← Infinite Loops
Loading summary
Transcript109 lines
- [00:00]
A
This is a 18 wheeler headed straight for us. And if we don't, you know, get out the way or figure out how to stop it, we're going to get crushed. Things are going to have to be a lot more personalized. Mr. Beast has 440 million subscribers and I think there will be 440 million different MrBeast versions for each person based on the data. There will be creators that will be on their own and just like have that authentic connection. But I think there will also be people that don't exist that are creators. So AI creators that are making the perfect content for you and so just let the best win.
- [00:34]
B
Hi everyone, I'm Nick dewilde with Shaughnessy Ventures. I am still guest hosting Infinite Loops today because Jim is still being uploaded to computer. Today I'm joined by Revan Lazarus, the founder and CEO of Jamie AI, which is building at the intersection of media, AI and digital advertising. He's here to join me today to talk about exactly that. Please enjoy the conversation. Hey Robin, how's it going?
- [00:58]
A
Good, how are you?
- [00:59]
B
Good, good. So why don't you walk me through exactly what you're to and what you've built at Jimmy AI?
- [01:05]
A
Yeah, well, first, thanks for having me. Super stoked to be here. So at Jamie, we are a AI platform for podcast networks and digital sales teams. So we essentially allow sellers and the whole content organization to understand their content and their portfolio at a deeper level and then use that content and that data to help make sales and production materials like custom decks, run of shows, stuff like that. And then be able to query across the entire content suite to figure out what's exactly working, what's not working, and then what's going to work for brands.
- [01:39]
B
So I was going to ask, so with that description, help me think about it in terms of like actual show production, pre production, pre production, production, post production and marketing and social. Totally the whole, the whole gamut, the whole suite.
- [01:51]
A
So we mostly focus on the pre production side. Typically what happens is we'll ingest your entire portfolio, but specifically on a show basis, we'll analyze what has actually worked based on retention. So figuring out this giant retention graph, which is like a L shaped graph that has all these peaks and valleys, like what will be on this video as well, and then figuring out at scale, based on those retention spikes, what is actually working for your audience, what's not working, how can you improve your hook, your transitions, your editing. And so when you are prepping for an interview, you know exactly what questions are going to create those Spikes, or at least more so of that. And then additionally put together a full interview map from your guests. So for instance, if you're researching me or somebody, typically our guests are a little bit more famous than myself, can look at their past interviews, can figure out what's actually worked well on social for them, and then do the full research. So watching up to 500 hours of content to figure out exactly what they've talked about before, what they haven't talked about, what actually did well, but maybe they've only spoken about it once. And then what you should really touch on based on what your audience typically likes, based on your past videos.
- [02:52]
B
You're famous to me. You're already so famous to me. So. So when you're, when you're communicating all this, are you working directly with talent? How big is that team? Maybe some examples.
- [03:03]
A
Yeah, so typically we don't work with talent. We work with the producers on the networks that work with the talent. And so we usually are, you know, with the producers that are making those run of shows, those interview maps. And so some of our customers are Way Sports and Entertainment, which started New Heights with Jason Travis Kelsey. They also do Carmelo Anthony's show and Andrew Santino's show. We also have Dear Media, which has the Skinny Confidential and podcasts like Khloe Kardashian's podcast. And then we work with a bunch of other great shows like Amon Rossin Brown show, who's a big NFL player, and a suite of those guys.
- [03:41]
B
So sports, pop culture, it doesn't even matter the content that you're covering. All that matters is that you have a system.
- [03:46]
A
Exactly. Yeah. So like an example is, you know, we work with a few different shows at Amazon's Wondery Network. So that's like, baby, this is Kiki Palmer, which is a more pop culture show. We also work with a show that is Zach Sang show, which talks to music artists and producers and songwriters or, you know, now they have other sports shows as well. So we work on those shows across the board.
- [04:09]
B
When did you start?
- [04:11]
A
We started officially March 1st of building this product. However, I'd been producing a podcast for six months before that and doing all the editing and pre production and doing all the ad sales. And so I feel like I've been in podcasting probably two years. But yeah, just building this probably a year and a half ago.
- [04:30]
B
What's the biggest change that you've experienced from beginning to now? Terms of where you thought you'd be providing media services or production services to what you're actually doing today. And how has AI affected that?
- [04:44]
A
Yeah, I think when we first started, I think there wasn't a lot of buy in from production teams with AI. I think that probably you and I are trapped in this bubble of like tech and you know, SF or even New York and LA is a big deal. And a lot of companies call it a year and a half ago, just weren't in that state of mind of like, oh, AI is going to change the way we do everything. And I think that's starting to happen. But I think even at the grand scale it's not the way the world operates right now. So that has slowly started to be like, oh, we need to work on AI initiatives. And I think more companies have had budget now allocated for that and are open to experiencing. And especially like top down, like really saying like, hey, this is going to change our business. This is a 18 wheeler headed straight for us and if we don't, you know, get out the way or figure out how to stop it, we're going to get crushed. And so I think a lot of those teams have been more willing to be interested in what we're doing.
- [05:39]
B
The difference in attitude, because I've come across this a lot in some of the media projects we do at Chauncey Ventures is what's the biggest difference in narrative when 18 months ago you're saying, hey, are you thinking about this? And they say no to AI why versus why they saying yes to AI today?
- [05:55]
A
I think a big portion of it is AI 18 months ago was pretty crap for what they do. Especially on the creative side. Ask AI to write you a story or a book. It's pretty horrible. And so I think it was people testing it out ChatGPT 4 or 3 and doing it once and saying, oh, this isn't good enough for what I want to do and sort of writing it off versus now I think if you can show, hey, we're trying to actually write line by line or give you very specific insights. I think showcasing it like that versus saying, oh, here's the whole story of how you should do an interview. Which is even how we started has really grown. And then I think additionally, like especially with us, I always had thought we would be a sort of like pre production content tool because I actually originally when I was producing a podcast, that was the thing I was worse at great at selling ads, but I was bad at doing pre production and so shout out the guys here for, for doing all the production. But so I originally thought like, you know, maybe a lot of other people have this issue or it's a huge cost driver. In a lot of other production companies, they're spending a ton on producers. And I was like, maybe that shouldn't be the case or they should spend anything less or producers should be working on more shows rather than just like, you know, three producers on one show, which is typically sort of how it was going. You'd have a executive producer, regular producer, and associate producer. And now I've realized that, like attaching yourself from a business perspective to costs versus revenue is like a huge. And so we've been really focused on how do we make companies more money instead of how do we save them more money. I don't really think anybody's interested in saving money when they can just make more.
- [07:29]
B
So that's fantastic. Well, something I heard you say is they were just using it wrong or the AI capability wasn't there, or it's just not something they were able to spend enough time on before they realized the value. Now it's just much more apparent. But also you're bringing that to the forefront for these people who may not be spending 247 thinking about it.
- [07:49]
A
Definitely. Yeah. I think that, you know, I would never say, like, you know, our users are using the product wrong because they're using it how they think they should use it. Right. So they're always using it in the right way. We just have to adjust. And so I think that like a great deal of it was, you know, the insights that we were pulling were just sort of lackluster on what a really good producer could do. And I think as AI has evolved, and especially we've evolved with their customers to figure out what do they actually want. Because obviously me running a podcast, that's one podcast versus a producer who's done this for 10 years and works on five podcasts, is a little bit different in terms of the output they're looking for. And then also, like just doing everything through a data lens can be really challenging with creatives because creatives obviously are like, this is an art form to me and I know exactly how my audience thinks and how they work. And so really convincing them to like take a look at the data, which I think has happened a lot more recently when you look at shows like Diary CEO or even Caller Daddy, like huge data centered focus. I think a lot of it. Yeah, you can, you can tell. So like when you look at the comments, sometimes you'll see even on Diary CEO of like, Stephen doesn't ask enough follow up questions or it feels Sort of like heartless. And it's because he's like optimizing for retention. So he's not asking the follow up question. That's a weak transition. He's just cutting straight to the next question. So when you feel like, oh, I'm sort of done with this answer, you're already onto the next question. And I think, like, it isn't the natural flow of a conversation. So when you watch it, you're like, this guy's mom died and Steven didn't ask wasn't that heartfelt about it. And it's because he's like, how do we keep the viewer staying for longer? And so there's a give and take to everything. But when you're really optimizing for retention and for views, I think it does work. His channel has grown dramatically in the past year, year and a half, and I think a lot of it is because of their optimization on the data
- [09:32]
B
side, the willingness for a show to take on a data analytics approach and like, really, really, really, like, focus on that. Does it matter how big the show is or is it like literally any size show? Any podcast? This is alternative media, right? So, like, doesn't cost a lot of money just to produce one episode of a podcast. If you want to do it on the lower end, does it matter or does it matter?
- [09:54]
A
Does.
- [09:55]
B
Does the size of the show matter?
- [09:57]
A
Yeah, I mean, I think always the size of the show is going to show more impact on the changes that you do. I think a lot of the time that on a smaller channel, you're really looking for that human connection. So you're really like, hoping that somebody falls in love with the content that you're making. And I think the optimizations can actually happen later on. So for instance, if you're starting a podcast, what you're really going to want to do is just define who your audience is and make sure that they love your content, regardless of the optimizations. I think the optimizations come a little bit later when you've grown your audience and now you can say, okay, how do we grow to a bigger audience? Or how do we expand our pool? One thing I love about Mr. Beast is like, you'll look at his old videos. There's no optimization, retention, anything, right? He's saying Logan Paul for like a million times, right? Or something like that in a row. It takes him like 48 hours, something crazy. It's like there's no retention there. Like nobody cares about if you're going to watch the next second. And especially a lot of the early stuff is like, how do you just build that audience? But now that he has the audience, he's like, okay, how do we optimize this to keep people watching for longer, to attract more people? You know, he's got this purple cow theory. He wants to do stuff that nobody's ever done before so that you click on his. I trapped, you know, two pilots in a plane for a hundred days. The winner gets the plane. It's like, all right, there's not a video like that that exists on the Internet. And so then it's like, okay, how do we optimize? But initially you just have to have an audience that truly loves what you're making. And I think a lot of that just comes from being authentic and being yourself and then figuring out the, the, the optimizations later.
- [11:22]
B
It's no secret that people want to know everything about their audience and demo matters. What are the most non obvious characteristics of demo that you think have an outsized impact on a show? Well, let's walk through some, let's walk through some of the obvious ones. It's like geography, age, gender, what else?
- [11:45]
A
Yeah, I don't know if like, you know, if you have any control over your demo and really like, you know, obviously you can do paid media to a certain demo and market towards that demo, but I'm not sure if that actually gives you an advantage. I think there's like unique advantages to every demo and it's just about how you position, right? So like older demo going to have more money to spend, right? So you can say that to advertisers. Younger demo is going to be more interactive with your content. They're going to watch more of it. So I think there's pros and cons to each of that. What I would say is like you just want a demo that like is obsessed with you in any realm. The same way like you, you know, the older generation is obsessed with the news and you know, Anderson Cooper and whoever, right? They, they'll watch anything that he does versus, you know, you have younger audiences that will. One I say is a great one is this guy, Jake Shane. Jake Shane is probably, I don't know how many followers on Instagram that he has now. Maybe 2 million. But his audience is so deeply connected to him, will watch whatever he does and performance at such a high level where it actually doesn't matter that he has 20 million followers. His 2 million are real. 2 million that all love him as their favorite. And so you'll look at the. Forbes puts this out on Creators who made the most money. Jake Shane is continuously at the top is because his audience really cares about everything he does. So whether that's a pro brand promo or an event, his audience shows up versus you'll have other creators that have 10 times as many followers, but it's sort of like loosely following to where they don't really care. A lot of examples I give is like men or women that are very attractive but don't necessarily have that connection with the audience. And so they have a lot of followers because people care to see what they're up to, but they don't really interact with their content versus somebody that just really cares about their audience, their audience really cares about them and then has that massive pull. So I don't really think it's a demo that necessarily matters. It's just like, how strong can you get people to connect with you? How many people can you get to show up for you if you need? And so I think that is the, the main thing I would look for when like building an audience is how deeply are people connected to your authentic self.
- [13:50]
B
Do you put a lot of thought into like, literally people showing up for that, for that content creator that or for that person, like in terms of real life events or physical events? And does that even, is that within scope of what you're, what you're thinking about?
- [14:03]
A
Yeah, I mean, on the sales side, you know, a lot of the stuff that people are starting to sell is personal appearances for social media creators or comedians and anybody doing podcasts. And I think that, you know, obviously having that, that backing of like, hey, we can get, you know, 20,000 people to show up for a venue is huge. Versus like, I think we're going to struggle to get people to come to this thing. And I think you'll see that at, you know, big temple events, whether it's the All Star game or the super bowl or Grammy parties or whatever it might be that saying to a brand like, hey, we can get real pull here is actually the difference of millions of dollars versus like not having that in your back pocket. And I think there's plenty of creators that can offer that because they just can't get people to show up. But there's also plenty of creators that probably can get people to show up and should. And I think that's a big lever that's not being pulled right now when,
- [14:54]
B
when, when these creators are, are pulling on these, like, real life events. How often do you recommend or how often do they want to have some sort of third party sponsor as a part of that event always or are they just, there's, are they just saying, you know what, this is going to be somewhat closed off?
- [15:12]
A
Yeah, always. I think, you know, there's not a lot of opportunities for creators to make money outside of third parties or starting their own, you know, their own company, which is super challenging. And I think it's great that creators are now like full scale businesses. They are the marketing operations, they are the product. They are, and they create separate products. But I would always say like, you know, if you can get people to show up for you. I think like doing personal appearances with a brand is super underrated. I know obviously you have things that are, you know, maybe like a closed off thing, but I think that's because the creator just wants to have some authentic time with their fans. But I would say like when it comes to temple events, it's Coachella and stuff like that. Make your money, you know, so I'm, I'm all for that.
- [15:58]
B
Tell me a little bit about that. We were talking about that the other day how the use of social media hasn't really changed like one generation to another. And it's like the first time in history where you've got one, one older generation kind of doing the same thing that a younger generation.
- [16:11]
A
Yeah, I think just in general with technology, like you don't really see generations shift that quickly to new tech, whether it was software or the iPhone or social media. You know, with, with Facebook, I think a lot of it has been sort of slow growth and slow adoption. And now that I think, you know, with, with short form content, I think it's allowed plenty of the generations to sort of come together in the sense of like, you know, 8 year olds are watching TikTok and Instagram and 80 year olds are watching TikTok and instagram. And I don't think that's ever really happened in a technology shift. I mean even if we think about AI, we have this group of, call it 12 to 30 year olds that are all in on AI and then maybe a small section of tech executives that are really big on it, or just company executives, but that's really it. This middle section doesn't really exist. And then the upper section, like the 60 plus is not really like interested in using AI or if they are, it's a very slow adoption rate. And I think one of the things that's super interesting is like how simple and easy short form content has been and also how addicting it's been for good or for bad. But you know I will, I will say, I mean, you know, whether it's like my parents or my friend's parents, plenty of people are constantly, you know, on the use of short form content. And I think that's why the digital market in terms of the advertising world has grown so dramatically. You know, I think it's something like $700 billion on digital ad spend in the past year, which is insane.
- [17:52]
B
It's so large, you can't even really,
- [17:53]
A
can't even measure it, can't even like
- [17:55]
B
really appreciate how big.
- [17:57]
A
Yeah, and so I think it's just like, it's a really interesting time, both with AI, but even just in general on social media that you have like basically everybody in the world on one technology, which is a new technology. I mean, short form content is, you know, really started to pick up 2019. So I guess 2020. And so within six years you have probably 100% of the demographic using a piece of technology. I think it's pretty insane.
- [18:24]
B
How are sales teams differentiating themselves then in this market?
- [18:26]
A
Yeah, good question. I think a big thing with sales is like trying to prove that your audience has a unique fit with this brand. And I think a lot of it has been typically like, hey, X company, you have budget and we have a great product and there's not much overlap. And I think that we're starting to see this world of advertising with short form content, long form content, content in general really see a dip in perceived roi. I think for a long time it was like, oh, we'll have creators or we'll have podcasters, promote the product and everybody will know about us and we'll do more sales because everybody knows about us. So brand awareness. And I think now, I think brands are starting to realize like one, we have the power of social media too in our own right. We don't need massive creators. We can use micro influencers or hire our own marketing team to do influence where we don't have to pay them more than their base salary. And so I think you're starting to move to a world in which everything is turning ROI based. So those are like direct response brands. But even for the brands that aren't direct response or don't need to see a huge ROI like Pepsi or Coca Cola, what they need is like real proof that there's an audience overlap. And I think that's what a lot of people are struggling to do. I mean, if you're a sales team and you're managing 25 shows, good luck watching all of the content. Good luck looking at all of the personality social media pages to see, like, who's actually using X products, who's actually talking about certain things. It's just at the scale, even with the AI models, it's impossible to figure out all of this. And so that's something we're really focused on, is providing proof behind the pitch. So figuring out across people's socials, like, hey, maybe you're an agency and you rep a thousand creators and you want to look who's drinking Pepsi, right? And so you'll look and you'll say, oh, wow, we have 50 creators that in, you know, 40% of their vlogs drink Pepsi. Great, we can clip that, send that to the brand and really have a differentiating factor rather than just like, hey, this makes sense because, like, we have all these creators and you should do a deal with us. And then I think additionally, like, there's a world in which like sales teams need to get creative about the pitch. So like, that is like appearances, that is doing short form personal, which is just starting to, I think, ramp up. I think some companies, like our customer Wave, has been doing that for a long time and now I think sort of the general market has realized, like, oh, we need to both be selling personal and podcast. So like, you know, both on the company level, but also on their individual. And that's how we're going to get the most bang for our buck and figuring out deals to which we can get some sort of portion of that to really incentivize this full holistic integration.
- [21:00]
B
I see how brands may struggle to one, identify roi, or if the budgets are so big, whether that's even a priority. It's more about, like you say, audience overlap and real interest, where their brand awareness actually makes an impact. But for podcasters, for creators, the ROI is the brand sometimes, isn't it?
- [21:20]
A
Right?
- [21:20]
B
It's their sponsorship dollars. So how do creators think about roi? Is it only ad sales and sponsorships?
- [21:29]
A
There's starting to be this boom, at least with the bigger creators of figuring out how can I be a part of the marketing engine in a deeper way? Or how can I be part of the company in a deeper way? And I think that's allowing creators to ask for like, way better terms. So, for instance, like, hey, like, I'm going to promote this, but like, I also want equity in this company or I also want backend on my promotion. So, yeah, you're going to pay me a million dollars to this promotion, but then I want two and a half percent of everything I sell and so I think it's starting to get to a point where creators are recognizing their value if they have it right. Like, there's still a lot of proof to be had. But if you are recognizing that this is working, like, how do we get some sort of percentage of the upside? And then additionally, I think with creators, a lot of it is working with brands that you see a future with. You know, one thing that I said to a lot of the creators that we work with is there's no point to work with a brand, you know, for three months and then never work with them again, because you're essentially like branding yourself alongside them. And what you want to look for is brands that have, you know, that are on the up and up, that you can grow with. And as they grow, you know, maybe there's more opportunities for you to grow alongside them. And so I would really encourage creators to sort of like define what type of creator they are and work with companies that they truly love rather than just like for a check, which I think a lot of people fall into because they need to. And it's like, if you built a social media following and you want to do it as your full time job, you're going to have to make money somehow and the platforms aren't going to pay you enough. And so really figuring out, like, what brands do you see a future with is, I think, super important.
- [23:04]
B
While podcasting is alternative media, it's also been around now, I'd say, for a good hot minute, what does this look like in 20 years?
- [23:11]
A
Yeah, on that note, shout out to my dad. My dad had a podcast in the 90s. He would do interviews with Broadway producers and put it on a website, find the link. But yeah, shout out to my dad. I think in 20 years it looks super interesting in the fact that I think things are going to have to be a lot more personalized. I think that, you know, we might start seeing versions of creators or podcasters that are hyper specific to the individual. One thing I've been saying for a long time is, you know, Mr. Beast has 440 million subscribers, and I think there will be 440 million different Mr. Beast versions for each person based on the data. And so I think this has to be a platform overhaul for the platforms to be really looking and measuring the data. And I think the platforms potentially have the ability to now work together in a cross platform manner of like, how do we without getting too Big Brother, but like, how to attract people from one platform to another to figure out what content they're actually interested, what are they watching, and then to make sure they have the most personalized experience to them. And I think really there's sort of this like hockey stick curve moment that I think is coming with personalization. And so whether that starts out with there being four different MrBeasts for every 100 million people, I think could be an interesting play. And I think that you'll probably start to see that in the next five to 10 years and then hopefully 20 years. You're only watching the content that is most applicable to you and that you love every moment of it. And looking at the retention data to make sure every moment that you could be watching content, you're watching the best content for you. I think it's a little scary, but I think it's sort of the way the world's moving of like just personalized everything. I mean, you have AI agents working on everything to make your life as personalized and perfect for you. So I don't think content would be any different.
- [25:04]
B
Mass personalization, mass customization. Fascinating. But doesn't that go counter against one of the things that's most important between the creator and the audience, which is keeping like an authenticity or a trust factor? And if they think they're just watching some AI tweaked version of Mr. Beast, does that make it less compelling?
- [25:25]
A
Probably. I don't know if people will care is the thing. Okay. You know, I think like content, short form content, like it's like a drug. I think if you're just watching like the perfect thing for you, I don't know if you like actually care about the authenticity and that you feel this connection. And I think there will be creators that will be on their own and just like have that authentic connection, but I think there will also be people that don't exist that are creators. So AI creators that are making the perfect content for you. And so whether that's actually Mr. Beast or that's Mr. Beast style videos or whatever it might be, I think you'll have this duality of AI creators and regular creators and just let the best win. And I think that you probably will see just because of the bulk content that these AI creators can make that is much better content and perfect for you, you'll probably start to see a shift in adoption there.
- [26:15]
B
So I want to say it here. Just because the creator doesn't exist, I. E. It's an AI Creator, doesn't mean there can't be in real life events.
- [26:25]
A
Yeah, totally.
- [26:25]
B
I mean, there's holograms there's I mean, the robots. I mean, there's a lot of ways to bring an AI creator into the physical world on. In some sort of location.
- [26:34]
A
Totally. And I think a lot of these events that are based around creators, whether it's temple moments, so Coachella in the super bowl don't actually have a lot to do with the creator themselves. I think it's people that like the creator all being around each other. And I think the creator as the center point is super interesting. And maybe there's a meet and greet and stuff now, but it's actually not the driving factor. It's like, oh, so. And so is throwing an event, and great, I'll go and I'll support them. But also I get to be around everybody that's sort of like me that we all have this shared interest. So I don't see it why it wouldn't be different. And yeah, you could do holograms and all that stuff. And, you know, maybe 20 years we'll have robots. Who knows?
- [27:12]
B
In terms of super bowl halftime shows, this year's Bad Bunny like this when Bad Bunny, like, the hedges around his performance and people were literally in the. In the stands saying, I can't even see a show. But the show is epic in so many ways.
- [27:26]
A
Across social media, people.
- [27:28]
B
Okay, there you go. But you've probably seen a million clips on social media about it. And that's. That's the. That's the point.
- [27:34]
A
Yeah, I think actually this was super interesting. So I was there, and, yeah, I couldn't see anything on the halftime show. And so, you know, I'm like, watching the big screen, but I think, like, there's so many, you know, those hedges, like, funny story where we're sitting, we're like in the end zone and we're watching these people run out in, like, you know, hedges, like, in grass costumes.
- [27:56]
B
Yeah, yeah.
- [27:57]
A
And we're like, oh, my God, like, the hedges are going to start dancing or something. And we just figured out, like, I
- [28:02]
B
thought the same exact thing.
- [28:03]
A
Right.
- [28:04]
B
And I was waiting for it to happen.
- [28:05]
A
Yeah, totally.
- [28:06]
B
Never happened.
- [28:06]
A
Me too. Never happened. And I started thinking about it and I was like, wow, it's just so much more efficient than rolling out these giant, you know, hedge, you know, fake hedges, like, on wheels. Just like. Yeah, like, dress people up and then just have them run and it'll be much more efficient. And is that why.
- [28:21]
B
Is that why they do. Yeah, 100%. Oh, God.
- [28:23]
A
And just, like, think about it. You can have all these people just run and move and make this you know, stage without actually having to build the stage, which I think was really interesting and also created like some great social media content for them. But yeah, I think like, you know, everything is becoming hyper tailored to like the growth sort of economy in the sense of, you know, I think old super bowl halftime shows were probably great. Like you watch Michael Jackson or Rihanna, the, you know, even whatever that was five years ago. Yeah, it's like that's probably a great experience for the people at the stadium. It was a horrible experience for basically everybody at the stadium this year. I mean, music was phenomenal, but like impossible to see anything that was going on. And so I think like, you know, it's really saying like, hey, we're gearing this. And I actually thought this was like a great stance. But like, because it was so controversial and so many people didn't like it, it was like, we're going to make sure everybody sees it and, and so we're going to tailor it for the perfect TV social media content and we're going to blow this out because who cares about the 80,000 people that are here. We care about the 800 million people that are going to see this, which I thought was awesome. And so big bad bunny fan. First bad bunny concert 2017. So I'm an OG. Yeah.
- [29:42]
B
And human stages.
- [29:44]
A
And human stages. Yeah. Um, it was, it was really great. Yeah. What a show.
- [29:49]
B
Let's talk about the AI creator system overseas. What's happening here? China, generative AI generally. I know, I mean we talk about it all the time already just within our teams. How are you seeing that use in your own workflow? But also what are some competing forces there that kind of raise your eyebrows?
- [30:11]
A
Yeah, I mean, I think short form AI content in general is super interesting. I think you're starting to see more and more of it on Instagram TikTok. But yeah, especially these Chinese TikTok competitors. I think a ton of the content, maybe the majority is AI driven content. And I actually think there's a huge growing force there in terms of we don't really care if this is a human or not. We just want the best content. And I think that you're starting to see a lot of that. I will say sometimes I'll watch a piece of content on Instagram reels or something and I won't realize it's AI until halfway in. And I think the models are just getting that good. And I think even TikTok's competitor in China actually has a video model that they are building. Talk about Kwai or yeah, so Kwai, I think, has a AI video model within their app. And so I think it's just starting to become more and more. And so I wouldn't be surprised if TikTok then allows you to make AI videos and has their own model and stuff like that just to encourage the bulk of content, because these platforms are in the business of showing you the most content and keeping you on longer so they can show you more ads. And so it's to their benefit to just have more content on there because if you can service more people, they'll watch for longer. It benefits them. And so I know there's, you know, like meta has like, you know, AI content flags and stuff like that. And I think they're doing that to make sure people are safe. And there's like a difference of like, misinformation and stuff like that. But I don't know at scale if that's going to keep up because I think people will just want to watch
- [31:54]
B
or if it's even defensible.
- [31:56]
A
Exactly.
- [31:56]
B
I mean, it's like you can't stem the tide.
- [31:58]
A
Exactly. And so, um, I think that's super, super interesting. And again, like, if you're giving me the best content, for me, I don't really care if it's AI, if it's like informational and I'm getting like my latest news that's perfectly curated to exactly what I want. I don't care if an AI is telling me, like, here's the news and this is everything that happened versus, you know, some talk show host or some news anchor.
- [32:21]
B
Like, how are brands thinking about it though?
- [32:24]
A
Do they care either way? I don't really have a good answer. I don't have a good answer for that. Yeah, I don't know. If I was a brand, I could probably have a better analysis. But I mean, I think, look, in terms of how I see brands using AI is like there is so much marketing material that is now being pushed out. Whether it's Nike or Porsche, they're all using AI B roll. And so they're taking photos of the car, of the shoes and saying, with AI, let's make this look in certain ways. And so I think they're doing a lot of marketing around AI video and AI photos and stuff like that by. But I'm not necessarily sure how they're thinking about it with creators and podcasts, but I know the assets they've been able to make and how quickly they've been able to make them is unparalleled.
- [33:06]
B
Yeah, the Commercial industry has been one of the earliest adopters of generative AI and infusing that in the commercials and brands don't seem to have a problem with it at all. The economics are just really compelling. And so it's interesting how generative AI1, the quality of it, but also where it takes hold in certain parts of like entertainment and media has. It's not necessarily like a completely like linear adoption or like an all in one adoption. It's happening in some places faster and sooner than others. That's just a result and I think a factor of like the quality of what you're dealing with. And as that stuff gets better, I think one of the holy grails is going to be actual long form film. And, and once you start talking about that, I don't know, I don't think it necessarily replaces Pixar, tech or other types of film formats, but it's just, it's just another type of film that people become used to and start enjoying consuming.
- [34:01]
A
Yeah, they put Paul Walker in like one of those Fast and Furious movies after he was dead. So it's like, you know, that was cgi. Cgi, yeah, it's all the same thing, right?
- [34:11]
B
It's a different version, different tool.
- [34:13]
A
And I think there's a lot of like AI is catching a lot of flack for the, the name, what it is. But people have been doing some version of making things that didn't exist for a long, long time within content. Whether that's making something look like New York, but it was filmed in Canada, it happens all the time. And so I think just because it's tech doesn't necessarily mean it's wrong or it's different tech. And so yeah, it's been happening forever and I think it's just part of the process. And I think even for these animators, right, that their jobs getting taken. It used to take them like four years to make a movie. Now it takes them in terms of the animation, a couple of weeks. That's pretty awesome. Now they get to work on a bunch of projects. I know Pixar in general. They had to pitch three movies. Each writer or each animator had to pick. They had to pitch three different movies. And so now. And they would choose the best one. Right? And so now if you have a team of 10, instead of pitching three, why not pitch 30? You have all the tools to do so. And so I think it actually just makes content better and I think it makes the level of content you can create that much more interesting. And so I think it just benefits the whole industry and benefits the audience. And I think that, I think if it was, if it had a different name or if it was associated with something more positive, I don't think anybody would have an issue. And I think, you know, one thing I've been seeing is like these commencement speeches and how the graduating class is booing anytime AI or artificial intelligence is measured. Have you seen those?
- [35:43]
B
No, I haven't.
- [35:44]
A
Okay. So it's like, you know, it's like somebody speaking at UCF or at really like any, like any school. Yeah. And I saw this whole compilation. It's like they, they'll mention like oh, the growth of AI or artificial intelligence and the whole graduating class booing. And it's like, I think it's insane. And I think we are also trapped in this tech world and venture world and so we see all the positives, but I think a lot of people are just seeing this as a negative. And I would hate for that to actually be the case because I remember watching that and I was like, AI did your homework. Why are you booing it? This helped you get through college. So I think that's my take.
- [36:27]
B
That's great. No, to go back to even go back to movies even how I think certain types of formats of movies will come back into fashion. Like the original like When Harry Met Sally romcom like doesn't really get made anymore or if it not for the big screen, if it does, it gets made like an episodic or like a miniseries format or it's on Netflix and. And it's just a different totally format. But like something like that comes back into play when AI can totally generate the setting and in every single scene and it's a new and it's a specific spot in New York or in Paris or wherever and like suddenly like production value of the production costs of that film just go way, way down. And yes, you can make it faster but also you can make a different type of movie that people probably haven't seen in 10 years or 20 years
- [37:12]
A
or yeah, you can make a movie you want to make without the budget of the studio to give you, you know, $300 million to make smoothie.
- [37:18]
B
Yeah, there might be a way. There might be a renaissance and like the pull away from like Marvel Cinematic huge big budget movies where that becomes 90% of what you see in theaters and maybe goes back down to a lower percentage. But because other types of films come
- [37:33]
A
out, I mean I think something that's super interesting is probably, you know, it's not Content related, but just the, I think the overall job creation that will probably happen with AI.
- [37:45]
B
Tell me more about that. Because everyone talks about AI and like the scarcity mindset, like it's going to replace jobs, but I actually think it's going to. Anytime we talk about innovation, I think there's a, there's a net positive that happens.
- [37:55]
A
Yeah, I mean, I was talking about this last night at dinner, but like, even if you just think about like the H Vac industry, like, you know, many jobs are going to be created to cool all these data centers, like hundreds of thousands of jobs in the US to like in, in H Vac that will probably pay pretty damn well. Or you know, I heard a bunch of actually different people give this analogy, but about radiologists and it's like, oh, AI can now analyze X rays so all radiologists won't have jobs. It's like, no, people are just going to get more X rays or people are going to need to understand every little thing a tiny bit better. So there's just going to be more jobs or there's going to be more software engineers, not less. I think this is a huge one that everybody's like, oh, rest in peace to all the software engineers. Which I don't necessarily believe is true at all. I think there are so many things in our daily life that we could use software for that we don't because it's too expensive to build for a marginal benefit.
- [38:46]
B
It's a technical barrier of knowledge.
- [38:47]
A
Exactly. Or it's like, hey, that doesn't really matter to have software because like, I'll just get by and like plenty of things in my life that I'm like, could be improved by software that I'm just like, I put up with or I do the process because it is what it is and like there's not worth building a piece of software for it. But now because you can build software for anything, I think there's just gonna be 100x the amount of software needed to be built and software engineers will need to build that because in order to make it production grade, it can't be vibe coded. Like it just won't happen. I think what you'll see is a revolution of everything just getting software. And so I'm very excited about the industry in general. And one thing I heard Jeff Bezos say yesterday was this is like digging out your basement with a shovel and somebody comes along with a bulldozer and it's like, let me help you. And it's like, why wouldn't you Say yes, it's just going to help. Or even the fact that right now most households are dual income, but it might not need to be, it might need to be single income because you can just be that much more productive. I think there's this mindset that AI is going to hurt and I think that it probably will hurt if you think that or if you don't adopt it. Right. And I think it's like all a mindset. Like I, I just turned 25 and I was probably going to work a regular job and like I was not a programmer. I was like, I was going to be working in podcasting and selling podcast ads, which is a great business, like make great money. But I think the level of impact that I've been able to have because of AI and just adopting AI and it's like I get to work with phenomenal people, hundreds of people that are impacted by our product and by what we do is the most rewarding thing ever, money aside. And I think so many people don't realize you can change thousands of people's lives by saying, wow, I have this pain point. If I solve this for myself, I can probably solve this for a lot of people. And I think it's never been easy to start a company. I think it's never been easier to build a product and get it out to millions of people with short form content. And so I would say all of these new grads that are booing AI is you have the opportunity to one make a shit ton of money right now because you are the right generation for it. But also you can just benefit your own life. If you're going to go work a corporate job, build AI agents to automate the stuff you hate about your corporate job, there's a million things to do. And so I think you just have to dive in because I think this is the biggest revolution in history. And you know, I've been going to like AI conferences for like four years before like AI, I remember thinking like,
- [41:27]
B
I'm sure the nature of those conferences has changed even four years.
- [41:29]
A
Oh my God. It used to be like AI is the biggest thing that's going to happen and now it's like, okay, it's here and we like thought it would be like a quarter of the size. And I think like this is, this is in my opinion 10 times as big as the Internet. I think this is going to be every company. I think you're actually going to start to see a, a breakup of these semi large companies. I think there's going to be these huge companies which are These model companies, OpenAI, Anthropic, Google, et cetera. And then you're going to see all of these small niche companies in AI and in software that are much smaller market cap. And I don't think there's going to be any $500 billion software companies that exist because they're just going to get crushed by the models or they're going to get crushed by the really small company doing it better. And this is, I think this is a big thing. Like everything that can be like hyper vertical is going to exist and then everything that isn't is just going to be these large language models. So insurance, large language models, but you know, maybe specific type of, you know, unique insurance that's like a super one off, like very much these small companies
- [42:39]
B
or even in like podcasting classic cars insurance.
- [42:42]
A
Exactly, yeah. And so I think you're going to have these model companies, these five model companies that are going to be worth tens of trillions of dollars. I mean Anthropic is going to be worth a trillion dollars this year. They're going to do a trillion dollars revenue in a few years. And then you're going to have all these smaller companies that are super hyper vertical and have the ability to just service their customer better than large language models. But the companies in the middle that are just doing regular software that are now worth couple hundred billion, I don't think stand a chance. And I think it's a very concerning time for those companies. But I think everybody else that's fresh, it's like, how can you pick something that's a smaller industry? And I think the investing world too, from your side as a vc is there are companies that are going to be worth trillions of dollars and there's probably companies that can just be worth billions of dollars and that's probably it. Like there's, there's not going to be, you know, hundred billion dollar companies in software. I think it's just gonna be the models and like everybody else on the smaller side. And so it's really interesting time to be like a semi large company. And I wonder how Salesforce and all these other companies feel. And I think like, you know, they obviously have a great integrated product that's very sticky. And so I think the like short term is probably like pretty bright. I mean the revenue keeps going up and up and up. But I wonder like you know, 20, 50 years from now, what does that look like? Because obviously this breakup of like the small companies and the massive companies isn't going to happen in the next five years. People are still stuck in their workflows and stuff like that. But slowly like can have anthropic make you a CRM. You can have it do all your automated emails, you can have it build your entire Salesforce pipeline, you can have it do data enrichment, you can have it do everything. So it's an interesting time. Or like, you know, Figma, like stuff like that. Like Figma is great and it, you know, helps companies ship software. And I actually like, I'm sort of bullish on Figma. I think nobody is but like because of the software boom, I think you all need some version of Figma because you know, cloud design is great but you still need it to be productized. And I think Figma has a lot of room to grow. But even stuff like that, like it's not a $100 billion company, I mean the market just showed that. And so I think you're just going to have these companies that are like super niche and these companies that are extremely general and nothing in the middle.
- [45:04]
B
I think if you have nothing in the middle, what you have is really a larger SMB economy. So small medium sized businesses that oscillate between making a run at growing but also our lifestyle businesses run by you know what, we would have considered mom and pop maybe 10 years ago but being run by mom and pop at that level and they become lifestyle companies that are actually profitable and it's just a different type of middle class to anywhere from like lower to middle to upper middle class that, that comes into existence and, and if that's the case, you know, I, I think you have to think about how that eventually gets consolidated because there's just always going to be a, like a recycling, there's a cyclical play when you know, you have, when you have a decentralized set of companies and they're all maybe hyper, hyper specialized. I think you always have another person that comes in and says how can we combine, how can we aggregate, how can we optimize even this? And so know, I don't know if that happens when you have the, you know, tens of trillions of dollars type companies that are in the background. Because especially in, in, in advertising, in advertising technology a decade ago, one of the most common refrains when an ad tech company was pitching an investor is what if Apple decides to what, cut you off, cut you out? What if Google decides to cut you off or cut you out? And that, that's the same exact narrative is going to begin happening where like great, you're this type of hyper specialized, vertically vertical, verticalized company. But what if you lose access or what if anthropic or anthropic or open AI decide to do what you're doing tomorrow? And, and, and so there's always a risk there. But what I think people realize if, if you kind of maybe go a level deeper is that Apple and Google
- [46:58]
A
never really kicked a lot of people out.
- [47:00]
B
And so it actually ended up being like a huge net positive and an additional part of the economy that, that was able to, to just to live.
- [47:10]
A
Totally. Yeah. I think, you know, obviously like the, the model companies need the small businesses to pay their APIs and to keep them alive. Like it's where the majority of the revenue is coming from, is from businesses. And so I think the, I'd be less worried on like the niche stuff of like what if they cut you out but what if they do what you start doing or what if the models become so good that your niche company doesn't matter? I think is super interesting. I do think though that like the, the big company is never going to be able to service the specific exact customer and make them feel like this product was extremely built just for them, which I think is the difference. And I think customers will be willing to pay for like this product was made just for me. Which is why I think these companies will be smaller because you can't make that many just for me companies. And so like you know, anthropic is not going to come and take your, you know, like even like my business, like the market's not big enough. Right. But like are they going to do insurance? Are they going to do a healthcare? Absolutely. Are they going to do finance? Absolutely. There's a million of these other verticals that are trillion dollar industries which is why like I'm sort of bullish on this take is like you need to pick a niche, you need to monopolize the niche and figure out growth later. But you need to be really good at doing something that yeah, maybe short isn't the biggest market in the world, but that you can make the for me product. And that's sort of what I say is like I want the, the customer experience to be like wow, they made this in house for us. Like that's how I want it to feel then there's only so much you can grow with that before you have to change up the way you think about your company and, and doing products. But it's also why the companies like Palantir and all these Forward deployed companies that do really well because they go in, they look at your business, they spend a couple of months doing it and they say, okay, great, we know how everything works, we're going to make you custom software for you. And so they're actually like, you know, Palantir is like a consulting company, like they're a software consulting company. There is, you know, as much as Deloitte or McKinsey that come into your business and okay, yep, we know what's wrong, we know how to fix things and we'll make you software. Yet they're also getting you know, the ARR multiples like SaaS companies, which is kind of crazy but I think that's sort of where you're going. It's just like how can you make custom solutions for like specific companies and specific like niche industries? And then I think sort of what you were saying on, on that growth side is I think that like these companies are going to be huge and I think that like they're going to be more nation states than they are companies. And I think that like it's, we're playing a very dangerous game where like you're giving the power to like five people and it's just like you get into a point where like they don't want to be public companies because being public company now sucks and their value. Yeah, exactly. But I mean I think SpaceX would be a private company if they thought they could raise money at 1.75 trillion in the private markets. They just can't. But you have anthropic that's going to raise 900 billion where in the public market it probably can't get 900 billion but in private markets definitely can, no problem. And so I think they're going to try to stay in the private markets for as long as possible which leads them to just have less customer favorability when it comes to like feeling a part of this company. I think that like you'll get to a point where you know, like if anthropic or opening I was a public company and people could invest and they'd be happy for the growth, they'd want to see more and so you wouldn't have I think so much like booing at graduations because people, people could feel like they could take part in the upside. But I think they're going to try to stay private companies for as long as possible and you know, potentially not even when they go to fundraise in the public markets, not raised that much. I think SpaceX is only raising 75 billion out of 1.75 trillion. And so they're not giving up really any percentage of the company. And so I think it's going to be really interesting to look at how these model companies operate and then also how they give back in a way to the people, or else there's going to be riots. One thing I heard this week was when Hershey's was making their factory in Hershey, Pennsylvania, they made a theme park. Amusement park. Right. Like they have like roller coasters. It's like, yeah, sorry for like creating all this traffic, noise and pollution in your town. So here's an amusement park. It's like, what is the equivalent to that for the model companies? Like, hey, sorry for putting a data center in your backyard and sorry for like creating all this havoc. Like, what can we do to give
- [51:28]
B
back, like universal basic income?
- [51:30]
A
Exactly, exactly. Exactly. But like, exactly.
- [51:32]
B
Well, as these companies get huge
- [51:36]
A
back
- [51:36]
B
in 2022, 2023, there was actually a lot of discussion around how actual nation states would have their own nation state models, their own AI models. I think that conversation's gone a little bit by the wayside just with the growth around the privates, like anthropic and opening eye. But that could very well just come back into the dialogue, into the fold. These companies are getting so big that maybe they're just partially owned or they're heavily regulated to the point where it becomes like a phone company.
- [52:10]
A
Yeah. You know, I use, I used to think there was a scenario in which OpenAI could no longer raise enough money to continue to grow. They wanted to grow with, with their burn and so they were going to have to like, rely on the government. I no longer think like that. I like, I mean, I used to think like this six months ago. They're like, yeah, I think they're just going to grow so fast and I don't think they're going to be able to raise enough money and they're going to be the most powerful company potentially in the world. This is before Anthropic obviously went on tear. But who's going to fund them? The government. And then the government's going to have an AI and then you're going to be asking all your personal questions. And so all of this stuff in the same way of. I think there was, a couple of years ago, there was this fear that, you know, quantum computing would basically like break everything. Break everything. And so that like, you know, it would break all your communication systems. And so, you know, there was this scenario I had thought of in my head which was like if it happens and for instance, like you can't send any encrypted messages and. Right. Everything is just out in the public. The government has to create some anti quantum computing phone. Right. So now you're on a government phone. Right. And now they're listening to everything you say. And there's a world in which we're
- [53:22]
B
back to where we started.
- [53:23]
A
Exactly. There's just this, this world in which this like it's an infinite loop of just like, you know, corruption of like, you know, privacy and like I think there's, there's a world in which like there's stuff like that that's scary and I think these AI companies are like verging on like the, the scary side for like most people. I'm obviously like super bullish on, on these companies. I think these companies are great for the world, but I don't think if they don't start giving back to the public in some sense, I think there will be riots in the streets for real. So it's going to be interesting.
- [53:58]
B
I mean short term disruption would lead to higher rates of unemployment across younger populations, which is always a recipe for volatility and also a recipe for more socialist mindset, I would argue.
- [54:15]
A
It's been a pretty fun conversation. I know.
- [54:17]
B
I mean you speak at like a thousand words a minute.
- [54:19]
A
Yeah.
- [54:20]
B
So we got a lot here. Well, it helps when you're an expert in something and you've devoted your life to something to be really, really good at it. I think you speak very, very quickly. Anybody would speak very, very quickly because there's just a high degree of confidence and knowledge around it. So I think that comes through. So thanks for the time, thanks for coming out and maybe we'll do, we're going to do, we're going to do another one of these.
- [54:42]
A
I'm in. Cheers.