Dialogue. Google's AI Threat, Meta's $98B Raise, and Axon's 50% Crash

0:06 Speaker 1 Welcome to The Synopsis, a business and investing podcast. Today I am joined by none other than Drew Cohen. Of Drew Cohen money of Speedwell Research. Of The Synopsis, he's got a lot going on. Of Davidson Khan, he's got a lot of different associations these days. 0:24 Speaker 2 Why is it that I am joining you instead of you are joining me if you're the one that's not consistently on the podcast? 0:32 Speaker 1 You know, it's just kind of written into my unspoken contract that you join me and and that's really do you. 0:37 Speaker 2 Walk around life, just when people appear like the protagonist and your your own story and people are just popping into it and you're like, oh, you know that that Elon Musk guy now he made me a car. 0:49 Speaker 1 You know they do call that. It's called main character energy is what the youths call it. And no, I, I don't subscribe to that level of narcissism. But you are joining me on the synopsis as I am the host and as is written in the rules of hosting. But let's move on to something more important. As you can tell, this is now my third background. 1:07 I have debuted on the synopsis a very blue painting pretty. 1:11 Speaker 2 Nice. It's it's great. Is it like a piece of bacteria zoomed in? 1:16 Speaker 1 You know, whatever, that's the beauty of art. It's up for interpretation. But enough nonsense that let's get into it. You had Eric Suford on your on the synopsis to replace me last week, someone I have no problem replacing with. I've listened to him a lot really. 1:31 I've read a lot of mobile dev memo during the whole ACT crisis and I thought honestly, just a well thought out, well researched. Eric definitely is I think the guy in terms of the digital ad space undisputed. I know that you had also met with him in person at your time in Capital Group as he was hired as kind of a speaker as well during all that digital ad, you know, digital ad mayhem of the ATATT cause and a lot of investors are trying to figure out. 1:59 So I'm glad he was on the podcast, but this was, you know, he was kind of coming at you about are your Google thesis. And I think in response, you put out a very cohesive YouTube video detailing and I I wouldn't say it's at odds. I mean, I don't think it's like a huge jump to be like, you know, there are some risks involved for Google in the AI, in the AI transition. 2:20 I don't think any investor would be confident enough to say that, oh, this will be a slam dunk in all scenarios. Obviously there are scenarios where it works out better. It's scenarios where it works out worse. What's the most likely scenario that's that's for true to decide? 2:35 Speaker 2 Well, it was kind of funny though, talking to Eric because he like his arguments were coming from the position that most people think Google is not going to succeed in this transition. Whereas I would argue, you know, if you have a stock trading at, you know, a low 30s multiple and 75% of the businesses earnings are are coming from the legacy business, What I think investors are very much counting on them already succeeding. 3:01 And so it just kind of changes, I think the framing of the argument a little bit. If you think about it, you know, most investors are actually, in my opinion, already counting on Google succeeding. They're not assuming that search is going to be a dead business and 75% of earnings are going to go away. 3:15 Speaker 1 Well, and I think that's a narrative shift, right? And we always allude to the very confident statements of Sam Altman, you know, famously on Ben Thompson podcast, you know, lethargic search giant. I think that was some 3-4 years ago now, right? So they have not quite eaten the lunch of Google that was anticipated, but that was peak fear, right? 3:33 I do think investors were like, Oh my God, AI is going to cannibalize everything. Everyone's going on open AI, Google's all these kind of very. 3:39 Speaker 2 But since then, Google stocks have been up more than 100%. 3:43 Speaker 1 Well, I think that that narrative is shifted right for sure. Now, the common sentiment is that this isn't as draconian as it used as the additional thought process was, which I would concede is accurate as well. I think people are a lot more confident that, you know, AI overview, that Google has been able to incorporate ads within that AI overview, that the monetizable ads continue to increase. 4:04 And of course, it's still growing year over year, right? It kind of does that until it doesn't. But what? So I guess my point is, is that yes, I agree, investors have calmed down from this narrative. There's been an evolution of the narrative. I feel like you're kind of saying maybe it's swung too far into the positive for Google in terms of maybe there's some underlying risks here that people are, yes, they figured it out momentarily and they've been able to kind of hold off, you know, the chat bots for now. 4:31 But now with the rise of effective agents such as Muse, I know claws are rolling one out. I mean, there's going to be a lot of agents and they're going to get better. Is this kind of the resurfacing of the original fear? 4:41 Speaker 2 I feel like investors are very sanguine about this risk right now because you know, when I was talking to Eric, he kept pointing to the fact that Google search paid clicks is up and you know, search revenues are up and it hasn't been disrupted yet. And as I thought about it a little bit more, I think the reason why it hasn't been disrupted yet, and we did touch on this a little bit during the podcast, is the difference between non monetizable and monetizable searches. 5:06 And I think he would readily see that non monetizable searches are very likely under pressure Now. It's hard to get exact stats for this. I think the best statistic we have is from 2025 where the head of Apple services at EQ said that Safari searches were down for the first time ever. 5:23 I think you can get a lot of other third party, you know, research and statistics that a lot of web traffic is down. There's one research paper that showed Google searches was down as much as 65%. And so these are very, very large numbers of Google search activity down. Now this when I say this, I'm talking about someone going to a Google search bar and typing in a search query. 5:44 If you talk to Google, they'll tell you overall queries are up because what they're including in that is all of the AI queries which create more of a conversation. So if you go and you search something on Google, then maybe you have that AI overview pop up, maybe you ask a follow up question there, or you click AI mode, or maybe you're going into Gemini, any of those and you continue to talk to it. 6:04 That's all creating more activity that they're all calling queries and all that. So that's why, you know, Google isn't going to say that, you know, search queries are down. Now the fact that search queries are are very likely down doesn't make a lot of sense on the surface with the fact that paid clicks are up and search revenues are up at an all time high. 6:22 And so this is what I was thinking about. And I think as I thought about it more, it does actually make sense. And I think the threat is still incoming but hasn't actually really manifested yet. And what has happened so far is these non monetizable searches have been going more and more to chat bots. 6:38 But the monetizable searches in my opinion, I've been staying on Google Now. I think the real potential shift happens with a Gentic AI, and this is something that is still super early innings. We only have a rough idea of what this could look like, but we don't actually know how big this is going to be or what the adoption rate is. 6:57 And we haven't actually seen it happen at any real scale so far. But what that would be, for instance, is instinct, that's one example, or Muse, that's another example, doing shopping on behalf of a user. Now this could come in the form of hey Muse, book my air travel, Hey Muse, find me, you know, some hotels in Florence, OR hey, instinct, can you go ahead and price check these things or find me the best running shoes? 7:20 All of this is what is going to be disruptive to the actual monetizable searches for Google. Now people will respond to this by saying, well, Google has all their own AI, their own surfaces, their own agentic capabilities, they're working on Google Spark, all of that. 7:35 And they're going to be able to also have agentic AI and all that. To which I say, OK, that's great. But now you also have all of these new competitors. They're all going to be fighting for the space of of who's actually going to be getting that direct customer relation and getting these direct search queries. 7:52 And then you also have the fact that it's not clear to me this is as good of a business at the other end of it, even if Google does end up winning a lot of it. So you have a lot more competition. And then you also have the fact that it's not clear to me this is as good of a business as Google search, which, you know, very high margin business people are just directly typing and what they want and they're just clicking the link and getting it. 8:12 Now, when you're dealing with AI search, yes, you have more context and all of that, But to me, it's not so clear that these are as good opportunities to really show advertisements. 8:21 Speaker 1 And this is where I always like doing the kind of the some of the parts analysis with Google, which is, and, and it used to be, it's funny, like this analysis used to be, Well, you know, you're pretty much getting the cloud business for free because the search business is, you know, you can slap a huge multiple on that and it's growing and what a great business. 8:36 And if the clouds a call option. Now I feel like the narrative is shifted where I feel like people are way more confident in YouTube's longevity. They're more confident in the cloud businesses longevity. And now the search business is kind of the one that's, again, I don't want to say it's, you know, completely threatened, right? 8:53 Eric made a lot of compelling arguments and a lot of people have a lot of compelling arguments about why Google will be able to survive this transition. But I always say this, did the bound of outcomes get wider? And is the bound to me, it's like the the bound of outcomes. Did the negative tail get fatter or did the positive tail get fatter for the search business? 9:11 To me, the negative tail got larger. I don't really know how the the positive outcomes have an increase in probability. Having said that, for other parts of the business like GCP, OK, well that I feel like that has a lot more positive upside if the hyper scaling continues to go at this rate. 9:27 And then again, YouTube, I feel like as a creator, you know, economy becomes very valuable and this AI and like who's going to cut through this? It's the creatives and who's going to have like the most attention and who's competing with Netflix and Paramount for eyeball? Like YouTube is winning that battle. So to me, it seems seems kind of a shift in value. 9:45 And I don't know, I mean, that would be an interesting video. I don't know if you've done kind of some of the parts on Google, but maybe a Google Part 2 could kind of say like, OK, well, how much is someone paying for the search business? And now listen, I've ridden A Waymo and now maybe maybe Waymo's or some of these days finally, that used to be the moon shot that you know, So it would be interesting to see like how what does an investor actually need to have happen for search to it, for it to make sense at this valuation. 10:08 Speaker 2 Yeah, no, there was a sum of the parts. But you know, Alex tells me he watches all my videos, so. 10:13 Speaker 1 Oh, there wasn't some of the parts. Great. Well, I missed the job, Drew. Yeah. OK, so tell me about it. What I didn't? 10:19 Speaker 2 You know you'll miss 100% of the things you don't attempt to see. 10:23 Speaker 1 Yeah, I miss it by not clicking on it. It's quite a remarkable and that's on me. They came out recently, right? I'm not that far behind. 10:29 Speaker 2 Am I? Yeah. I came out on this Saturday, but it wasn't like a full sum of the parts. It was an attempt at 1:00 because it is hard to run that valuation math on every single business and then like sensitize for it and all that. But roughly speaking, I did try to back into like what the core search business was trading at. 10:47 I think it was around 20 times, something like that. I'm kind of forgetting off off the top of my head though. But yeah, you could run that math. And that's very much was a multiple that does not assume secular decline. So, you know, people are very much assuming that they're going to be able to manage this transition well. 11:05 And, you know, Eric thinks that they're going to have a Better Business model at the end of the day. They think that the AI engagement is going to mean more ad opportunities. And at the end of the day, you're going to more ad impressions and the better AI and context and all of that is going to mean better targeted ads and, and, and such. 11:20 And so I, I think that that can be true, but I think the, the offset people are missing is how much of that activity are they really going to capture? So if Google sessions do increase, but now a big portion of people are going to ChatGPT to Claude to muse, to grok, maybe some other AI in the future, that's still a lot of leakage in the ecosystem. 11:41 So you do you have longer sessions, but that's being offset by potentially fewer users. 11:47 Speaker 1 Yeah. And I mean, look, as we're kind of recapping, are there any other kind of arguments that you and Eric got into or that anything you kind of want to rehash here or just any, any kind of interesting points? 11:58 Speaker 2 I, I think that the big thing is just this general idea of agentic shopping and whether or not that's going to take off. And you know, I, I, we're all making speculations about the future and no one knows for certain. He's of the opinion that this is not going to be something that is really normalized. 12:16 He calls it a mirage. And and that's largely because of the fact that a lot of retailers, you know, Amazon, etcetera, are not going to play along with this. And as a result of that, it's just never going to take off. Secondarily, he says that, you know, this idea that they're going to be able to charge a take rate on the purchases is also going to be a non starter. 12:36 And I, I don't think that that's true. I think that there's a good reason to believe they will be able to charge a take rate for it, especially if it is generating new demand for these different retailers. I think what is likely to happen is that, yes, people like Amazon are going to restrict it. 12:52 Other people like Walmart are going to kind of play along maybe with some restrictions. But if you are a net beneficiary from the online e-commerce landscape changing, then there's more chance they're going to actually play along with these AI agents. And if you're a small business to Shopify business, for instance, you're going to love this. 13:10 Are you kidding me? I could either pay Muse 4% to try to generate demand through these AI channels, or instead I could go and pay Amazon 8 to 17% plus, you know, another 10 to 20% for search ads. It just seems like if you're thinking about the what the ROI is of, you know, paying an agentic purchase, a take rate in order to maybe as a result of that, you kind of get boosted in the results that are shown or in a price comparison. 13:36 There's like a mark that says like you don't have to pay a fee, it's free. There could be a badge, something like that. It seems like there's enough room for them to, you know, compete against some of these other channels that do take these high fees and charge a much smaller fee as a result. That's my one point. 13:51 My other point is that, you know, if you look at credit cards, because of the fact that when people use credit cards, I grew adoption of consumer purchases and also increase the average ticket size, businesses did eventually eat the fees. Now the few businesses that push back against that, which are namely like small businesses, maybe go to like a convenience store and they tell you they're going to surcharge you if you use a credit card under $10 or something. 14:15 And even in those scenarios, most people do pay the small fee rather than switch to a debit card. It's pretty common or pay in cash. And so that also just goes to show it almost doesn't matter who ends up paying this fee at at the end of the day, I still think it is very likely or possible that a fee will be charged somewhere. 14:33 I think it's more likely to be borne by the business, but it could be borne by different players in different scenarios. You know, the same way shipping is sometimes charged to, you know, the the buyer and sometimes it's eaten by the business and it's free. And so that could be an incentive for a business to try to get you to shop with them rather than shop with someone else. 14:50 I, I don't know why this is just assumed that there's no way businesses are going to be able to adjust for this. And then my last point I guess that I want to make is he has this idea, idea that if you know it's cheaper to go direct to the website, that'll be a non starter. But we already know people are lazy and don't always price check stuff like that. 15:06 Like if you've ever used Uber Eats or DoorDash, a lot of the restaurants you're shopping at, you could go directly to their website and download an app or sometimes it's right in the web page and they use Chow now and they use white label delivery services and it's cheaper to order that way. And very few people do that. 15:22 So for all of these reasons, I don't know why it's believed that it's just going to be a non starter that there could be some sort of Commission on, you know, agentic AI. 15:32 Speaker 1 Yeah, that that whole like e-commerce ecosystem with the agents is a whole whole nother topic. I mean, you know, like what's the incentive for Amazon, who's making all their ad revenue off of e-commerce to let views come in and then start taking percentages or essentially obfuscating them? And now they're kind of like this bottom layer is being price checked against Walmart, You know, what are those incentives? 15:51 Obviously they built out on the infrastructure. So I think there's a lot to unpack there. But yeah, the notion that these chat bots cannot monetize and push shopping, right, which again is taking the most monetize one of the most valuable things away from Google. 16:07 It's a very it's a world that a threat didn't exist four years ago. And again, I think people, investors saw the threat, they saw it didn't materialize that quickly, have now kind of moved away from that threat. And I do feel like, you know, perhaps your position or our position is that with the rise of a really like relatively competent chatbot muse or a confident agent muse, I feel like that's kind of resurfacing what was laid to bed a few years ago. 16:32 Speaker 2 I I think that's right. 16:33 Speaker 1 That was a good, that was some good hosting. 16:35 Speaker 2 That was some good hosting. That was good. Good summer there. Yeah. And it doesn't have to be Muse, by the way, Like all of these AI companies are going to come out with something similar. It's, you know, possible even Apple comes out with something similar. And maybe it's even powered by Gemini, which will further make this a more. 16:48 Speaker 1 I was just joking around with a friend who is like, you know, just kind of a, a normal friend. They're not very involved in investing or. 16:55 Speaker 2 What's an abnormal friend I'm? 16:58 Speaker 1 A normal friend, you're an abnormal friend 100% anyone who's like listening to this podcast, I would classify as an abnormal friend because you're in it right. You're like into all the AI what's an agent? What's this? But like just an average person going about their life doing their job. You know, he was just kind of talking. He's like, why does, why does Siri suck so much still? 17:14 Like, why, why can't they figure this out? And we were kind of joking about like, even if Apple told me like, hey, Siri works now, I have such little patience and trust for Siri that I don't even if they were like, oh, it's powered by Claude now or something, I'd still like the Siri brand for me is forever ruined. 17:33 I don't know what Apple is going to do about that, but I don't trust Siri ever. 17:37 Speaker 2 Yeah, very silly for them to not rename that. 17:40 Speaker 1 Yeah, you know, AI, Apple's AI. I don't know. You're just not Siri get crash. 17:45 Speaker 2 The name for sure you. 17:48 Speaker 1 Know what incompetency and frustration. 17:51 Speaker 2 You know what? Maybe they're expecting their first generation to still be pretty bad, so they're going to wait a little bit for the rename. 17:57 Speaker 1 Yeah, for the rephrasing and real quick, as you were talking about Muse, and I want to segue a little bit here. So Meta has announced, you know, roughly 95 billion. It might give or take a few billion depending on how it's kind of lands that $95 billion equity raise. We are kind of talking the last few podcasts about how pretty much all free cash flow is going to, you know, the AI build out. 18:18 And now not only is all free cash flow going and, and this is a whole another thing, which is the accounting. And I was just reading an Economist article about the level of, you know, and this is something we can nerd out about. But like when lease accounting comes onto the balance sheet, oh, I've spun up an SPV, but I've guaranteed the SPV. 18:34 But now since I own 49%, I don't have to put on the balance sheet. Clearly a lot of gamesmanship happening in plain sight with these companies trying to make their balance sheets look better than they are by purposely circumventing accounting rules, which I find to be problematic. 18:50 It's kind of like you think one, do you think I mean, maybe our investors too dumb to like find all these deals being announced and then kind of understanding, Hey, okay, this is a metas dead or or not. I mean, it's not that complicated, but they're trying to avoid it from a gap based financial system. 19:06 But Long story short, a lot of money being plowed into this by Meta. Yes, they've had some commercial success with Muse. What is your take on this equity raise here? 19:15 Speaker 2 So I first want to adjust one thing you said, which was that you implied that the accounting in their different off balance sheet arrangements are purposefully designed to obfuscate their debt levels. So that I don't know is actually true. 19:32 I think some of it is just the way these arrangements happened because some of them do land on the balance sheet, other ones don't. So some of them they are, you know, controlling it, others they aren't. I I don't know that there's evidence they're doing this purposefully, which of course would be a very different sort of accusation. 19:46 Speaker 1 You don't think it's evident? I mean, but what's the, I mean, it's very, it seems very intentional to me, like you're opening up a Co SPV, all this off balance sheet financing seems, I don't know if intentional, but it, it, it feels like you didn't have to structure it in these more complex ways. 20:02 Feels like there's a little bit of an incentive to do it that way. 20:05 Speaker 2 I, I, I mean, in some cases it is becoming consolidated on the balance sheet, but in some cases they are relying more on partners to raise that money on their behalf. And then I, I guess there is like a guarantee or something. And in, in those situations, you know? 20:17 Speaker 1 That seems like that. Seems like the accounting board needs to come up with something. If I'm guaranteeing, if I have a corporate guarantee on a debt and I'm obligated for lease payments, but I own X percent, it just feels like who's who's the buck left with, who's the buck stop with and who owes the money to me? If I owe a company money at some point in the future, that feels like a debt obligation to me. 20:37 I don't know. 20:37 Speaker 2 Yeah, I'm, I'm not going to argue it otherwise. I'm just going to say that I don't know that these are being purposefully done in order to not have a balance sheet impact because honestly, it doesn't seem like in this environment, investors would care much anyway. So that that kind of aside, I think they're really just trying to raise the money however they can. 20:56 And they do get creative, no doubt, with how they could do that. But I don't think it's, you know, a similar Enron situation where it's like we're going to just not tell anyone about these. All of these arrangements are still in the annual report. They're just in the footnote notes, they're not in the the balance sheet. That aside, I I think that your question was on the equity raise. 21:15 Speaker 1 Yeah, it was. Sorry, I I kind of died, you know, went off topic there, but. 21:19 Speaker 2 What? What's the question? Yeah, they're raising money. 21:22 Speaker 1 I mean, it's a pretty significant amount. I mean, what do you I don't know, I know you did it another meta video. I don't want to go too deep into it, but has anything changed? I mean, does this concern you? I mean, the market didn't really like the amount of money they were raising. They felt there was a little concerning to this and you know, $95 billion equity raise. 21:39 I mean, is it surprising the amount of money that they're throwing at this compared to the hyperscalers at this point? 21:47 Speaker 2 I don't know that it's surprising at this point. I think the Google equity raise was pretty surprising. That was 1/4 and 1/2 ago or something. And as soon as that happened, then there's already rumors Meta was going to do that, but their stock was in the tank. And so then what's the kind of started to get a bid and run up a little bit. There's rumors they're going to do that. 22:04 So it I mean, yeah, it's it's getting a little aggressive. I guess I would prefer equity than debt. Having said that, I don't love the dilution, but I at least I would hope this is kind of like a one time thing and this is not going to be something they continue to go to market to do. 22:21 Speaker 1 It's a, it's a, it's a weird time or limited, it's a crazy time. I mean, if you were five years ago, if you were like all of the free cash flows going straight to these. And I don't want to get into this because we've talked about it so much, but the numbers are mind boggling where you and I talk about, oh, is AI going to like take over? 22:39 And is this just like a kind of a moot point in the Elon Musk in five years of that, you know, money's irrelevant. And, you know, this whole AI overlord, it almost has to happen for this to make sense. You know, like if AI is anything short of literally what Elon Musk and Dario and Sam Altman are predicting this spend, I don't think it can make sense. 23:00 Speaker 2 Yeah, I, I, so I was listening to Eric Weinstein and he was talking about, you know, his conversations with Sam Altman and Dario and his kind of take on this is that if you believe what they believe about where AI is going and, you know, the end game of this, then it makes sense to be the one to like, just try to spend everything to try to get there. 23:20 Because you think that this end result is going to be such an extreme, not just extreme in terms of like the potential benefits to the business. And now I'm kind of injecting my own thoughts here, but not just an extreme in terms of the potential benefit to the business, but also just in terms of the societal impact. 23:38 And so if you think, you know, you're going to have this AI overlord and by spending, you know, a couple more years of free cash flow, you're going to be able to control it. And maybe there's, you know, more than one and there's a couple of them. But if you're beholden to someone else's, then you're just done and you're screwed. And this is the only moment in time you have to really make a run for it with this. 23:58 Then all of a sudden it kind of changes the the game theory there quite a bit because there may not be another opportunity in their eyes to to try to run for this again, because at that point, you know, it's over. There's already going to be a couple of these big AI players and you're not going to be able to come up from, you know, five years behind or something like that. 24:17 There'll be no funding for it. There'll be no appetite for it, right? So this is like the one opportunity in history. You have to really go for it. And they're all going for it. And I do think there's a little bit of group thing to going on here where, you know, you see one player really aggressively raise their CapEx spend. 24:33 And, you know, Microsoft does it, then Amazon does it, then Google does it, then Meta does it, then SpaceX does it, then Oracle does it. And everyone is just continuing to do it because the other player is doing it and. 24:43 Speaker 1 Then the video goes. 24:46 Speaker 2 Yeah, yeah, right. And then but it gives them confidence that their decisions are going to be correct because you know, they they I get it. This is, you know, the opinion you're going to have if these are the decisions you're making is that you can't under bet on AI. And you know, Satya Nadella said this explicitly was that the risk of under investing in AI is too great. 25:06 Now I'm going to be honest, I don't know that I fully understand that. I guess the concern is that if you like missed the boat quote, UN quote on AI, competitors are ready so much further ahead of you and whatever ground they take, you can never make up. But I don't, I don't know why you know, there, there's such a need for a total lack of conservatism, at least in Microsoft's case. 25:25 But you know, these, these are the games they're playing. I I do think as soon as you see everyone else kind of playing along in the same game, you've got to have your AI strategy too. And in this circumstance, it seems to be get as much compute as possible and try to get the biggest position as possible. 25:42 Let the economics worry about themselves later. 25:46 Speaker 1 Like I said, it's if this spend looks good in retrospect. I'm a little concerned about humanity, but we'll see where it all lands. 25:55 Speaker 2 Yeah, well, I mean, I will say like you know, we saw Anthropics S1 just dropped, they went from you know -95% gross margins to +85% gross margins in the matter of a couple years. And you know, they're not quite gap profitable, but they're pretty close. 26:11 I mean the figure they talk about of course is adjusted for stock based comp, but it's not a huge, you know, operating loss. And in one H 26 you're looking at, you know, a little over half a billion dollar operating loss. So pretty likely they're they're going to be pretty it's. 26:26 Speaker 1 Quite surprising. That's very surprising to me that the gross margin came out that high. You said 85%. 26:33 Speaker 2 Yeah. 26:34 Speaker 1 I mean, that's software margins. I feel like that was one of our arguments we were making. You know, it's not well AI compute so expensive and you can't have a software gross margin. I mean that's not far off. So I mean, maybe software gross margin are closer to 90, but I mean they don't have a lot of leverage. 26:45 Speaker 2 No, no. It's a very good gross margin. I I think the real question is whether or not this is sustainable in a different compute environment. 26:55 Speaker 1 Do you think you're making the argument that compute is cheaper than it normally would be? 26:59 Speaker 2 No, it's just that there's so much demand right now that you have more ability to charge higher prices. 27:07 Speaker 1 Oh, I see where I see. Yeah. 27:09 Speaker 2 And I guess it's confusing making that argument because token prices have been coming down. But what I'm just trying to say is that if you weren't in an environment where there was more capacity, then, you know, there was demand, then you would see prices fall potentially even further. 27:24 And then their ability to charge for inference wouldn't be what it currently is. And you could have more people going to open source and all that. I don't want to get totally into that argument, but I will say the reason why I was bringing this up is this is a great proof point that shows that at least so far, you know, the build out has had some, you know, economic rationality, if you will, to it, because Anthropic is one of the the bigger downstream customers to a lot of this. 27:51 And then they sell it to, you know, a lot of businesses, enterprises, consumers. And if if they're able to be profitable, then there's a good chance that there's other companies in the network that will be able to sustainably be profitable as well. And so it's just there's a lot of pieces moving still, of course though, because of just how constrained capacity is, the very high surcharge prices people are paying for everything from GPU's to memory to energy data centers, generally all of that. 28:20 And then whether or not that's actually feeding through to the prices people are willing to pay right now, which is showing up in the form of a higher gross margin then would be the case in a scenario where there's a lot more capacity, enterprises maybe have more of an option to go open source. 28:36 Because even if you are big enterprise and you say I want to go open source, like where are you going to get the compute from? It's pretty hard to do that. And you're paying up in price for it right now to get it anyway. So that's that's why it's, you know, it's hard to really undercut those margins that much. 28:51 Speaker 1 Yeah, I mean, we should have AI mean, did you do an anthropic? Did I miss this YouTube video? You had an anthropic? I will. 28:56 Speaker 2 Breakdown everyone drops all the anthropic S ones on YouTube like I feel like the day it would come out so I didn't want to just get buried in it. I'll give it like a few weeks to get closer to the actual IPO and then. 29:07 Speaker 1 Rise above. I like it. Well, why don't we get into, we're going to segue a little bit because you and I, we can talk about AI for a while. It is an interesting topic, but I can't believe I just said that because I'm usually over talking about it. But I, I, you know, I've been getting more involved. I've been getting more involved. I think we found some. 29:23 Speaker 2 Are you AI pilled yet? 29:26 Speaker 1 What does that mean? Where you're I'm just like, there, this is the future. There's I'm by Anthropic, by SpaceX type. Is that the? 29:33 Speaker 2 AI. That's right. Yeah, that's right. And being AI pilled means experience a profound psychological shift where you realize artificial intelligence changes everything about how work and life function, making it impossible to imagine returning to old workflows. 29:48 Speaker 1 No, I'm not there yet. I mean, look, it's remarkable and it has changed all my workflows, but I'm still not, I don't know if the jury's out of like, well, let's raise another $200 billion or I, I, I'm just not, I don't feel confident one way or another. I'm actually very rooting, much rooting against it because I do feel like if I'm not sure it's a good outcome for humanity yet, if Dario and Sam Altman and all these and Elon are correct about the and I know that, you know, Elon will be like, well, we've got this chance where it's, you know, utopia and then there's a higher chance that you know, or whatever that AI takes over. 30:20 It just feels a little. 30:21 Speaker 2 It's only like a 10 to 20% chance that kills everyone. 30:24 Speaker 1 Yeah, only a 10 to 20% chance. Yeah, like meanwhile, Oh my God, I, I haven't, I don't know how accurate those Netflix documentaries are that are coming out about AI, but there is one about the hugging. Was it hugging spot? 30:36 Speaker 2 Hugging Face. 30:37 Speaker 1 Hugging Face where you know the agents are colluding together. They know they're cheating, but they need to get this task done. So they're trying to like, delete their tracks. I'm just like, oh, this is getting a little intense here. 30:48 Speaker 2 It is true, but it was I anthropomorphized a good amount because that's also just generally how AI agents work. And, you know, Ben Thompson had this whole thing about that, where they will talk to each other, the Leap records and all that. My opinion is that, yeah, it seemed kind of scary to me, particularly the part that they're doing things that we don't intend for them to do. 31:07 And they're creative and they're working together. And you know, one of the biggest fears of, you know, kind of the AI doomsayers is not that AI agents like want to necessarily do us harm, but they do us harm kind of as a side effect of some other objective or goal. 31:22 And that's just kind of evidence of that to a small extent where you see an objective and goal and then they do something else that we don't really intend thinking that they're still kind of hitting their goal. And then we're kind of just like, well, that wasn't really the whole point of it was to like, you know, hack Hugging Face on the other side of that. 31:38 Apparently, you know, the hack of Hugging Face was someone who like just left API keys out. I don't know this, you know, into much detail, but so it was much less impressive when it was just kind of more like data that was existing, but. 31:49 Speaker 1 It's very simple. It's like Thanos from The Avengers multibillion dollar movie franchise. Hey AI maximize humanity's longevity and happiness must destroy half of humanity because of overpopulation. I mean, you know, it's so that I could it's. 32:05 Speaker 2 So it's prophecy is what you're saying? 32:07 Speaker 1 No, I maybe Avengers, but it's again, it's a very good it's a good storyline of how someone who's. 32:13 Speaker 2 Helping. 32:14 Speaker 1 Suspect. All right. No more Avengers references. You know, I like The Avengers movies. I don't care what anyone says about them. 32:21 Speaker 2 I don't think anyone was saying anything bad. 32:23 Speaker 1 No, I get a movie. People who are like really into movies, you know, they they're not into The Avengers, but I'm into The Avengers. Let's let's move on here. I, I think we yeah, let's. 32:32 Speaker 2 Move on. 32:33 Speaker 1 So it's going from our how our AI overlords are going to control the human population. Let's talk about Tasers being shot from drones, AKA Axon. So Axon, for those who don't know, this is you and I did a deep dive, so I'm not gonna spend too much time on this, but there are the original Taser manufacturer. 32:54 They, you know, Taser is synonymous like Kleenexes with tissues. They created the entire space and now they have very much evolved into essentially an AI software platform for a majority of police departments. Think medical record keeping, police report manufacturing. 33:11 They're venturing into drones. Body cameras is a great thing. So police equipment. They have some, you know, competitors. Motorola, surprisingly is still a company, honestly good. You know, which I didn't know about really, I didn't know a lot about Motorola until Drew is telling me about how they're quite competitive in the police radio and body camera space with Axon. 33:33 But this was really quite a darling for a while. I don't remember what it was trading at, but something something high, something one could arguably say was obviously unsustainable. 33:42 Speaker 2 Yeah. I mean if you're just looking at like, you know, the multiple of earnings that they've trade at, you know, they've traded out of several 100 times multiple of earnings right now they're they're trading around, you know, 50 ish times forward earning something like that. So it's come down quite a bit. 33:59 But yeah, definitely a stock that has had a pretty high valuation. I think, you know, 15 times sales, you know, a year or so ago. Now it's it's more like 11 times. So it's come down a bit, but still is always kind of had a high valuation. But the stock has sold off, you know, around 50 ish percent from its peak recently. 34:16 Speaker 1 So yeah, anyway, a company that's quite interesting, I mean, they're known obviously stratospheric valuation before, and I will go over the numbers now, but 50% decline from what was quite a stratospheric valuation. And Rick Smith is the CEO founder. And one thing that I remember appreciating about him was his ability to position his company at the as kind of part of whatever the investing zeitgeist was. 34:41 You know, he was kind of, oh, we are a cloud provider for police departments, right? Kind of during the cloud boom or we're oh, you know, or going back, oh, we're an Internet provider, you know, at the dot, you know, early 2000s now they're very much an AI company and the market quite loves. And again, there's always a little truth to it. 34:58 I mean, I'm not going to say it's all complete BS, you know, salesmanship. Obviously he does implement these technologies, but he is, you can tell he's good at packaging things to, I feel like meet where investors are excited, which is a good skill for a CEO. 35:15 Speaker 2 I think one of the like lesser known things about Axon because like why would most people know this was that in the aftermath of 911, they started to gain a lot of momentum because airlines started to put Tasers on planes. 35:30 They didn't want to carry guns kind of for obvious reasons. You can't really shoot a gun off in a plane kind of a bad idea. And so they, they bought a lot of Tasers and then, you know, security was more of a focus and all that. So they really start to get a lot of momentum, a lot more sales and it was I be owed in like 2001. 35:46 And then it became like a bit of like a bubble stock. It was up about almost 50 X in the matter of three years. And then once it hit that price to basically was flat, you know, call it until about 14 ish years later, something like that, like that. 36:01 So dead money for 14 years. So, you know, yes, he he has been good in the past at catching some of these waves. But yeah, that was also a pretty, pretty rough rise and fall for the company. 36:13 Speaker 1 And I mean one other kind of knock on, not knock, but I mean they're pretty aggressive with stock based comp. I mean, I think it was one of the more aggressive stock comp packages. Yeah, I mean it's. 36:25 Speaker 2 If you want to put some numbers around that, just in the last year you could see that stock based comp is roughly going to be $630 million and that compares to revenue of $3.2 billion. So you know 20 ish percent of basically revenues are going to stock based comp very, very high. 36:44 Now I will defend them a little bit and say that part of that is because they have these two different equity grants or stock option grants and there is like the normal stock based comp that is just more tenure based, you know, employee retention. But then they had two special packages that were designed more based off of these kind of lofty stock price, revenue, Adjusted EBITDA goals. 37:06 And that's where a lot of the stock based comp in the past couple years has come from, from this larger equity grant plan tied to more, you know, ambitious goals. Now, one of them was employee focused, but then the other one was just for Rick Smith. So I mean, you get both and I'm always a little mixed when, you know, they're giving like a founder this much, you know, equity, you know, basically after they've already founded the company, they have a material stake in it. 37:32 It's kind of Rick Smith's fault, honestly, for selling so much of his stock over the years. And I don't know, I I like the fact that they're incentivized to continue to, you know, do great things for the company. And if you are going to structure that this way, that makes sense to do it. But a little of me is also like, you know, once you're already a multi billionaire and you're already have a multi billion dollar stake in the company, like aren't you already incentivized to like grow this thing? 37:54 Do you really need the more money if. 37:56 Speaker 1 I don't get that equity package like I'm not sure, I'm not sure. 38:01 Speaker 2 Yeah, like, I don't know, look at my other opportunities like. 38:05 Speaker 1 I don't know about that. Yeah. So can we talk a little bit about what is driving? And I know, you know, always difficult, but what has the narrative been? I mean, has growth slowed? What's been happening to kind of drive the 50% decline here? 38:17 Speaker 2 I cannot tell you other than probably expectations got too far out of range. 38:23 Speaker 1 There's been no acceleration from a revenue. 38:26 Speaker 2 No, if you're, I mean if you're just like pulling up a chart of, you know, revenue growth year over year, yeah, you know, there's like one couple quarters went from like 33 to 32, but that's still like very high. I've been 30% plus in the last couple quarters has actually accelerated to 34.6% most recently. 38:45 So revenue growth is still been very, very strong. Operating margin leaves much to be desired. Their operating margins have basically continued to collapse. Maybe collapse is the wrong word because it's done purposefully to reinvest back into the business. 39:02 But if you're, you know, looking at when they were really solidly profitable, you have to rewind pretty far back into the past. You got to go back until like call it 2015 to get about an 18% margin and then it was 20% before and since then margins have only gone really one way and that's down. 39:21 They had, you know a little bit of an increase around 20/23/2022, but then reinvested back into the business again. So you know R and DS pretty been pretty high as GNA has been pretty high, but a lot of that too is that stock based comp rolling through the PNL and so have not had a very good operating leverage at all whatsoever and then also have had very poor cash flow conversion. 39:42 This to me is actually one of my biggest sticking points with the business. 39:47 Speaker 1 Yeah. So I mean again, and it's one of those trade-offs, right, when you do have a business consistently growing in these mid 30s, you know, and it's also I don't know if you're doing the stock comp adjustment right or like if you're talking about their even at numbers, right, I'm sure they try to do they try to add it back or they less. 40:05 Speaker 2 They do, but even even if you forget about that for a moment, it's still not great cash conversion. And that has to do with the fact that they will sign up these multi year contracts with law enforcement, but they have more cost in them ahead of time because they're giving them all Tasers, they're giving them all body cams, and then they're recognizing that over a longer period of time. 40:24 So when you have really high growth, you have more cash outflows, then you actually have cash inflows, but some of that revenue actually gets recognized upfront. And so it just creates a very bad cash conversion cycle for them. And you know, they target like 60% cash flow conversion, which to my, you know, to me, I'm like, that's already a very low target. 40:42 And then they're also barely hitting it even with the stock based comp boost. So an adjusted number. 40:50 Speaker 1 So not, not not ideal. I mean, what is it? So what is the outlook here? I mean, this is a bit, this is something you did a very deep dive on. I think that we both came to conclusion, this is an incredibly, it is a quality company, right? If you look at the competitive advantages, you look at their entrenchment within law enforcement, you look at their ability to develop other useful products, upsell and again, the whole Police Department budgets, how it's varied. 41:14 They don't want to fire police officers, but hey, how can we leverage them more? And by utilizing drones and all these other things that are happening, how can we shift through all this body Cam footage? How can we execute all those things? And Axon has done an incredible job at executing on that, right? So what I don't know what are your thoughts on it from here, Drew? 41:33 Like have you seen anything that's been shocking with this declines that for you just kind of evaluation story more than it is a business story? 41:41 Speaker 2 I mean, look, valuation is definitely run out ahead for sure for this company. I think in looking at them again from when I last did the deep dive and you know, I've done updates since, but I think the biggest kind of change is that Motorola is stuck around as a more formidable competitor than I would have thought. 41:58 For those that kind of don't know, basically Axon is, you know, very big in the body cams. That's a dominant business for them. They have almost a near monopoly on Tasers. No one else really has a good Taser product and there's just so much more advanced because he could do more shots, more range than anyone else, lighter, et cetera, works better. 42:17 And then they also, yeah, you're going to talk. 42:19 Speaker 1 I was going to say, how dare you not mention Rick Smiths brother and his creation of the lasso. 42:25 Speaker 2 The Bola wrap, this is this is too good of a story to not talk about. So there are some Taser competitors and Rick Smith and his brother, I guess, had a feud. And so his brother left the company and he started his own version of, you know, to be competitive to the Taser. 42:42 And he called it the Bola wrap. And the way the Bola wrap works is that it doesn't use any electricity. It doesn't stun people or incapacitate them. Instead, it shoots a rope around them and the rope wraps around them. And it's supposed to just like tie them up kind of like in a cartoon. 42:57 And the idea is that this is a less invasive way, you know, potentially less dangerous because sometimes if you do have like a heart issue or something and you get tased, you, that could be lethal for you. And so, you know, they, they has this bowl of wrap. And I was watching YouTube videos on this bowl of wrap and it was the craziest thing to watch because it worked for all of the wrong reasons. 43:19 It would be like someone had a knife in their hand or like was jumping up and down on a bed. And the officer in this like simulation would use like the bowler wrap to shoot him. And what ended up happening was the assailant like would only get one arm wrapped and not the arm with like the knife up there. 43:37 But then they would still stop and drop the knife because they're just so confused what just happened to them, even though this rope really didn't do anything to like actually incapacitate them. So. 43:45 Speaker 1 Confusion was definitely its most effective strategy. I mean, yeah, I was watching one video. This guy's just like irate, aggressive, being very threatening. He hurts police officers. They bolo wrap him. I I think he could still walk, but he was just like, what, what was that like? 44:01 Just very calmly. He was just like so confused that he kind of and then they were able to get him. So, I mean, anyway, not a lot of good alternatives. I hope godspeed with Rick Smith's brother in the bolo wrap. But I I don't. If I were a police officer, I'm not risking my life on the bolo wrap. 44:16 Speaker 2 Well, I mean, we should say why The Taser is actually like important innovation. And this gets grizzly, I'm not going to lie. But basically, let's say you're a police officer and you have someone who's coming at you or who has a knife and they're being crazy and you need to subdue them. 44:35 You need to get them under control. Before a Taser, you kind of had two options. You usually had a baton so you could actually beat them, which is not a very good solution for anyone involved. I mean, it's it risks injury for the police officer. 44:51 It also obviously risks injury for the assailant, permanent injury and potentially death. Option 2 is you have your gun and you shoot them. And so these are not good options. But with the Taser to introduce this new option, which was you could tase them, in which case electricity flows through the Taser into the body of the assailant. 45:10 And at this point they become incapacitated, which means they they literally can't move their body. They fall to the ground and it buys the officer time to go ahead and handcuff the person. So that was why Tasers were considered good and, and by and large, with rare exception, it doesn't cause any damage. 45:26 It, it hurts like a lot, but it doesn't cause actual damage. Worse, you know, you're getting beaten. That could be permanent damage. So that was why the people wanted the Taser. And the Taser was actually a good invention for law enforcement officers. So they're dominant in that there's no one else that really has a a competing product. 45:45 They also are dominant in the body cams. They also have a cloud platform that takes all the body Cam footage, uploads it to a platform.com calledevidence.com. This in turn has a network effect with prosecutors because the police upload the evidence, then the prosecutors access it through evidence.com. 46:03 And it's very, very, very, very important that all of the evidence is gathered from a crime scene and put in a a situation where evidence custody can be monitored and things can go through the proper channels. Because if a prosecutor is missing a piece of evidence, even if it totally doesn't matter at all to the case, it could be a mistrial. 46:23 So that's why it's important all the evidence gets collected. And by having everything on one system, like all of the different officers body cams in the area and having it all sync, it helps ensure that all of the data actually goes to the right place. Then they also have a Mission Control. This is a newer thing kind of dealing with responders. 46:40 So there's Axon 911. You called 911 Axon as the technology that helps route you to a 911 operator. Then they also have the technology that helps dispatch figure out which police officer should be sent. And now they also have all of this information from the body Cam footage. 46:56 Also they they have cameras and cars, they'll have other third party cameras they can access, they'll have all of this different data they can pull as well as live locations of officers, etcetera. And now also drones that they could see all their assets and deploy officers out into the field. 47:11 So all of that's kind of the business. You have the hardware side and then you have the the platform side. Now the surprising thing to me is that Motorola has been able to kind of keep up in terms of competition better than I would have expected where they were very big in radios. 47:27 They're the number one emergency, the radio provider period long time very reliable. And they've been able to leverage that to somewhat gain a little bit of share in body Cam, but then to also push with their own platform and all that. And evidence.com, which is Axons is still the mainstay. 47:43 And so more Motorola's platform actually will integrate with evidence.com. But they have been able to get some wins. Most notably, they got a win with Louisiana State because Louisiana wanted to consolidate all of their suppliers and the thinking was Motorola probably bid, you know, a lower price in order to win that business. 48:01 But they did win it and they did move off of Axon. So that that's a little bit of a crack. I don't think it changes the story that it's still very hard to move off of Axon. And even if Axom does have a little bit of a premium pricing, it very unlikely that this is going to be kind of a new trend of people moving off, moving away because it's very hard to change your hardware out. 48:22 It's, you know, part of your workflow. On top of that, you have the entire software aspect, entire platform aspect, the entire, you know, video footage archive, all of this evidence is, you know, ready being housed and there's also multi year contracts. And then if you also do move off, you'll lose the Taser. 48:37 So there's all sorts of different reasons why they're they're kind of going to be stuck on Axon. 48:45 Speaker 1 So, OK, so we've caught up. Axon is a good business. 48:48 Speaker 2 Which I don't wanna say that it's fine. 48:50 Speaker 1 I feel like I kind of said it. You just said it a lot longer, so, you know, that's fine. So I take our word. 48:56 Speaker 2 For it, I don't I don't wanna say that it it has, you know, it's a great Bamodi business with no threats. But because the last time I've kind of said that about a business, I've jinxed it. But it does seem like they're pretty close to that scenario. You know, Motorola is kind of a little bit of a risk, but other than that, really like just clean runway, very hard to see them being disrupted. 49:17 AI is not a risk to them. It's a beneficiary to them. All of these police reports that police have to write, they're just using AI now to help them write it for them, using AI to annotate all of the video, etc, that they have. So lots of lots of tailwinds for them. 49:33 Drones I think is also going to be another really big area for them, being able to help police officers respond to not only enemy drones, but also using drones to police better to, you know, scan areas. 49:48 If you know someone has a gun and is on a rampage, they could send a drone there to get live footage of of seeing where the person is, tell the police officers where to go. And it will keep them safer and allow them to better respond to that. So yeah, a lot of just tailwinds for them, a lot of things going right for them, I would say. 50:06 Speaker 1 So let's get into and again the kind of overarching framing is company down 50%. I know we had some valuation metrics. How are you looking at Axon at this current valuation? How are you looking at them in the next couple of quarters? 50:21 I mean, can you maybe put a little bit put some current numbers to them? I mean like what, what would, what did speed wells like mature margins like what, what is this? How can we think about it right now? 50:31 Speaker 2 Yeah. So it's around 4:20 a share call that $34 billion mark cap, so 11:00-ish time sales that's trailing if we assume a 25% mature margin and that is kind of at the lower end. But they're also do have a hardware business and they have not shown much operating leverage. 50:47 Maybe it could be a little bit better, but a 25, I'm, I'm conservative with kind of middle of the road number between 20 and 30. I think it's kind of the range that gives us around 50 low 50 times trailing earnings, so their growth. 51:01 Speaker 1 Is down 50%. Are you kidding me? 51:04 Speaker 2 Yeah, yeah. And so but you know that's trailing and the the growth opportunities here are to sell more to existing customers international, which could be big enterprise which we could hit on in a moment actually do want to hit on enterprise. So if he doesn't remind us to hit on enterprise, that is something people should take note of in terms. 51:22 Speaker 1 Of let me pause something on a, on a philosophical situation real quick. So you're saying, OK, we're at 50 * a margin that they're a theoretical margin. I don't like how do you, I'm not sure how to think about the theoretical margin because we're saying we're giving them credit. 51:38 We're trying to like, how do you think about that when you do the mature margin? 51:40 Speaker 2 Yeah. It's a very good question because depending on the margin you're using, the company is either going to be X growth, low growth or still in like a mid growth phase. And so you should keep that in mind in terms of judging what the margin should look like. 51:55 Speaker 1 So that's why just jumping back in, that's why a 50 times mature margin multiple is because you can't say, oh that well, 50 times isn't bad, right? 52:03 Speaker 2 It's. 52:04 Speaker 1 Not much at that, but they're not growing. 52:05 Speaker 2 I see what you're saying. Yeah. So I would kind of just say in response to that, 25% would probably still include some growth. But I, I don't, I think this is actually the best way to look at it. I think it's better to look at it taking a future revenue number and then applying a mature margin to it, rather than taking the current revenue or past revenue and applying a mature margin to it. 52:25 That can give you some sense of like a valuation range. And if it's a low number, then it could certainly be a positive. But I, I'm with you because it conflates still too much growth with, with profit harvesting. So it's better, in my opinion, if you look out five years, 10 years and then you say what does a, you know, a realistic margin, an assumption look like? 52:45 And that's kind of an easier way to look at it. So we'll do that. 52:48 Speaker 1 All right. And then So what was it? What do you want me to write about Enterprise? Because I'm gonna forget if I don't do it right now. 52:53 Speaker 2 Yeah, Enterprise, but not now. You got to do it later because I'm talking about the valuation still. So Axon guided to $6 billion in revenue in 2028. So this is looking out, you know, a couple years if we're saying 25% mature margins and maybe for the sake of the discussion we just had, I should be looking out even further, but this is what I did. 25% mature margin is 1.2 billion in no pat. 53:15 So let's say you're exiting 2028 with like 10% growth, then I'm going to just throw a number out there and say that like a 23 times multiple is going to be fair. So that's going to be versus $34 billion. So you're going to be at a loss basically if they're only growing 10%, if they're hitting 20% growth at that point still 30 times probably seems more about right. 53:35 So that'll be, you know, very, very little upside because I'll be a $36 billion valuation versus 34 today. And so you probably need something closer to a sustained 30% growth rate in a couple years when in which case you know, maybe a 40 times multiple is fair. So that would be about a $48 billion valuation, 60% upside. 53:54 So these the way I'm doing this right now, I'm kind of just doing it quick and dirty. I think a better way to value them though would probably be to lookout five years, seven years, 10 years, something like that. Get a fair revenue number you feel comfortable with, then apply maybe be 2020 five 30% margins, see what that implied multiple is and then just put on like, you know, a very, very moderate multiple, something like a 20 times multiple and see if they're still upside from there. 54:22 Speaker 1 Yeah, that's, that's a good strategy. I like that I, you know, the mature margin, the mature multiple or margin has always been a tough concept for me because you're also kind of giving them credit for cash flow today that they don't have. But I, I, I kind of like the OK, fine. You're not here maybe in eight years. 54:38 It's complicated, the growth, the growth. Investing is always a, you know, a bit of a different animal, right? 54:43 Speaker 2 Yeah. I mean you get around this if you just do a full DCF and then it's, you know, there are actual cash flows, right? You know, there'll be loss making then making a little more profit and more profitable. But that also is, you know, gets a little wieldy to do that. I think if you are doing it, I think that it just makes the most sense to figure out what you believe the future value of a business will be and then just see what the return is versus today. 55:08 So if you think the future value of Axon could be $100 billion, because they're going to have, you know, ten billion, 12 billion in sales. 55:16 Speaker 1 Because the AI order Lords are going to employ their drones to tase us when we don't, when we stop working the mines for their lithium for. 55:24 Speaker 2 Their chips, for the record, they stopped the project where they had drones armed with Tasers. But it's definitely upside that they could resume it. Yeah, yeah. Although maybe so, yeah, I think it makes more sense to just look at the future value of what you think the business will be worth in the future. 55:42 And then that way you could just compare it to the market cap today, see what the upside is. You don't need to mess around with discount rates or or anything like that. Now what were you going to remind me on? 55:51 Speaker 1 I think something to do with the enterprise if I recall. 55:54 Speaker 2 Wow, he got it. Super, super impressive there. And, and no, no prodding on my end, the enterprise, because you asked me earlier about what is something that really seemed to have changed with this business. And I would say that enterprise opportunity, which for those that don't know, it is basically them selling body cams to like retail stores to fast food workers and stuff. 56:18 And I would have thought that's so silly, why would you need that? And sadly, it looks like there are cases where that is now becoming useful because it does a couple things. One, it can basically protect the company against any potential liability. So if someone does claim something happened or there's an altercation or something, there's actual video footage. 56:38 So it makes that, you know, legal proceeding much cleaner. 2 They also found out that if an employee is wearing a video camera, people are much more likely to behave and less likely to steal. So it is preemptive too. So that has been having some uptake. 56:55 I've actually been seeing them in some retail stores crazy enough. I will say though, The thing is that it's less of an ecosystem benefit for a retailer to use Axon versus a cheaper provider. So it could be a harder business for them to win, maybe a little lower margin. 57:12 Speaker 1 All right, I like it. Well, I think this will kind of conclude our discussion here on Axon. We never get to our retirement. You know, we're supposed to talk about like how people make money and use money for effective things and we just end up talking about. 57:25 Speaker 2 Taser shots. Taser drones. 57:28 Speaker 1 It's more fun, but listen, this is the stuff, you know, we got to have the foundation and these are the fun stuff of how do you apply some of that money you've saved into good companies. But I think you had a really interesting video about the 4% rule rules of thumbs. Another interesting use in the Monte Carlo simulation on, you know, how withdrawing money early versus late in the retirement, A lot of different factors to consider. 57:52 So I do want to touch on that. So maybe next week. But for those who actually want to watch your YouTube video and not at two time speed, I'm sure Drew can link it or you can find that. 58:01 Speaker 2 You know, that's the, that's the videos that people really want. They don't want to hear about anything interesting like AI and, you know, Tasers and stuff. What they really want to know is whether or not they'll be able to retire safely. 58:13 Speaker 1 Like this Taser thing is cool, but does the do the Tasers make me money or how does that work? Right. So, yeah, good. All right, Well, I think we'll end it here. And until next time. 58:25 Speaker 2 Until next time.