55:37] Speaker A: And so I've got lots of people signing up. It's going really well. [00:55:42] Speaker B: How many, how many on the waitlist? [00:55:44] Speaker A: That looks pretty good. [00:55:45] Speaker B: Looks great. [00:55:48] Speaker A: So let's See now. Wednesday, March 11th at the AWS Builder Loft in San Francisco, which has been so nice of them to help us with this. [00:56:01] Speaker B: Yep. [00:56:02] Speaker A: We have 173 people now. [00:56:04] Speaker B: 173 going. [00:56:08] Speaker A: Yeah, 173. And they've got fireside chats with some well known VCs. Yeah. OLIS and Charles. Charles was the president of the Venture Capital Association. I guess it's nationally. I don't know if it's global. It must be nationally. So they'll both be speaking. We'll do a fireside chat with them. And this summer, something very different. We're going to have a pitch competition. We're going to let five startups get there. With 90 seconds and pitch and I'll be on judges. My buddy Arjun will be one of the judges. We are looking for another judge. I'm looking for someone really interesting in the startup world. Possibly and probably not an investor because investors, it's kind of a conflict of interest for them to be on the panel. It just doesn't necessarily work for them. So I'd love someone who is a startup enthusiast who, you know is in a really interesting position. Like I am. Right. Like I run a newsletter. I'm not an investor. I don't plan to be investor. I just love startups. I'm technically not a founder either. Right. I'm just a media guy. I'm just a guy who loves startups. So people who do or in a situation like that, I saw or I went to an event where the CEO of Product Hunt, one of the people. So someone like that would be. Would be interesting. Yeah. And then we'll, we'll. We'll have five people pitch and then, yeah, we'll crown one the winner. And we've got just a ton of great. So a ton of great people going. So here are the investors going big list now. The founder list is just off the hook. Crazy. This time we've actually got some people from large enterprises. We didn't really have this, we didn't have demand at the last event for this, but someone from Tesla, Microsoft, Meta, Google, figma, arm, all will be attending the event as well. So, yeah, just go to Luma Apply. You know, ideally you're a founder or an investor or work at a large. Work at any investment firm. That's the ideal fit. If, if you're not one of those people and you still want to go, you can apply. Maybe just do something interesting on your LinkedIn page or in your submission that caches RI. That explains like why we think or why you think you'd be a good fit. [00:58:34] Speaker B: Awesome. So apply today. Luma.com [00:58:41] Speaker A: yeah, the link's on my Twitter. Link is everywhere. I don't even post the link anymore. I just say like DM me for the link, just. [00:58:50] Speaker B: Or Google Viper SaaS founder mix two. [00:58:53] Speaker A: There you go. Does that come right up? Yeah, I'm sure it comes right up. [00:58:56] Speaker B: Yeah. It only does this the first one. Right. [00:58:59] Speaker A: Dude, that was fun. Any else, Any other announcements? [00:59:01] Speaker B: No, I think we got it. [00:59:04] Speaker A: Okay, thank you.