044 Open AI is Making a Phone With No Apps?

[00:00:00] Speaker A: All right. We're life. [00:00:02] Speaker B: Ready? [00:00:03] Speaker A: We're ready. We're life. [00:00:05] Speaker B: Okay. Well, hey, welcome back to the Gregory and Paul Show. I'm Gregory. [00:00:13] Speaker A: I'm Paul. [00:00:14] Speaker B: And we stream live every Thursday now at 11:00am Pacific. What's that, 2:00pm Your time? Eastern. [00:00:22] Speaker A: 2:00pm Eastern. [00:00:23] Speaker B: Yeah. Yeah. So for today's show, we've got a bunch of exciting topics that I'm looking forward to speaking to you about. What's on our list here? Let me see. [00:00:38] Speaker A: Okay. [00:00:40] Speaker B: OpenAI. [00:00:41] Speaker A: You're excited. [00:00:42] Speaker B: Could be making a phone. Yeah, I'm really excited. The AI first phone, maybe. No application, some leaked information. We'll talk about that. We talk about it, and then we'll jump into the trial of the century. Elon Musk versus Sam Altman. [00:01:01] Speaker A: That's a good one. [00:01:02] Speaker B: We're going to talk about the cost of compute. Wasn't Jensen saying $250,000 for tokens per engineer? [00:01:10] Speaker A: Yeah, it's not as cheap as we hope. [00:01:15] Speaker B: No, it's not cheap at all. [00:01:16] Speaker A: Just getting expensive. [00:01:19] Speaker B: We've got some confessions of a vibe coder who had an agent delete his entire code base. We want to talk about weight loss, drugs impacting food shipments. There's some analysts out there who are saying that's a cool one. It's actually true. There's data on this. We'll talk about. Ooh, the Clipping Industrial complex. I know you're excited about that one. [00:01:43] Speaker A: I'm loving this one. [00:01:46] Speaker B: Google's cloud revenue topping $20 billion. And of course, we'll end the show with Gary Tan's wonderful post giving YC founders basic accounting advice. [00:01:58] Speaker A: Basic accounting advice. That's right. Yeah. What is mri? [00:02:02] Speaker B: Oh, and the. The. Wait, wait. One thing I wanted. I wanted to discuss before we jump into our news items was the pitch competition. [00:02:13] Speaker A: Oh. [00:02:13] Speaker B: So we'll be doing a pitch competition on May 12th at Entrepreneurs first in San Francisco. It's an excellent space. We've got a wide selection of pitches that have been submitted. In fact, tomorrow is the deadline to submit your pitch. If you have trickled in this week. Yeah, I think I've got like, yeah. [00:02:35] Speaker A: How many? [00:02:35] Speaker B: 5, 65 pitch decks, which is a lot. It's almost double what we had last time. Yeah. You know, the first time I did this, someone, I did a couple events for startups and had people demo and someone suggests, like, well, why don't you make it a pitch competition? I'm like, that's cool. So we did the first one. I was like, am I gonna get any submissions at all? Will this be. Will this be interesting to people? And like lo and behold we had a bunch of great ones for the event. [00:03:01] Speaker A: Yeah. [00:03:02] Speaker B: At Amazon this time I have even more and the quality of the submissions is excellent. [00:03:08] Speaker A: Amazing, amazing. [00:03:09] Speaker B: Really fantastic. [00:03:10] Speaker A: Very excited to see who you let through. What are the best startups. [00:03:15] Speaker B: I like to joke that the fact that the competition is so fierce for something where there's no prize money, just my little O. [00:03:24] Speaker A: Well there are some prizes. Yeah, that's right. [00:03:26] Speaker B: That's the trophy. Right. That's what makes it so competitive. [00:03:30] Speaker A: Know back VCs will be there. So there is some exposure, right? [00:03:35] Speaker B: Correct. We have two speakers. So the founder and CEO of Signifier, an excellent early stage investment firm will be there. [00:03:45] Speaker A: Perfect. [00:03:46] Speaker B: And then we have this guy Kyle from Bling Capital, another really active early stage investor. So they'll be speaking and then the list, we have about 35 people from other venture firms. They're scouts. Maybe some corporate development people. Sure, yeah. You know like an event like this, we get a lot of founders. Right. We've got a couple hundred founders registered. I expect to there to be at least 80 to 90 founders there. Investors will show up but they like to perhaps keep a low profile. Of course they don't want everyone. [00:04:21] Speaker A: Of course, yeah, it's like you know, the dating, seeing San Francis, you know, same, same amounts of ratio. [00:04:29] Speaker B: Correct. [00:04:31] Speaker A: So. [00:04:31] Speaker B: So if you do go be nice to people who are investors, not drive them crazy. We want them, no swarming, come back. Yeah, yeah. Oh, I'll give one piece of advice because I've had other people reach out to me and you know, have questions about like if I'm looking for an investment, it's hard to get in front of these investors at the event, like what should I do? And so the simple answer is like you should be networking and reaching out in advance. [00:04:55] Speaker A: Right. [00:04:56] Speaker B: And you should have a process. You should have. Yeah. Or, or just, or cold intro in the sense like hey, are you going to be there? I'm going to be there. This is what I'm doing. Like try to kick off the conversation in advance, make sure that they are in your space, they're interested what you're doing. Investors have a lot of different areas of special specialty. They don't always invest in what you're doing. So try to pre qualify all that stuff rather than just walk around the van. [00:05:19] Speaker A: Right. [00:05:19] Speaker B: You're going to run into somebody. [00:05:22] Speaker A: Has anyone asked you to make intros? [00:05:26] Speaker B: This great question, does this actually come up as like part of the process or something? So, so actually What I'm thinking about and what I'd like to do in the future is this, like, now that I've got a rather large list of founders, investors. [00:05:39] Speaker A: Right. [00:05:40] Speaker B: Yeah. I want to do an event where we would match up, let's say a group of five founders with a mentor and do kind of like a round robin type of thing. [00:05:50] Speaker A: Cool. [00:05:51] Speaker B: Maybe on a weekend or something. Yeah. Very different. This is like a mixer. This is after work. This is pitch competition. It's really fun, it's exciting, it's entertaining. I'd like to do something else where I maybe take all those founders and put them into groups, cohorts, and then give them founders. And that way you would get feedback from like five different founders in a. In a session. [00:06:12] Speaker A: Like a founder focus group. Yeah. [00:06:15] Speaker B: Or just I was thinking more like maybe even have mentors who have different areas of expertise. Right. So someone who's like a product person, who's an AI person, someone's a go to market person, someone's a sales person. [00:06:25] Speaker A: Yeah. [00:06:25] Speaker B: And they'll just give you a different perspective on like, advice, things you could do to improve what you're working on, like how to go to market more effectively, how to build a better team, like that kind of thing. And then everyone have different. A different advice. And I think that could be really good for people because at the end you could sit down and kind of like compare and contrast what you heard. [00:06:44] Speaker A: Perfect. Exciting. [00:06:45] Speaker B: Yeah. All right. All right, let's jump in. Oh, and if you want to apply and you want to attend, go on to Luma. It is called the vibe, your SaaS pitch competition and VC founder mixer. I'm sure you can find it. If not, just send me a message. I'll send you a link. All right, let's jump in. OpenAI could be making a phone. I'm sorry, do you want to say something? [00:07:13] Speaker A: No, no, no. Say more. Jump into. [00:07:16] Speaker B: Okay, so the phone. Okay, here we go. So we all know that Sam and Johnny Ives from Apple made an incredible and very romantic video about. [00:07:29] Speaker A: I remember that. They made a video, right. [00:07:32] Speaker B: About. Yes, it was. It was almost like a, like a, like, like a. A trailer for a romantic comedy or something where they're like walking around San Francisco and then they run each other and they're like hanging out. [00:07:45] Speaker A: It's a pre wedding video, you know, announcing their engagement. [00:07:50] Speaker B: Exactly. [00:07:50] Speaker A: Hello, world. [00:07:51] Speaker B: About a year ago, a year ago, they made this wonderfully romantic video. There was an announcement. They gave Johnny Ives an ungodly amount of cash and equity to help produce some Kind of device. [00:08:07] Speaker A: That's right. [00:08:08] Speaker B: A lot of rumors at the time. Yeah, the rumor was like I heard a bunch of rumors actually. It's going to be an earpiece. [00:08:13] Speaker A: It was going to be a puck. [00:08:15] Speaker B: A puck, A pendant. It was going to take all of these different. [00:08:18] Speaker A: Some kind of wearable form factors. Right, right. [00:08:22] Speaker B: Who knows? Underwear. I don't know. Apple underwear could be something like that. [00:08:26] Speaker A: Yeah. So. [00:08:28] Speaker B: So they. The most recent. I wouldn't even call it leak. I don't think this is an official announcement is that it'll be a phone. I actually think this is the most accurate announcement. And they've like pivoted. They might have tried a bunch. I mean, if I was Johnny Ives and I was Sam, like I would probably make prototypes of all of that stuff. I would not make the AI powered smart underwear, of course, but I would definitely make the puck, the earpiece, the phone, the watch, the, I don't know, the, the marathon running powered AI shoes or whatever. I would try a bunch of stuff until I hit on something that we thought was going to work. And it looks like they're zeroing on a phone. [00:09:13] Speaker A: Which makes sense, I think is really. [00:09:15] Speaker B: It totally makes sense. I, I think it's hard to beat the ui, the form factor of a [00:09:22] Speaker A: phone in your pocket. [00:09:23] Speaker B: The iPhone in particular. [00:09:25] Speaker A: 100% touchscreen phone. [00:09:27] Speaker B: I mean, like I, I'm very obsessed with this actually. I've thought a lot about UX and UI over the years and I like to always talk about the keyboard and mouse like that, how that survives. Like the original inventor of that thought for sure that something would surpass that. I think that like approach is going on like 60, 70 years depending on [00:09:51] Speaker A: how you, how you calculate. I mean, typewriter, right. Like originally originated from the typewriter. The keyboard. [00:09:58] Speaker B: Yeah. [00:09:58] Speaker A: So. [00:09:58] Speaker B: Right. So the keyboard is a hundred plus. That's even older than the mouse. Yeah. So like sometimes we hit on these kind of like UI UX things and I would say that they're just far superior to anything else that's tried to replace them. Right. People got into like these kind of, I don't know, you know, pads with a stylus. That's like one approach that's been tried that I'm not a big fan of. And the touchscreen phone interface, I think that's another one that's just going to be really difficult to do something better. [00:10:33] Speaker A: Yeah. [00:10:33] Speaker B: I do think glasses and all these things are interesting. Like maybe they will surpass it someday. But none of the, none of the models that have come out have ultimately proven popular with even. Even the narrowest set of consumers. Really? Besides, like a bunch of geeks. Nerds in Silicon Valley. [00:10:49] Speaker A: Yeah. [00:10:50] Speaker B: Yeah. I just have not seen those things anywhere in the wild outside of like a select group of people in San Francisco. So I'm skeptical that those will ever get adoption, at least maybe in my lifetime. So the idea that they're doing a phone or some type of phone makes sense. But, like, at least the leak or the rumor is that the phone is going to have no apps. [00:11:13] Speaker A: I heard about that. But what does that even mean? It's going to be like an AI native phone. So all of the apps are going to be baked into AI, which kind of makes sense, right? Like, it's. [00:11:27] Speaker B: Yes. I think it's exact words headed. [00:11:29] Speaker A: Yeah. Right. [00:11:30] Speaker B: I've thought about this. I've wrote like a year ago stuff about this. Like. Yeah, I think it's. I think it's. I think it's inevitable. I think Johnny Ives a hundred percent sees this, that like, AI is all you need on the phone. It might be too early. It might not do everything, like, but, like, you know, like. So, okay. I actually believe so strongly in this that it's my theory that that's why they never made a lot of progress on Siri. So I think that Apple is really. [00:12:05] Speaker A: Yeah, I see where you're going with this. Right? [00:12:06] Speaker B: Yeah. I think Apple is very clear. They understand that, like, vibe, let's just call it vibing apps. Like, just make me an app and then it makes you, like, a game you want to play or whatever is like an obvious threat to the entire iOS and app ecosystem. [00:12:22] Speaker A: Right. [00:12:23] Speaker B: So Siri was like, really good. Like, so Siri was like, show me the weather. Play this song, do this, do that. [00:12:28] Speaker A: Like, do all these getting rid of Apple apps. Right. Like weather app, for example, it's competitive to apps. [00:12:34] Speaker B: Like, you can see it right now. Like, a lot of the basic stuff, like record my run. Right? Like, I mean, there's a whole ecosystem of like, run tracking and fitness and health tracking apps. Right? [00:12:45] Speaker A: Yeah. Use CL to record this. Totally. I use a claw as a financial tracker now. I upload receipts. It spits out a custom app directly inside of clothes. Before that, I would download an app for this. [00:12:59] Speaker B: So it totally makes sense that, like, that at least OpenAI and Johnny, I was be experimenting with phone that does that. It's a phone, it makes phone calls, it does texting, it does messaging, and it's got some super Siri that kind of does whatever you want to do. [00:13:14] Speaker A: Yeah, I think it's going to be somewhere in the middle. So what I think is going to go away is the App Store. What they're really trying to do is get rid of the App Store. So whatever this interface is, whether it's ChatGPT built into the phone, that's gonna be the platform. The distribution. The distribution platform now. So you can vibe code apps directly in it, or other people can publish their apps directly inside of this interface. I think. [00:13:45] Speaker B: Yeah. I mean, what's into. I don't. So what I don't know is if it's. If, if apps do go away, like, that's the part that, like, I don't [00:13:53] Speaker A: think so, because people are not going to be creating. Not everyone is going to be creating their own apps. That's just not. That's not. I don't think that's going to be the case. [00:14:05] Speaker B: Well, so. So I'll give a couple examples of, like, apps that I use already in conjunction with AI. So I. I was really excited when they announced. I think they called. They have so many names now. Connectors. They were. I know they had GPTs in ChatGPT, right? [00:14:20] Speaker A: Yep. So custom. [00:14:23] Speaker B: Technically, a GPT. They announced an integration with Spotify. I got really excited. I think I upgraded to a Spotify paid account to test it out. [00:14:33] Speaker A: Right. [00:14:33] Speaker B: And so I use that. I use that to create all my playlists. I actually have like this complicated recurring prompt that has. I trained it on all of my, like, existing playlist and music selection, and then it creates concepts for playlists that I pick one and then it makes it. And then it, once I approve, produces the list on Spotify for me. [00:15:01] Speaker A: I see. [00:15:02] Speaker B: So it's a good example. [00:15:02] Speaker A: That's probably why Spotify dropped 10 yesterday. The stock. [00:15:07] Speaker B: Yeah. Crappy playlist. [00:15:09] Speaker A: No, probably because, like, you know, a lot of people are doing this. What's the remote now? Right. Like anybody. [00:15:15] Speaker B: So, I mean. Well, they have all the music licensing, so you couldn't. Technically, you couldn't stream it all through ChatGPT, right. [00:15:23] Speaker A: Yeah, but then, you know, Universal can publish an API that lets you access their entire catalog and what's the point of Spotify anymore? [00:15:32] Speaker B: Yeah, but they're not going to do that because Spotify has a lot of features. [00:15:35] Speaker A: Sure. [00:15:35] Speaker B: That calculate the. So the whole model around how people get paid. Is this. This is a. I think this is actually excellent discussion. Like. Like, where are the nuances between what applications will get subsumed by AI and which ones will not? So I would bet that Spotify will exist because of the complexity of the business relations relationships and how payouts get managed in streaming music. [00:16:07] Speaker A: Yeah, I see. [00:16:07] Speaker B: Means that you have to have an official entity. [00:16:10] Speaker A: Yeah. [00:16:11] Speaker B: That has legal agreements in place with everyone. [00:16:15] Speaker A: Sure, sure. But OpenAI could have this relationship. Right. Like, I mean it seems like Apple, I mean like Apple bought Tido. Right. So Apple has this relationship. So if they decide to get their shit together, release a Siri that actually works, they could have this relationship. [00:16:39] Speaker B: So I mean this is every platform operators. Like I don't know if it's a conundrum or challenge. Like if you're the platform operator and then you have apps on top of your platform, the number one or most popular use cases are ones that you want to just build into. [00:16:57] Speaker A: Yeah. Platform Y. [00:16:58] Speaker B: And so music is that popular that maybe OpenAI would just build their own music application layer or something into their system. It's totally logical. And the thing that I always point out with this is like this has existed since hardware. So the idea that like the platform operator subsumes different features and functionality of other like third parties has been going on since the very beginning. Like used to buy all these different independent chips and then the chip manufacturers figured out how to like take these third party features that other pieces of hardware did and put them into their chip. Right. So this has been happening a long time. So I think you're correct that like OpenAI could do it. Whether to do it or not, I don't know. I tend to think that the relationships and the complexity of billing and the music industry is actually significant and ultimately OpenAI wouldn't want to do it. It doesn't make any sense for them to like take that on. There's a lot of just like accounting and complexity. You got to negotiate independent record. There's a whole bunch of stuff you have to do. And then network effects with like, you know, people like me that make crappy playlists and stuff and sharing. Maybe, maybe share those. Maybe there's. Maybe there's some impact from that stuff. All right, this is pretty interesting. Anything else you want to say on the the Johnny? I've. [00:18:19] Speaker A: Sam, what do you think the timeline iPhone later this year, 2027, 28. [00:18:25] Speaker B: Oh, great question, dude. I think this thing is indefinitely in like a lab and it's a prototype for a long time. [00:18:34] Speaker A: So you think it's going to be vaporware basically? [00:18:38] Speaker B: Well, I think they have something real, so I don't think it's vaporware. I think they really are making prototypes. I think they're really testing things I think Johnny, Johnny Ice is walking around north beach in San Francisco. [00:18:48] Speaker A: Something shoeless, shoeless. [00:18:52] Speaker B: I think, I think they're, I think they're building real stuff. I don't think it's vapor in a sense that they're not building anything, but I think that open AI clearly needs to refocus their business. And they had a lot of side quests. I think this is, this is another side quest. [00:19:07] Speaker A: Right. [00:19:08] Speaker B: Like, I mean, it's the ultimate side quest. I think it's a. I think it's one of the few things that they should continue to do. I wasn't a big fan of like, Sora from a business perspective. I played around with it. It was kind of fun. I made some videos. But I don't think it was a good use of their, like, resources. I think this is a good use of resources. I think if they pull this off, [00:19:24] Speaker A: I mean, they already paid the money, so, like, minus fall. [00:19:28] Speaker B: If they pull it off, if they pull off something that, like, even gets some traction, like, I think, I think it's a make or break technology, like, yeah. Opportunity. That's, that's the way to frame it. They're, they're like. So I guess we'll keep talking about this because I think this is really interesting. I think it took a company like Apple to create the iPhone, so there were lots of people experimenting with like, interesting smartphones. Yeah, breakthrough. I don't think it's the kind of innovation that can come from, like, a typical startup. No one's going to invent this kind of thing in their garage and be able to build a company around it and market it to do the AI phone correctly. It needs OpenAI. A company like Johnny Ives. It needs a lot of money and resources. They've got to perhaps make relationships with, like, mobile operators and carriers. Yeah, because there's a famous story where, like, Jobs is the only one that could get the operators to agree to, like, release his phone. And supposedly, and I believe this story, it wasn't there in the room. I don't know. He wrote a check in the meeting and said, if you don't accept the iPhone onto your network, I'm gonna either buy a competitor or build my own. And he wrote a check for like a billion dollars or 10 to some crazy number. He literally had the check and he's like, if you don't do this, I'm just gonna compete direct with you and I'll buy a mobile operator that has the iPhone on it. And they were like, okay, supposedly, that's how he convinced ATT or. Dude, I, I believe story. [00:21:02] Speaker A: Yeah. I totally believe it. Yeah. [00:21:04] Speaker B: Because I also think what I like about that story, it illustrates, like, that's the kind of leverage you needed in order to, to win in that negotiation. And only Steve Jobs could do that. [00:21:14] Speaker A: Yeah. So let's see. 20, 27, 28. What's your, what's your final year prediction? [00:21:19] Speaker B: I don't know. I mean, I'd love to see something next year. I would give it two years. How's that? [00:21:24] Speaker A: 20 tears. I agree. Yeah. [00:21:26] Speaker B: Christmas time, Christmas release. [00:21:29] Speaker A: Yeah. [00:21:30] Speaker B: That's what I'm Target. Yes. [00:21:32] Speaker A: Yeah. Okay. [00:21:33] Speaker B: That's fair, right? Okay. [00:21:34] Speaker A: Fair enough. [00:21:35] Speaker B: Okay. [00:21:35] Speaker A: Okay. Okay. What's the next thing we said? [00:21:37] Speaker B: All we. That we could possibly say on that topic. Oh, dude. Musk versus Altman. [00:21:44] Speaker A: Staying there. [00:21:45] Speaker B: Have you been following this a little bit? [00:21:47] Speaker A: I saw the, the, the part that I paid attention to is the subpoena. No. From Brockman where he wrote in his diary back in like 2017 that says, hey, I think we're screwing Elon. I think we're lying to him. If we keep doing this, this is probably going to be lead to a nasty fight. I. Yeah. That's why I don't write anything in my journal. I don't keep a journal. [00:22:16] Speaker B: I mean, this is going to be. Dude, I think that I, I think about that all the time. I don't want to document too many things. Like, what if. [00:22:25] Speaker A: Why would you. [00:22:26] Speaker B: Like, why would someone reads this stuff? [00:22:28] Speaker A: Like, yeah, if you get subpoena, it comes up in deposition. You're. This is literally all evidence. This entire. I think Elon is going to win because Brockman wrote this down. His diary. [00:22:40] Speaker B: All right, so I'm looking at the article here, so there was something where I saw like, most of the stuff was thrown out, though, like, ultimately. Well, this, this quote is great from Open AI. We can't wait to make our case in court where both the truth and the law are on our side. I don't know about this lawsuit has always been baseless and a jealous bit to derail a competitor. Yeah, I like, I think the characterization. Characterization is good. I mean, I don't know where I, I do. [00:23:13] Speaker A: Okay. [00:23:14] Speaker B: I, I guess my personal opinion about all this stuff is like. And I don't know the, the, the nuances of what's legally happening here. They do think that, like, Elon has a. Has a. Let's. I'm gonna put it this way. He has a. He has a. He has a point that the intended purpose of OpenAI was to be a nonprofit that would pursue. [00:23:37] Speaker A: Yeah. [00:23:38] Speaker B: AI and that is not right. I think in my opinion, like that is super clear and like I'm. Like I said, I'm not a lawyer, I'm not a legal expert, but it certainly seems to violate perhaps law that you've taken what is clearly defined as a non profit and you've turned it into a for profit company that, that seems to be like not what this is supposed to. Not supposed to happen. [00:24:04] Speaker A: That's right. Yeah. That's not what Elon signed up for and invested into. So. Okay, so question. Right. [00:24:10] Speaker B: Yeah, I think it's what anyone, if anybody starts a nonprofit, I don't think the perception is that this is someday going to become a profit, up for profit company. Like, I don't think that that's what the. So I like the word like intended structure and I think intention matters a lot. [00:24:26] Speaker A: So legal. [00:24:27] Speaker B: I don't know. Legally. It'll stand though. Lawyer. Right. [00:24:30] Speaker A: Legality and intention aside, do you think. Who do you think is winning the public opinion court? [00:24:43] Speaker B: Oh my God. That's a great question. I think they're both losing pretty bad. [00:24:46] Speaker A: They're both losing. Okay. You don't think there is a winner in this fight? [00:24:52] Speaker B: I. I don't know. Maybe my judgment's clouded. Like I ultimate like. [00:24:55] Speaker A: Well, who do you like? [00:24:57] Speaker B: I hope, I hope Elon will prevail. I think that like. [00:24:59] Speaker A: So you hope Elon will prevail from like, you know, from all of the missing or partial information that you have gathered so far. Right? We're talking about it. Who do you think is coming out ahead in your mind as of right now? [00:25:14] Speaker B: Yeah, I mean like, look, dude, I think, I think Elon's right. He's like, dude, you guys took a non profit. He took a bunch of money from Microsoft. You turn into a for profit. Like what is going on here? This is not, this is not how the game is played. Yeah, I completely agree with that. That that's. It's a crazy thing and that I guess, like, I guess for me open AI should remain a non profit and it shouldn't go public, which is what they're planning to do. That's not what. That, that's not. That should not happen. [00:25:43] Speaker A: Yeah, I agree. I agree. I think like out of this fight I literally don't know anything except a snippets of stuff. I sided with Elon so far. I, I really don't care about Sam. Whatever. Like all of the stuff that he's doing. I, I wish I want to Talk to somebody that takes the other side. Right. That's like negative, short on Elon law. [00:26:05] Speaker B: And why, yeah, why is it okay to take a non profit and turn into for profit? [00:26:11] Speaker A: Well, my guess is, I mean, investors obviously, right? Like investors looking at this, they're like, why not like I want my money? Very simple. [00:26:25] Speaker B: Yeah. I mean, but do they have like a rational argument like, other than like, like, like why is it okay to transform a nonprofit into a for profit company? [00:26:35] Speaker A: Vision. Company's vision change all the time. It's pivot. [00:26:38] Speaker B: It's like, yeah, I mean, like I said, I don't, like I'm not a legal expert. Like I, I, I guess like they, they probably set up an entity and so it's okay to create a new company. [00:26:47] Speaker A: Like there's financial engineering that went into it. There's legal engineering that went into it. Of course they, I'm sure legal, putting all the legality aside, really, this is a court of public opinion. I think for the normal people, Musk is winning. For the investors, they don't care. They support. [00:27:06] Speaker B: Okay. [00:27:08] Speaker A: They support. [00:27:08] Speaker B: I know how to close, I know how to, I know how to close this one out. So when you're a startup, in the beginning you want to have a big product team, then you grow and you want a big marketing team. And then when you dominate the market, you want to have the biggest lawyers. Legal team. [00:27:20] Speaker A: That's right, lawyers. [00:27:23] Speaker B: That's, that is like how the high tech world works. Okay, let's move on. This is the cost of compute. This is your quote from Jensen or [00:27:34] Speaker A: a vp, like a senior VP at. [00:27:36] Speaker B: Oh, it's not Jensen. [00:27:37] Speaker A: Okay. Someone else at Nvidia saying that the cost of compute is definitely more expensive than paying human worker. So like whatever, whatever relevant, like quote this comes from, it kind of like confirms our suspicions so far. Right. From my personal experience as well, I think you can speak to this. Sometimes token costs are going through the roof. They're increasing price. Claude Copilot OpenAI all of these companies are increasing subscription price. The subsidies is ending. Token price is gonna keep going up. At some point we're all gonna collectively wake up and say, hey, maybe I should just hire somebody to do this instead of using AI tools. I think this is gonna mark the end of, or the beginning of the end of vibe coding. Vibe. Anything where people who really has no idea what they're doing just throwing token at the problem. I think this is a wake up call to all of us. [00:28:40] Speaker B: Are you, are you telling me I got to change the name from Vibrassas something else. [00:28:44] Speaker A: Viper sass. Well like, I mean, are you through? I think your usage of vibe is [00:28:51] Speaker B: a little bit more. I, I'm glad you said that. That's exactly what I think. That's the way. Because I get this question once a while I'm like look dude, that was the reason I went with it cuz I thought it was like funny. [00:29:01] Speaker A: Maybe you should, maybe it's the end. [00:29:03] Speaker B: I literally wrote it down somewhere. I was like there's going to be a 12 month run for this thing. That's basically where, where we're at. Yeah, but I still think, I still think the vibes matter. [00:29:13] Speaker A: The vibes do matter. But I think the definition is done now. I think we need to redefine it a little bit. So maybe it is. [00:29:21] Speaker B: I agree. So I agree 100 that like token costs are crazy and like it looks like maybe you won't save any money with AI. [00:29:32] Speaker A: That's right. That's. [00:29:33] Speaker B: Yeah. So that's what I found with claudebot. I was like, everyone's all excited about it and I was like looking into it and I'm like God, I could probably just do this. We do that so talking to people and they're like oh yeah, the token cost. And I was like what? I was like I could just hire [00:29:47] Speaker A: spending 2, $300 a day. [00:29:50] Speaker B: A day. I was like, yeah. I was like you can get someone in upwork for 50 bucks or a hundred bucks. I just like why would I do that? So that was what I ultimately did. I hired someone to help me with a few things I wanted them to do. And I did not automate it because I just, I just didn't see the value or the need in like doing it. So. So the only thing that could change in this equation is like if they build out all these data centers then the costs will go down. [00:30:20] Speaker A: Yeah, that's a ten year horizon though. That's the thing. [00:30:24] Speaker B: Maybe. [00:30:25] Speaker A: Of course, of course. That's the direction that we're moving towards. Right. That's like in the early Internet days that says hey, if everybody banks online the value of in person banking is zero. But that took like what, 20 years to materialize. [00:30:42] Speaker B: Well I was thinking more like high speed Internet. It took like five years for really become pervasive and like it was more expensive in the beginning. [00:30:53] Speaker A: No, no, no, wait. High speed Internet like DSL was available what late 1990s, 2000s. [00:31:00] Speaker B: Yeah, but it wasn't very good and it was expensive. [00:31:02] Speaker A: It was expensive and then it was [00:31:05] Speaker B: expensive and it Was not available everywhere. It was hard to get. It was a pain. It was pain. Like no one liked it. I remember like istn, all that stuff. They all suck too. [00:31:11] Speaker A: They'll. At least in Canada we didn't have stable high speed Internet until maybe like 2010. No, is that true? No, no. [00:31:23] Speaker B: That's what I'm saying. It took, it took time for like took like, I remember like New York first. Got it. It took like a couple of years. Yeah, it rolled out fairly quickly. So that's what I'm saying. Like, will tokens be that, like, sure today? [00:31:37] Speaker A: I think so. Yeah. For sure. 10 years. It's going to take us 10 years to where anyone in the world can have billions of tokens. [00:31:47] Speaker B: Maybe. So the, the flip side of this is that the costs are super. Do we have that? Do you have that quote? That. But there's a quote where Jensen is saying like $250,000 a year in token costs. [00:31:57] Speaker A: Yeah, that guy's. That guy's just saying things, you know. Yeah, yeah. [00:32:02] Speaker B: He's saying this is like a crazy. That's like a crazy number. Right? That goes to 50k. Is that even still worth it? [00:32:07] Speaker A: So there is a nuance to exist. Exactly. The quote that he says, he says you should be giving your best people unlimited amounts of tokens. So that part, I agree with him. If you kind of like read between the lines, what he's saying is like if you are not giving your best developers, Your top developers $250,000 in token spend every single year, then you're. You're shooting yourself in the foot. I kind of agree. [00:32:35] Speaker B: Isn't that what's important is like, we've deter. We figured out that like, AI makes the best people way more productive. [00:32:42] Speaker A: Yes. [00:32:43] Speaker B: And because it's high cost, it doesn't necessarily. It's not. Well, because it's high cost, it's not necessarily valuable to improve the skills of low productivity roles and people. [00:32:55] Speaker A: Correct, Correct. So basically we're lifting the top, but not particularly lifting the equal amounts off of the bottom. At least I don't think so. [00:33:06] Speaker B: This was a big debate at one point. Like, will it do that? All right, all right, let's move on. Let's do this. This AI agent deleted my code base conf. [00:33:17] Speaker A: Oh, man. Okay, so this tweet got, I don't know, six, half a million or almost seven million views on Twitter. It's going super viral. Got reposted to Reddit as well. Lots of discussion. So the long story short, this founder came out. This founder of A company called Pocket os, an app. He seems like he vibe coded the entire business and he essentially documented how he used Claude while running some normal day to day usage, deleted his entire production, database backup and all off of his hosting company called Railway. And he went through step by step exactly what happened, all of the mistakes that he made and kind of people went wild on his post. Some people said, well it's crazy that this could happen. It's fault of the agent model, it's the fault of the hosting provider. Lots of people of course piled on to him, says, you know, it's his fault for, you know, vibe coding his backend in the first place. And I. This is, this is, this is a tough topic to dissect because. So in the old days where whether if you're a junior developer or senior developer, the puck stops with the developer, right? Like the puck stops with you. Regardless if you have like checks and balance in place to not let you delete the database, you could always point to somebody and say, hey, look, you ran the command, you're responsible for this. But in this day and age, if you're a founder, like we just previously said, you have no idea what you're doing, you have no idea what SQL query is. And then you're deleting a database, deleting the entire backup volume. Who's to blame here? Is it the person who's running the command or is it the hosting company? Is it the model company? It's like we're in this state where there's blame but nobody knows or agrees on who should own this fault. So I think it's going to get a lot worse before it gets, it's [00:35:42] Speaker B: a little bit like the self driving car. [00:35:44] Speaker A: Like if it's exactly like that fall, [00:35:46] Speaker B: it's kind of like it falls into that. [00:35:49] Speaker A: That's exactly the argument with, with AI usage right now. Because AI is in the cloud, it's not a real person, it's a. [00:35:58] Speaker B: Why is it not the guy's fault for running this agent? [00:36:01] Speaker A: Because he doesn't know better. That's the problem. [00:36:04] Speaker B: Why did the agent delete it? What did it think it was doing to do that? [00:36:10] Speaker A: Because is that explained? So he made a series of mistakes, right? He went through the whole bunch of step by step and if you're kind of like a senior developer by the middle stage you're like, okay, I'm going down the wrong path here. Let me like, let me back off. He literally went through step by step and then he said he instructed the AI to do A few things that could be misinterpreted as just delete everything. And that's what the AI did. [00:36:46] Speaker B: That's funny. [00:36:48] Speaker A: So, yeah, not for him. It's not for him, but let's try. [00:36:51] Speaker B: Yeah, not funny for him. [00:36:52] Speaker A: Not funny for him. Also, Railway, the hosting provider, kind of got blowback as well. They already came down from hack or security leak a couple weeks ago. So they're just consistently getting bad press because some people say, why does Railway even let you do this through the API? It doesn't make sense. Right? Like it doesn't make sense, but it's also such an edge case where if you're building for a normal developer, nobody does this. [00:37:25] Speaker B: Yeah, I mean, I don't know. I mean, so what do you think this means for Vive coding? Is this the end of Vive coding? Like, is this the beginning? So we've had this, like, we've had the, the narrative arc, all new technology adoption that goes through. [00:37:39] Speaker A: Yeah, of course. [00:37:40] Speaker B: And exciting. It's going to change the world. Everyone needs to do it. If you don't know how to use AI, someone who does will take your job. [00:37:47] Speaker A: Right. [00:37:47] Speaker B: We've gone through this whole, and now we're at the whole, like, oh my God, AI does actually all these bad things. It's got all these drawbacks and there's challenges and there's issues and lo and behold, you do need to know something [00:37:59] Speaker A: to use, of course, because on the other end, you're a customer. Right? Like, think about the customer. If I'm buying a piece of software from a Vibe coder or a business that says, we Vibe code everything, am I like, going to throw money at this? Probably not anymore. The fun excitement of vibe coding and moving fast is a little bit over now, right? So this guy, so the difference is that this, this guy had real customers. That's why he's so frustrated, right? He has real customers that depend on his software. He literally blew away all of their data. So, like, he basically killed his business. If I was a customer of his, be, like, what, like, why would I keep paying you to do this? [00:38:51] Speaker B: Yeah, I don't, I don't know. It seems silly to me. [00:38:57] Speaker A: Well, I mean, I think. [00:38:59] Speaker B: What, so what, what do you, what do you, what do you think this means? So what do you think this means for the industry? Does it mean, like, vibe coding now gets guard railed? Vibe coding is something that is not for everyone. I mean, that's kind of where I go with this. But, like, I'm curious, like, what do you Think happens. [00:39:16] Speaker A: Yeah. So I think the, so the floodgate for anyone, the idea guy to just Vibe code their idea into existence, that's gonna, the window for that is gonna start to close a few ways. One token price is gonna go up. So a lot of people who just has an idea has zero go to market strategy, zero understanding of the market, they're gonna get priced out, right? So okay, if I heard that max subscription is going to go to $1,000 a month, people that's Vibe coding are not going to be able to afford that. [00:39:50] Speaker B: Like, why even bother? [00:39:50] Speaker A: Why even bother? The second thing is the customer base. The customer's appetite for vibe coded B2B applications, let's say, are going to start to go away. They're going to start asking questions on sales calls is like, who's your engineering team? Do you have a cto? We're going to slowly revert back to the pre VI coding days where you demand some kind of engineering person whose sole responsibility is making sure nobody is actively deleting production databases. Oops. Oops. Yeah. So I think customers are going to start asking these questions and that's going to price out or push out some people. So like, so this is interesting because essentially like what we just said previously, the top people, the top percentile people are going to reap all the benefits and the bottom people are not going to see as much. [00:40:55] Speaker B: So yeah, I, I agree. I, I, I, it's interesting, I think seems to always come full circle, like it's going to turn out that Vibe coding is a great way to accelerate your engineering team. But the same token, it's like I keep wondering this too. Like how fast can they go? Like if the expectation is that they're going to guard rail things and QA things and like pay attention to things and understand how things work, right? They can only go so fast and they can only go as fast as they can comprehend, right? So it limits the, it limits some of the benefit. Right? And now we're seeing the fallout of like a less carefully planned and executed product that has risks. AI, yeah. Has huge issues with it. [00:41:38] Speaker A: Like these, these things are sub cost, right? You're, you deleted the database. There was a whole bunch of security stuff that we've already spoken about previously. Every single week you're hearing security problems. So that's like the secondary effect. So I think customers or buyers, enterprises especially are going to wake up to that. [00:42:00] Speaker B: So it kind of reminds me maybe of like maybe the early days of blogging. So I remember early days of blogging it was like everyone wanted to or had a blog, but very few people actually like managed and updated them. And there was a joke at one point where I was like, every blog post reads with like, I know it's been three months since my last post. Every single one. [00:42:25] Speaker A: Every single one. [00:42:25] Speaker B: Right. So I think it's kind of like that, like in the sense that like everybody's gonna try this and a bunch of people are gonna be like, okay, that was a fun thing to do. And like, I'm actually not all that committed to being a professional full time vibe coder, because I do think so. Okay. I do think there are going to be people who did a lot of vibe coding and GTM engineering stuff and they're gonna be quite good at it. And they will have this new kind of interesting technical hybrid role. But I don't think there's a lot of those people. And they'll be, and they'll be, they'll be good at it. They'll understand what they're doing, they'll be valuable person on the team. Like we even have, we've always had these like hybrid like marketing roles where people were pretty technical and could do things in SQL, but they weren't ultimately on the engineering team. So it's always existed and it's a small segment of like business people. So I think that'll continue, but it's not going to be in a lot of organizations because I think people like that. In my experience, these hybrid roles ultimately always go away because of the challenges around working in a team and managing everything to standards. So eventually every company I've ever been in, there's a point where it's like, we need a standard way to do this. And once you decide that like there's a standard way to do things, specialize you have to be on. Yeah, you got to pick. Are you on this team Run. So this was even me way back in the day. I used to do a lot of like coding stuff and eventually they were kind of like, Gregory, you got to pick, like, what are you going to do? And I was like, okay, I'm going to just be on the design team. [00:43:58] Speaker A: I think I don't fully believe this. [00:44:00] Speaker B: I don't want to be full time. I remember like very clearly, like, like, I'm smart enough, but I'm not, I'm not like gifted. And so I'm not going to pursue this to the exclusion of other things. And so I made my decision. Right. And I feel like that's what's going to happen with A lot of people, they're going have to pick and then, and then I guess I would recommend, like, pick the thing that you think you're really gifted at. Like, don't pick the thing that you have an aptitude for, but like, aren't really comfortable with, like building a career around. How's that? [00:44:30] Speaker A: Good summary. [00:44:31] Speaker B: All right. Yeah, I'll come on this quickly. So this one was super interesting. I stumbled on upon on X the other day. It's not AI related, but it is newsworthy. There's been a lot of, like, punditry and discussion around the impact that Ozempic and GLP1s will have. So these are like weight loss drugs and you know, they're being used at like a fairly large scale. So there's always been this question, like, is this going to actually impact food sales? It's going to actually impact the amount of junk food people consume. And so there was a tweet by some guy who, like, I didn't look too deep into his background, but he does work for a company that provides analytics for freight shipping and different types of transportation. And, and he tweeted out some data saying that, yeah, there is actually a measurable reduction in the volume in shipping around the food and beverage category. This tweet even says, like, hey, I spoke with the CEO of a mega fleet, so I'm assuming like a big truck driving company. [00:45:55] Speaker A: Yeah. [00:45:56] Speaker B: Who said that everything was going really well in terms of their business, but. Except that the food and food and beverage segment had a huge drop in volume. So I love this one. Like, it's like a. It's something that a lot of people have, like, thought about and have like perhaps predicted or asked and were expecting perhaps to see in the data. And so now, you know, we actually see something emerging. [00:46:27] Speaker A: What do you think about this? So, so is this basically saying that the consumption rate for food and beverages is going down because of GLP1 or is it kind of like the other way around? [00:46:43] Speaker B: All right, so it says, it says exactly. Here's the number. [00:46:46] Speaker A: Okay. [00:46:46] Speaker B: Estimates that 851,000 truckloads of food and beverages have been removed from the market due to GLP1s, and this could go as high as 1.95 million in 2030. So a significant amount of food and beverage shipping is no longer happening, which they attribute to GLP1. Yep. [00:47:14] Speaker A: Who. Who did? [00:47:15] Speaker B: Which is that weight loss shrugging. The analyst. [00:47:19] Speaker A: The analyst. I don't know, man. Like, food prices are just going up, right? Like beef prices through the roof. I personally Stopped eating beef just because the price is so crazily high. I'm eating a lot more fish. So I, I don't know. I don't know. [00:47:39] Speaker B: For example, that would be the, that's the counter to this, right? [00:47:42] Speaker A: Yeah. Just because of the rising price of food and beverages. I stopped pretty much. I stopped stocking my fridge with beer [00:47:48] Speaker B: because so, so people are getting skinnier because it's too expensive. [00:47:53] Speaker A: Oh yeah, for sure. For sure. Like, I mean I'm. I, I stopped, I stopped eating junk food actually. [00:48:00] Speaker B: Buy that. I buy that explanation. I don't know. This, this is like, this is one of those things. Like it's just some random what I'm so point somewhere. So like, I don't know. I mean, I thought it was really interesting, so I want to talk about it. [00:48:12] Speaker A: It is interesting. [00:48:13] Speaker B: Yeah, I think it's possible. I'd have to see more data. I'd have to see this data repeatable. I'd have to see other people do. For sure. People with other. For sure people who track this. [00:48:23] Speaker A: Yeah. [00:48:23] Speaker B: I also would be, I would also expect to see this in like a Coke or Frito Lay sales. So like junk food companies would like ultimately have to have a decline of some kind of decline or flattening of sales for this to be, to be true. [00:48:39] Speaker A: Yes, that's what I think. Yes. Yes, yes. Do you think that there's a lagging time? So if you look at stocks right now, right, like for example, Coca Cola doesn't sell direct 100%. [00:48:50] Speaker B: That's why these like analysts. That's why these guys exist. Because like stock traders try to look at the shipping data to make determinations on stock trades. [00:48:58] Speaker A: But even then, like for example, Coca Cola, they don't sell direct to people. So shipping going up and down is just completely irrelevant to their, their stock price. Right. Because usually like if it's a, if it's a grocery chain, they would bulk buy or bulk stock anyways. I don't know enough to show up. [00:49:23] Speaker B: Like, it's. I know it's, it's con. [00:49:24] Speaker A: It's. [00:49:24] Speaker B: It's an interesting one. It looks like a big number. But like I said, that would have, we'd have to see. There would have to be other aspects of the economy that would see slowdown or decline in order for this to be true. And that, that, that hasn't happened yet. But like we're in, we're in the quarterly earnings. So like maybe Coke will come in. Hey, guess what? [00:49:43] Speaker A: I would be. So, okay, so if the hypothesis is that People are eating less because of GLP1 or they're consuming less. I would be interested to see if the restaurants either lowers their menu prices, I suppose, because like, if, if this is because people are eating less, then maybe, I don't know, Denny's will reduce their menu size. [00:50:12] Speaker B: I don't know. [00:50:12] Speaker A: Menu item size? Yeah, like, I don't know. [00:50:15] Speaker B: I do think so. I think I. I think you. So I think you're going the right direction. [00:50:18] Speaker A: I need more data. [00:50:19] Speaker B: Decline in. You would expect to see a decline in spending overall at restaurants? Of course, that one. There is a number you can get, of course, spending. [00:50:27] Speaker A: And then restaurants reacting to that declining spending by reducing. [00:50:31] Speaker B: Well, like, dude, was it Carl's? I mean there are like food. There are public food companies, food chains. Yeah. Chain. Like chain restaurants. Some. Some of them are starting to struggle. But again, it goes back to like, is it the economy or is it GLP ones? I think it's hard to actually tease out that. I think it's easy to say. Or there's data that indicates. All I know is people are spending less on food. [00:50:55] Speaker A: Right. [00:50:56] Speaker B: But why? We don't really know why. [00:50:59] Speaker A: As long as Costco keeps their hot dog at A$99. [00:51:03] Speaker B: There you go. [00:51:04] Speaker A: Everything is fine. [00:51:08] Speaker B: All right. Got a few more things we got to get through. Dude. The clipping industrial complex. Explain what clipping is to everybody. [00:51:16] Speaker A: Okay, so in. So clipping is when either influencers or you hire a team of people to take long form content like this podcast, this video content, cut it up into let's say 30 to 50 seconds, short form videos is. And just distribute it on their own accounts across social media like Instagram, TikTok, YouTube Live, YouTube Shorts. Twitch is also another platform. [00:51:47] Speaker B: Straight. Straightforward enough. [00:51:49] Speaker A: Yeah. So that's the mechanics of it. The secondary is basically the reason is that you're gaming the algorithm on these social media platforms. Right. By essentially hiring on a CPM basis, thousands of people to distribute your content for you. So I would call. [00:52:08] Speaker B: That's the kind of. That's what I would call the twist. So like we create a bunch of clips for this show and all right, people create clips. Like that's nothing new. That's nothing new. [00:52:16] Speaker A: Correct. [00:52:16] Speaker B: But the twist is that people are like taking that to this like crazy extreme. [00:52:23] Speaker A: Yes. [00:52:23] Speaker B: Where they're like giving. They're giving the videos to people to like create their. Create clip. To create a bunch of content from. [00:52:31] Speaker A: Yeah. [00:52:31] Speaker B: And then where do they publish that? [00:52:33] Speaker A: So they publish that on their own accounts. So these clippers have a small following, let's say typically a small following, let's say between 5,000 to less than a hundred thousand followers across these social media platforms. And they distribute it for you. Right. A lot of the times these things are not even publicly sponsored. They don't announce that this is sponsored only because the video themselves already have some kind of overlay that says this is a sponsorship or this is a clip from this company. [00:53:11] Speaker B: Yeah. Okay. I mean it makes sense. So why is this getting all this attention? [00:53:17] Speaker A: Yeah, so this is actually interesting. I knew about this for at least a year now. A lot of B2B, sorry, B2C companies, especially E Commerce, Live shopping, a lot of consumer applications are already doing this on TikTok. I think it's picking up now because B2B companies are picking this up. Recently a article on. Is it New York Times? I can. Forbes, I think Forbes. [00:53:49] Speaker B: Right, yeah. [00:53:50] Speaker A: Forbes wrote about this and a lot of people are now interested because they got a, I suppose a view behind the curtains how a lot of these companies are a group. Cluli is another company that famously did this to go viral across. [00:54:10] Speaker B: Yeah, they did a good job. Right. They, they claim that they achieved billions of impressions with Correct. [00:54:14] Speaker A: Through clipping. Yep. [00:54:16] Speaker B: Yeah. [00:54:17] Speaker A: And I think so secondarily the reason why is a lot of these platforms, they're called bounties. So this is when as a small time creator, they can go onto these platforms to claim bounties, which is claim missions or kind of like a upwork mission from these companies that says, hey, we have this video, please, we'll pay $2, $3 CPM, clip it up, distribute it, report it back and then we'll pay you. These platforms are popping up. [00:54:51] Speaker B: I think it's cool. [00:54:52] Speaker A: Very cool. [00:54:53] Speaker B: I like it. [00:54:53] Speaker A: Yeah, so I don't think so. I think this code has been cracked for a lot of B2B sorry, B2C companies. Cluli is a really good one for B2B but I think it's going to go even more white. [00:55:10] Speaker B: Yeah, yeah. So the big innovation is that it's not just like clipping and then throwing them up on your YouTube or Instagram, it's that you're like aggregating like a thousand different editors who all have a small following and then they're all making their own kind of interesting version of it and then publishing it on their platform. And they report back the metrics. [00:55:35] Speaker A: Yep, exactly. [00:55:36] Speaker B: And yeah, I love it. I love the creativity of it. I love the like the fragmented and almost democratized approach to distribution. I think there's A lot of cool things happening. I like the, the like to some degree, Like I would almost call like loss of control. Like, like this would have been like incredibly taboo 10 years ago for marketers to do because you're essentially like almost open sourcing it, saying like, hey, here's a, here's a video. Do whatever you want with it. And like in the old days, people wanted like brand a lot of control. [00:56:16] Speaker A: That's right, yeah. Brand safety. Like, so these had to be certain [00:56:21] Speaker B: way people were just these like, I'm kind of glad it's over. Like, exactly. I always thought they were too precious about the creative execution to the point is being like excruciatingly difficult to work on or work with. And I remember like being on these projects and was just like, what does it matter so much if it's like done this way or done that way? Like, I get that's how you want it. And they would always like create this narrative like, oh, it's about our brand. It has to be consistent. Everyone's always like, would go down that path. But it never, I personally never believed that. And I think that like this approach proves that like extreme levels of consistency actually don't matter a lot. [00:57:08] Speaker A: Correct. Because then people, if you see the same video over and over again without any variation, people start tuning them out. They get bored. [00:57:16] Speaker B: That's exactly what I think too. [00:57:18] Speaker A: They get bored very quickly. And then digital. [00:57:21] Speaker B: Go ahead. I'm sorry. [00:57:22] Speaker A: I think during the pandemic. So companies working with creators has been happening for decades now. Like hiring, sourcing creators to produce content, posting for them. That's been happening for a long time now. So a lot of company has built out ecosystems around how to manage that relationship. So let's say you run a campaign, you hire a couple dozen creators and you have to send out contracting. Maybe if you're an E commerce company, you have to send out in kind gifts to actually ship the thing to them. That was a really complicated campaign and marketers just got fed up with it. So like imagining, I think one stat is that in Rust's campaign got half a billion views from, I don't know, 20, 000 videos by 500 creators or Clippers. Imagine managing that campaign if you had to like approve the message. [00:58:26] Speaker B: Yeah, pay them off $100 or whatever. [00:58:28] Speaker A: And yeah, it's collect the stats, approve the content. That would be. How do you get all of the metrics, Aggregate the metrics, self report, keep them on schedule. Impossible. [00:58:44] Speaker B: You're not signing. You're not signing up for this Paul Impossible. Gregory and Paul show one billion impressions. [00:58:49] Speaker A: One billion impressions. [00:58:50] Speaker B: We're gonna clip it. We're gonna clip our way there. [00:58:52] Speaker A: Yeah, yeah. And also like this is a direct answer to the top of the, the, the ad cost going up. Right? Like Facebook is down. [00:59:00] Speaker B: Well, yeah, that. So okay, so. Right. Why would anyone even bother with this? Because like the ads are just getting astronomically expensive for sure. And like it. [00:59:09] Speaker A: Yeah, exactly. [00:59:10] Speaker B: I, I also would think that or would make an argument that this might be more effective. So it goes back to where we're saying that like people get bored and they see the same thing. So letting a bunch of people clip it, spend their own, like put their own spin on it, make it different, put in their channel. Like it might make it more interesting. [00:59:25] Speaker A: Right. [00:59:26] Speaker B: And it might ultimately be a better, A better approach. [00:59:30] Speaker A: Yeah, exactly. [00:59:31] Speaker B: All right, let's, we've got two more things. We'll just run through them really quickly because we're, it's, it's noon already here and in the west coast. So I put this one, that in the, in the, in our, in our doc that I wanted to mention. Google cloud revenue came in at $20 billion for the quarter. This was announced yesterday and it's a jump of 63%. So there's been a lot of haters out there, let's say over the years about Google's investment in their cloud product. And I, I think this is exactly the same thing that they did with YouTube. They've invested an enormous sum, they've put tons of resources into the Google cloud product for many, many years. It's always been these like haters who are like, oh, where's the profits? When's this gonna work? Like blah, blah. Well, like looks like there you go. Exactly the same way YouTube played out. Like they invested in YouTube for years. I remember like people were so critical of YouTube saying like this is a giant waste of time, like it's never going to work. I can't believe they spent all this money like five, six years in. But look at YouTube today, right? It's, it's basically the dominant platform and I even think probably they're going to maybe beat Netflix. [01:00:49] Speaker A: Yeah, they're, they're biggest IP for sure. [01:00:54] Speaker B: So like it's really cool to see Google like just like ignore all of the people like ankle biters out there, just like peanut gallery and watch them invest and watching them win. I really enjoy that part, like seeing them make these big investments, these big bets and keep at it. And I'm inspired by it. And I think like, everyone out there is being inspired by it because I think what it means is like, if you stick to what you're doing and you do a great job and you keep at it and you ignore all the people giving you. [01:01:23] Speaker A: Yeah. [01:01:23] Speaker B: I mean, in general, in general, you're, if you stick it out long enough, you'll. [01:01:28] Speaker A: Will succeed. Yeah. And also, like retrospectively, most people don't know what the fuck they're talking about. Google knows. Google is ahead of the game. [01:01:38] Speaker B: I mean, they, they knew that they had a win and now they've got like, they've got the AI initiative, they've got the cloud initiative, they got YouTube, [01:01:45] Speaker A: they have a social media team. Dude. It's really impressive. It's really impressive. Just, just like retrospectively. So they tried to expand into social media to get more data. Do you remember Circles Google? Yeah. Oh yeah, that was a complete Circles. I think Google Circles. [01:02:07] Speaker B: Called Google Circles. Is that what it's called? [01:02:10] Speaker A: Yeah, Google plus or Google Circles. It was, they kept changing the name. [01:02:16] Speaker B: I remember. [01:02:16] Speaker A: Plus. [01:02:17] Speaker B: And then, yeah, and then, then they hat, dude. [01:02:20] Speaker A: Yes. [01:02:20] Speaker B: Not, not a, not a successful venture for them. [01:02:22] Speaker A: No, not a successful venture for them. It like turned it into circles afterwards. It became like some instant. Anyways, basically they were trying to get into the social media game. My guess is because they saw the data part of the training for AI and they tried to do it with circle. They try to do A plus. And then they realized that they were sitting on go mine with YouTube and off, off to the races. So they are. [01:02:52] Speaker B: We got, we gotta, we gotta move on. We got one more. You want to do this one? [01:02:57] Speaker A: Yeah. [01:02:57] Speaker B: Okay, so Gary, Gary Tan's post. Okay. [01:03:01] Speaker A: Okay, so Gary Tan on it. Nobody asked him, but he essentially came out with a X post that says, here's the official YC advice about being truthful and precise about what is a pilot bookings, revenue, recurring revenue. What does any of this actually mean? He had to like, come out with the official YC stands on all of these terms. I think it's hilarious, the timing. Of course, I'm a little bit more cynical about it. I don't think it's because, you know, because of the integrity of Gary Tano and YC to come out and clear some of the misconceptions around mrr. I think it's because a lot of their AI portfolio companies are raising crsb. They need to potentially cover their own assets. So, so that's my take. [01:03:58] Speaker B: I, I, I deeply respect and admire Gary Tan for doing this. I think it's Great. [01:04:05] Speaker A: Yeah. I think they're just, I, I think they're cutting their losses. [01:04:10] Speaker B: I think it's great. I think like there's been a. We've talked about on the show, I've talked about it that like people are tossing around all these numbers calling all these things ARR. That part perhaps aren't ARR. And of course, yeah, he has other stuff in here with like non binding letters of intent loi. Like there's a lot of startups that you talk to try and raise money and they're like we've got these customers and we've got this letter, we've got this and that and they're kind of like stretching the truth and course trying to like position themselves correctly. [01:04:32] Speaker A: A lot of is like pass through GMV that go straight to token cost. [01:04:36] Speaker B: So, so which is, which is all. I want to be too hard on anybody. Like it's all normal. Like people try to position themselves and make their company look as, as positive and as attractive as possible. That's their job. That's their job. [01:04:49] Speaker A: Yeah. [01:04:50] Speaker B: And there are some, let's call it leeway that you have particular stage company, a startup company, you don't have to necessarily abide by the most strict change when it comes to. [01:05:03] Speaker A: Yeah. [01:05:04] Speaker B: Yes. So I think, I think we all recognize that but I do think that there was a point where this started to get really out of whack and people were perhaps positioning things and saying things and doing things in a way that like, I don't think, I think was misrepresenting what was going on. And I think that. [01:05:19] Speaker A: I agree. [01:05:20] Speaker B: I think it's great that Gary has just said like hey, just not pointing fingers but please don't misrepresent your numbers. And like by the way, this is how I recommend you calculate all this stuff. That's a very, it's a very, I think it's a very adult appropriate way to go about this. So you can call anyone out. [01:05:40] Speaker A: Agreed. [01:05:41] Speaker B: He just said like look at yc. We've, we've discussed this before. Here are the official approved y commoner way to look at this. [01:05:50] Speaker A: I agree. [01:05:50] Speaker B: Please take this under advisement. I think it's fantastic. [01:05:54] Speaker A: Yeah, I mean I think it's about time. So that's the only thing that I will say on it. [01:06:01] Speaker B: All right. All right, I gotta run. We're over. We're over for the show. [01:06:03] Speaker A: Okay. Let's end it here. [01:06:05] Speaker B: All right, dude. Hey, thank you all. We'll be here next Thursday. We're doing our a live stream Every Thursday at 11am what time in Canada? [01:06:16] Speaker A: 2pm Eastern. Canada, 2pm Eastern. [01:06:19] Speaker B: It. [01:06:20] Speaker A: But. All right. Perfect. Nice. [01:06:25] Speaker B: Awesome. [01:06:25] Speaker A: Dude. Dude, that was great. [01:06:27] Speaker B: So much fun. I got to go get my. I got to get my haircut. Very important.