Want to join the conversation? Connect with us on Tribe.
Find “Founders in Arms” on Apple Podcasts, Spotify, and YouTube.
Transcript of our conversation with Trip Adler:
The transformative rights, sounds really valuable. For example, something like Harry Potter has so much fan fiction that's generated off of it, that people love just to extend the characters and extend the world. I can see that being tremendously valuable.
[00:15.9]
And yeah, you can make a video out of fictional worlds and stuff. Yeah. Like with Harry Potter, you can imagine just how much you can expand the Harry Potter universe using AI. You want to make sure you build a model that expands the monetization potential of Harry Potter versus, restrict the monetization potential.
[00:33.3]
So that's what we're, we're trying to build and then, you know, bringing along rights holders, and AI, companies or building these things ourselves to sort of just, you know, create this, this future opportunity.
[00:45.5]
Hello everyone. Welcome to the Founders in Arms podcast. I'm, Raj Suri, co founder of Lima and Tribe. And today we have Trip Adler, founder and CEO of Created by Humans. Immad can't make it today. He has a flight that he's taking, but he'll be on next time.
[01:02.7]
Welcome Trip. Glad to have you on. Thank you. It's great to be here. Very excited about your new company. You're a longtime founder, right? When was, the first time you started founding companies? Let's see. I started my first company when I was about 21 years old.
[01:17.9]
I was a senior at Harvard. That was around 2006. We did Y Combinator, class of 06. That was a different era back in Cambridge. And yeah, built that company up to a pretty big company over 17 years and now I'm onto my second company. Yeah, you and I probably started around the same time then because I started my first company around early 2007, around when I was 21 as well.
[01:40.8]
So it's, same vintage, I guess. Yeah. It's been a good era for startups. Yeah, it's been crazy. Starting to get better too. Yeah, it's been great. And it's crazy to see how the ecosystem has exploded. Right? I mean, we, I thought it was big back then, but it's just like a thousand times bigger now.
[02:02.6]
Yeah, no, it's, it, it's really exploded and I, yeah, I think it's going to just continue. I, it's, it seems to me like doing startups is just going to be the new path for a lot of people and there's enough infrastructure in place to, to support a lot of startups. So. Yeah, I think it's just, it's still early days of the, of the big startup wave that, that we're on.
[02:20.6]
Yeah, I, I was, I saw something the other day. I think it was a tweet that said that Stanford kids coming out of school now find it easier to get into Y Combinator than they do to get an internship at a, at a company. Do you believe that? Yeah.
[02:36.3]
Then I saw, I think, I think Jared, the partner there is my former co founder. He yeah, he responded saying, explaining, explaining that I guess it's, it's, it's if you're, if you're dropping out of school to start a company, that's a very good signal compared to looking for an internship.
[02:52.1]
So, so I think it's just kind of like those people who, who are willing to drop out of school and start a company. That's a great signal for Y. And those folks do get a lot of opportunity. Yeah, especially I mean YC Index is towards the top schools as well. There's no doubt about that. Right. So it's like if you're going to Stanford, MIT or anything like that, then you have a much higher chance of getting into yc.
[03:10.9]
So yeah, those people have that option. Not everyone does, but those people do. And Stanford also acts as a filter. But yeah. Tell us about your new company. Created by humans. It's a big, big topic and very timely. Yeah.
[03:26.6]
So Created by Humans is working to solve Copper, in the AI era. It's a really big topic, it's really complex topic and we are trying to do our part in solving it through a startup solution and by getting a marketplace going for AI rights.
[03:43.6]
We believe that if rather than having these copyright battles all just play out in court, if we can actually build a model, a licensing model where creators can proactively license their works and AI companies can come and agree to terms that we can kind of connect these two through mutual agreement.
[04:01.4]
We have just a better approach here than through than through, through lawsuits. And you know, our ambition is just for all kinds of copyrighted content. We want to be the connective tissue that connects copyrighted content with AI. But we're starting specifically with books.
[04:16.5]
You know, books have been very central to this whole, this whole situation and there's a lot of tricky things to figure out with books. So our goal is to solve this with books and then we're going to expand other types of content from there. Yeah, yeah. And a very important topic, I still, I think very much unresolved to this point now, there was a big court decision, right, A few weeks ago.
[04:35.9]
I was thinking of you when I read this court decision. Well, there are a couple that came out. There's the, the one about anthropic using books and the one about meta using books. I mean these are just, there's, there are dozens of these types of lawsuits and I think that these are just two of the many more to come. So I think it's still early days and I think all these are yet appealed too.
[04:54.4]
So I think the legal process here is going to take a long time to play out any, any resolution. But there was, you know, there's some preliminary or there's some judgments on those, those two situations which you have some preliminary input on where this is all going to go. I mean the way, the way I read the outcomes, it was, it was pretty, it was pretty, pretty unclear where this is all going to go still.
[05:17.6]
You know, there was this like kind of a bit of consensus that like you know, training and just training is fair use. But there's other questions about like, well, but questions about how you get the books. So for example, in an anthropic case they're saying you can't pirate books, so there needs to be penalties for that.
[05:35.2]
In the meta case there's this question of over time, will there be market dilution if these AI companies are training on books and replicating the ideas outside of books? Because, there is market dilution. That's going to be a signal that it is copyright, infringement.
[05:50.7]
But basically it's really still kind of like it's still playing out. I mean these are two early judgments and I think that there's just going to be, we're going to learn a lot more, more of these cases over the coming years. Yeah, I saw a lot of AI pro AI people celebrating the anthropic one, right?
[06:05.9]
Saying that hey, it means that AI folks won't be held liable for just training on whatever they want. But that's not really the case. It's kind of what you're describing. Yeah, I think it's more nuanced than that.
[06:22.1]
I think the big one there is that they said that you can't pirate the books. And from what I read, Anthropic pirated seven million books, which is a lot of books. So you know, there will most likely be a judgment that, that, that anthropic is going to have to Pay for for each of those books and that that could be pretty big numbers.
[06:40.6]
So so yeah, I think, I think people are celebrating that maybe like yeah, training and just training by itself is fair use but the other whole, all of their actions around the whole experience of training really matter too. And I think also eventually what you're going to do with the content and how you monetize it also is going to matter.
[06:56.2]
So so I just really I, I, I think people were, were taking kind of binary positions this but I early and it's still unclear where it's all going to go. Got it. What's your progress so far in terms of how many you're building a two sided marketplace, right.
[07:13.2]
In some ways you're building the AI side and the book side. What's your progress in building this market? Yeah, so we started on the supply side. We figured we need to get some content available for license and we went to the, the book publishing industry.
[07:30.6]
They quickly, the, the, the publishers redirected us to the authors because the situation there is that since all these publishing contracts for the last you know, many decades have never mentioned AI rights, it basically means that the authors are the ones who still own their AI rights. So in order to unlock these AI rights, you have to actually get authors involved.
[07:49.7]
And, and that's why these lawsuits are actually by authors because it has to be the actual rights holder that files a lawsuit. So we built it as a direct author product. We partnered with the Authors Guild and we allow authors to, to to sign up and claim their books, select their licensing options and, and license their books.
[08:06.0]
And we've we've gotten quite good traction. We've got a lot of, a lot of really big names, and a lot of and just a lot of authors, a whole diverse group of authors. So we've got a nice little library available for license. And then we are in conversations with companies.
[08:23.0]
We've we've signed some deals, we've got many more in the works. We yeah, there's a lot going on so we're gonna have more to announce soon. But I think we're still, yeah, we're still building this marketplace. It's a really complicated marketplace to solve with a lot of kind of nuance issues on all, all parts of it.
[08:39.3]
But but we're you know, we're making good progress. Sounds good man. And I can see on your Website. Some are public. Right. So Walter Isaacson with the Steve Jobs books, James Patterson, like these, these are some of the authors, you know, presumably you've signed. Right? Yeah.
[08:54.9]
And like Doug Preston and Susan Orland and Sylvia Day, those are just similarities we put up on our, on our homepage to kind of like, to just to sort of, Yeah, it's got to showcase some of the folks we have. But there's a lot more and library is growing every day.
[09:11.6]
And yeah, we, we hope over time to have you know, a library of. A growing library of content where authors are, you know, willing to, to, to share their, to, to license their books as opposed to, just filing lawsu. You know, we're trying to flip this to getting people to share their books for license versus versus this taking legal action.
[09:31.8]
And yeah, I think we're making good progress. When we started this, the, the, the, the author community was very divided. Some people just didn't want anything to do with AI. Some people were kind of on the fence. Some people were enthusiastic. We're kind of starting with the enthusiastic ones and then we're going to expand, from there. Yeah, I mean it makes a ton of sense.
[09:47.3]
This sounds like very much like the music industry going, you know, and then eventually ending up at Spotify. Right. You know, it was all lawsuits in the 22 thousands and then ended up with the legal path to getting to monetizing their streaming rights. Right. Which people still complain about, but that's.
[10:04.5]
Everyone seems to be relatively, it's a relatively stable place. Right. Where Spotify. Is that a good analogy? Yeah, I think it's a good analogy because with you know, with Spotify there was, there was the Napster era. There was a lot of, you know, legal questions about that, lawsuits. And Spotify solved it by just building a really good product that got the labels and got the artists on board, and got consumers on board and bringing those two together.
[10:27.0]
And I mean that's, that's what we're trying to do is get. Build a legitimate model that everyone is, is on board for. And See, I think Spotify is a good analogy. We're trying to be like the, the Spotify of, of content of the AI era. Yeah. Maybe while they're doing that, you can also build on a consumer product to consume the book so you can compete with Audible or something, or you know, some other types of.
[10:48.3]
Yeah, we're going to be building all sorts of stuff. I mean, we've built, we've built a licensing product, we've built an infringer detection system so we can monitor the LLMs to see which books people have and that information available to authors. And we are actively thinking about building our own consumer products.
[11:05.3]
I mean there's, there's so many ideas and yeah, so whatever ideas is like just become a customer of our own, of our own marketplace basically and start building consumer products. So we're, we're, yeah, we're building a lot as a company and we've got some good, good stuff in the works. Sounds good man. Sounds good. And so you mentioned that this, this idea could extend to other segments and industries.
[11:27.8]
It kind of makes sense, right? Like the movie industry has had a lot of controversy over a, obviously music itself, has had a condo. Which industries do you see as like ripe for this type of disruption? As well as books? I think it's all human generated work and all copyright work.
[11:45.9]
So that's. Yeah, videos, music, art, podcasts, I mean, yeah, really, really everything. And you know we, we eventually want to work with all kinds of industries, all kinds of content, and be the connective tissue between all copyright works and AI.
[12:02.1]
And when we started this I thought we were going to expand pretty quickly, but as we got into books, I just realized there's just so much to solve with books. I think we're in a stronger position, really kind of, you know, succeeding in that one vertical, solving the problem and expanding from there. So we're laser focused on books. But once we, we you know, we make some progress there, we're going to expand and do this for all kinds of content.
[12:23.6]
What do you think that would look like? So let's, you know, right now a lot of content is created on TikTok and YouTube. You know, so there's a lot of like self created content, even twee tweets. Right? Like do you think, do you think, say if you're a prolific social media influencer on Instagram or Twitter or TikTok, do you think that like you start getting paid per volume of content?
[12:44.5]
Something like that? Like what would the numbers look like? I'm just curious, what do you think this would end up being in terms of like in the range, of course, dollars and cents, like you produce a thousand tweets, do you get $5 or something like, you know, for your AI rights? I mean, I have no idea what the numbers are going to be. I mean, the, the market prices here are just.
[13:02.3]
It's all over the place. I mean, some people believe that, you know, creators should not get paid anything for this. Some people believe it's like this should be, you know, data should be the most valuable resource in the world. We think the answer is somewhere in between and the market's going to decide. And, yeah, so how much a particular Twitter user gets.
[13:18.0]
I, I just don't know. I mean, I think it really also just depends on what exactly type of content we're talking about. So for something like Twitter, I, I think that it's really, the, you know, the, the Twitter is the only company training on that data. And there isn't really a way for other companies to, to access it.
[13:34.1]
I think from what I could tell. I think, I think Elon's been cracking down others, using that data. So, so I mean, we could potentially have a way that an individual Twitter user can export their tweets and then monetize, Monetize those in other platforms. There's things like that we could do.
[13:49.8]
Yeah, we just, we just haven't gotten there yet as we're still. Yeah, yeah. Very focused on books at the moment. Yeah, I'm just curious about, how the industry is going to evolve. On Twitter they have an API, actually, which we use at Tribe. And, you're able to suck in any data you want from that.
[14:08.1]
Oh, really? Yeah. And so it's a very expensive API, but you can do it. So maybe that's where other companies can train, but maybe there's a rights issue for big company like OpenAI not being able to, to train. You know, there could.
[14:23.7]
I mean, you know, at that scale, you're going to be worried about lawsuits. Right. So, yeah, so I think, I don't know what the rights are, but you know, we'd have to look at the rights, man, because our general attitude is we want to make sure that all rights holders are signed off. And since these AI rights have often been poorly defined, we want to get the actual humans who own the rights behind, behind it too.
[14:42.8]
So we could potentially. Yeah, we'll look at it and try to understand the rights of the, of the actual Twitter user versus versus Twitter and try to sort that out. But, but yeah, as I said, we just haven't gone down that path yet. Yeah, I was talking to a bunch of other founders about X now is part of xai. Right.
[14:58.8]
And it seems like the main value of Twitter now is to train their AI. Right. It feels like that has become the main thing. Right. It's not like advertising anymore or anything like that. Do you see that? I mean the quality of that data doesn't seem very valuable.
[15:15.9]
How do you look at the differential quality of data? Right, so books seem very valuable. It's very structured. Right. Especially like non profit, I mean nonfiction, books have a lot of useful things but even fiction books can teach AI LLMs how to think or how to like write.
[15:34.6]
But Twitter stuff just seems like a bunch of random crap to be honest. There's probably a few good news articles. It's useful for real time stuff. But LLMs don't really care about real time stuff as much. Like how do you look at the differential quality of like videos versus tweets versus books, you know, retraining LLMs?
[15:52.6]
I mean yeah, we, we, it's hard for us to say. I mean we're just trying to build the marketplace and then we're going to try to get the, the best market rates for all this type of data. And I think, I think if you, maybe if I came back on the show in a few years I'd have much more of a question, much of an answer to that question by being, by being further along.
[16:08.9]
But yeah, our point of view is that like longer form content that's highly edited, things like books should have a premium value. It's, it's, you know, there's a lot of people go into creating a book and the long form aspect, the structured aspect really increases the value. And the other thing is we're not just doing this just for training.
[16:25.3]
So we've actually, we've, we've created three categories of rights. The first is training. The second we call reference rights which is using content for rag. That's actually where we're seeing a lot of interest in is getting books for rag. And then the third is what we call transformative rights, which is taking a piece of a piece of copyright work and transforming into something new using AI.
[16:47.4]
So for example this could be taking a book and you know, talking to the book or talking to a character of the book or converting the book into a video or personalizing the book. There's all sorts of things you can do with transforming a piece of copyrighted content into something new. So that's another aspect of our marketplace.
[17:03.9]
So we have these three categories of rights, and they all, they all work kind of differently. And I think for something like, you know, books, something like transformative rights are a really good fit. Well, for Twitter, I think transformative rights wouldn't make near as much sense. So I think that there's. There's all sorts of potential for books and for other kinds of content, just depending on the type of content, the type of licensing.
[17:24.4]
And, yeah, we're just trying to build sort of the plumbing to enable all this to happen, because right now it's like, because it's so complicated and it's not defined, people are just pirating things. But we think that if we make this, if we, if we kind of define clear rules and make it really easy to get the data you need, that's how we kind of solve this problem in the long run.
[17:40.3]
Yeah. Yeah, that's fascinating. I had no idea about those different types of rights. The transformative rights, sounds really valuable. You know, like, for example, something like Harry Potter has so much fanfiction fiction, that's generated off of it, that people love just to extend the characters and extend the world.
[17:58.8]
I can see that being tremendously valuable. And yeah, you can make a video out of fictional worlds and stuff. So, yeah, that's very exciting. Yeah. Like with Harry Potter example, you can imagine just how much you can expand the Harry Potter universe using AI. And you just need to build.
[18:15.5]
You want to make sure you build a model that expands the monetization potential of Harry Potter versus, restricting the Monetization. So that's what we're, we're trying to build and then, you know, bringing along rights holders, and AI companies or, or building these things ourselves to sort of just, you know, create this, this feature opportunity.
[18:33.4]
Oh, yeah, yeah, you should definitely build it. There's going to be. If you do, getting the rights is like, you know, probably most of the work. The, the, the actual, the, the building part, I mean, that's probably the more fun part for sure. You know, we're definitely going to start building our own, consumer apps.
[18:50.6]
And like, you just said the hardest part was clearing the rights. And that's kind of what we've done in this is kind of like sorted that piece out. But now that that's done now it's like lots of, lots of opportunity for partners to build or for us to build using all this, this data we have. Yeah, especially at Silicon Valley founder, like, you know, we all know how to build stuff.
[19:08.9]
But, but like Very hard to negotiate. Right. So like at least I see that as a big schlep. Right. And you, you know, you've, you've kind of done that part already. So and yeah, then the other. Right, the middle one, the rag, like using books for rag. How does that work?
[19:24.6]
Maybe you could tell some of the orange parties understand how. Yeah. When you hear like about you know, let's say OpenAI doing a lot of these media deals, they're, you know, they're, I can't comment how the deals work but, but often they're, they're what they call a display deal.
[19:41.7]
It's about like having the content show up in an output and then linking to the, the source of that content. So a lot of these news deals are here like news news news companies and companies. Those are what they call display deals or, or rag deals. And that means the content was never used for training.
[19:56.8]
It was just used for, for like, you know, know, pulling the content for rag in real time if it needs the information. So we're finding that there's quite a bit interest in rag, rag deals for books. So you know, if, if somebody asks a question, the answer to the question is on page 224 of a book.
[20:13.0]
It can kind of pull up the answer from that, from that page and, and say here's the answer and here's the book where it came from. There's been quite a bit of interest in that kind of deal. Yeah, that makes a ton of sense. Okay, I get it now. And what do you think about how actual book, writing will go from now on?
[20:32.0]
Do you think there'll be a lot more books created now with AI? It feels like it's much less friction now to write a book. AI could write a book now in seconds. But, do you think that will mean more people writing or do you think the people won't? Because the value of books will go down.
[20:47.1]
It's interesting that Spotify, in today's world, we're talking about Spotify, there's tons more music produced now, now than there was ten years ago. I found that fascinating. It's like some crazy number of like tens of thousands of pieces of New music are released every week. Do you think that will happen to books as well?
[21:04.0]
Yeah, well, this is, this is one of those things where there's like More. There's, there's infinitely more reasons and it's even more easy to write books. But there's also lots of reasons to write books because, because of people reading AI Output. So it's like, who knows what's gonna happen? It's gonna be. It's. It's like, it's a. It's a new world we're entering.
[21:20.8]
But I would say it's like, you know, we want to create a world that incentivizes people to keep writing books. What we're concerned about is if the business model of publishing goes away and, people are just getting all the information they need from AI outputs. There isn't a business model incentivizing authors to write books.
[21:38.2]
And we think that's just not a good world to live in. Right. We want to be the company that champions, human creative work and champions the business model for that so we can keep creating great things by humans, forever. Right. And, you know, if we get into a world where the world of AI and the world of human creation gets too siloed, that's not gonna be a good world.
[21:55.5]
So we're trying to bring these, these two worlds together. Yeah, I mean, we're. We're thinking about like, also like, you know, if you look at a book, you know, in the future, like, parts of the book could be written by a human, parts to be written by A.I. like, do. Do humans wanna. Do they prefer to read something written by a human or prefer A.I.
[22:12.0]
you know, we're trying to like, answer those kinds of questions and then, and then sort it out so that. That people do get more comfortable with. With a book that' written by. Partially by AI So, those kinds of things we're figuring out. And I mean, these are. These are big questions without clear answers.
[22:27.3]
But, you know, we're positioning the company in the space to try to like, solve this and, and just, you know, bring, bring everyone along and make sure that, you know, humans keep writing books in this new world. Yeah, yeah, totally. I think that makes sense. You know, the interesting point. I'm going back to my last point about music.
[22:43.2]
You know, AI can now make a song very easily too. But there's no AI song that has become popular. Right. Like, there's this idea of, like, the. The AI stuff all becomes kind of generic. Whereas the music that. That pervades our consciousness tends to be fresh and new and deeply human.
[22:58.4]
You know. And, And there's just something stale about AI you know, AI Generated content. I think the same might be true for writing as well, especially fiction writing. Non fiction often needs to be fresh, to be, to take off the needs ever a differential point of view.
[23:15.2]
Right. So it's hard to do that with AI, but AI can help with like some, you know, a lot of times they say like a book could have been a blog post. Right. So like you have these, these like a few fresh ideas and then AI can help expand on that maybe. Yeah, well that's the kind of stuff we want to do.
[23:30.5]
Right. It's like if you're reading a book, maybe, maybe you want to go through part of it faster or maybe AI can help summarize that as you're reading it. Or maybe a part of it you want to go deeper and AI can help expand on that. Like these are the kinds of things that we want to enable with books. But yeah, I mean I think that if somebody, you know, there are people created 100% AI generated books and from what I can tell, no one really wants to read them.
[23:50.6]
But I could imagine though where somebody writes a book that's mostly written by a human that has some pages or some paragraphs written by AI and I could see that still being a best selling book. Yeah. So I think that there's going to be like a process of sort of like baby stepping to have more AI content within books. And you know, maybe in the long term it'll get to the point where books are fully written by AI, maybe with still some human involvement.
[24:13.7]
I mean, I don't know how it's all going to play out, but we just want to like, you know, build the plumbing, build the business model and you know, just build the, the ability for this to be figured out. Because it's just, it's a, it's a really complicated thing and I think everyone wants some, some clarity for how it's going to work. Yeah, for sure, for sure. I think, yeah, I, I mean I'm sure writers are using AI already, especially when they hit like a, a stumbling block.
[24:34.9]
Right. They need just some ideation or some new thoughts. I, I, I use AI a lot just for that purpose these days. Just like brainstorming. Sometimes you don't have anyone to brainstorm with, so you just brainstorm with AI. Right. And so for sure there's a good use case there, because you're basically talking to the entire history of human knowledge.
[24:55.4]
Right. In one go. And that's super valuable. I think humans and AI are already kind of like, we're already kind of merging quite a bit. Right. I mean, it feels like a lot of we're interacting with AI so much that we're already sort of, merging and AI is having a big influence us.
[25:13.8]
We're having a big influence on AI and I think that's sort of where the world's going to now. Yeah, that's really, that's a great perception. Yeah, we're certainly all becoming cyborgs slowly. It's already happening. Yep, it's happening. Yeah, it's definitely happening. And, you know, they're trying to make it even easier.
[25:29.6]
Right? They're trying to come up with a new hardware to make like, you know, put the AI. So this AI is so even closer to you. It's just like on your body. Right. So, you know, it's glasses and pins and things like that. So, I think for some use cases, it makes a lot of sense.
[25:48.6]
And so, trippier. How old is the company? A little more than a year old. Little more than a year old. Okay, so you're still getting started? Still in the early phases. And, have, you announced any public funding anything publicly?
[26:03.6]
Have you announced anything publicly at all apart from the website? Yeah, we've raised about 11 million. We, Yeah, we raised from like Floodgate and Giant Ventures and Gary Tan and Walter Isaacson.
[26:19.1]
There's a whole list of investors on our website. And yeah, just kind of enough money to, to get. To get started off the ground. You'll probably raise, a proper Series A at some point, but we're, we're in no rush. I mean, we've, we. It's, you know, we've got a good, good cash balance for now and we've got the size of the team we need to kind of get to the next step.
[26:37.6]
So, but yeah, we've, we've announced funding. We do have a product that' where authors can sign up to, to claim and license their A.I. rights. And we've got some more stuff in the works that's coming out soon. That's good. And how is it, like, starting company nothing?
[26:52.9]
This is your second company after Scribd? That's right, yeah. Yeah. So how's, how's the journey this time for you personally? Like, you have kids now. It's a different. You're a different person. You know, you're enjoying it more or is it just as stressful as the first time? I think I'm enjoying it more.
[27:09.1]
And it's, it's less stressful. I mean, it's been really, just quick to get a company off the ground. In terms of just, I mean, just like, you know, it's, it's just different era parties. I have more experience.
[27:24.3]
It's also probably just a different era. So in terms of being a different era, I mean, you know, when we raised for our first 40k at Scribd, we had to, to negotiate a term sheet and it took months. Right.
[27:39.6]
Now when you want to raise 40k, you just, you have a quick conversation, someone says, okay, I'm in. You send them a safe, they send you a wire, and it's done within a matter of minutes. Right. So, just like the concept of safes, like, really speeds up getting things started.
[27:55.3]
And there's just all these tools that you can use to sort of onboard your team, just get everybody working together really quickly. So it's a really different era. It's a lot easier to get a company started just given the era era. And then also, yeah, I think I just have a lot more experience. And, that makes it, that makes things go faster.
[28:12.5]
I mean, I just, you know, I think we're getting a lot of things right with this company from day one. That took me 17 years to figure out, at the last company. So I think we're just, you know, able to get, you know, really good momentum really quickly, which is nice. I mean, there's certainly things that are, that were, that are challenging.
[28:30.7]
But I don't find them as stressful as I did the first time around. Because, you know, the first time around you have a challenge and you're, you're thinking like, oh, gosh, is this the end of the company? But then you, you learn that you always figure out some solution or, or something comes along that kind of unsticks, you get to the next phase. So now when we hit challenges, I'm, I kind of like, I, I, I'm sort of just pretty comfortable at navigating them and, and, and knowing how to approach these challenges.
[28:54.3]
So, I mean, yeah, the, the, the long, the, the, the short answer is just, it's, yeah, it's been great to start a new company. And yeah, it's a good, good start of the journey. I'm excited to see where it goes. Yeah. Yeah. Yes. And, and you didn't take much time off between SCRIBD and, and this company, is that right?
[29:09.6]
Like, Only a few months Just a few months. Yeah, I just felt really excited to, to do this. The, the market up. There's just such a, a, a big opportunity and there was so much conflict and I just, I knew a lot of the, the people in the book industry and the AI industry who are in lawsuits with each other.
[29:26.0]
I'm just like, well, I just want to solve this problem. So I just started talking to folks and it just kind of, it happened pretty quick. Yeah. Yeah. Is there a reason, like you were really. I mean, you sound like you knew the book industry well. Is that from work at scribd or is this something else that, you had connections there? It's from Scribd.
[29:42.7]
I mean, at Scribd, I spent like, you know, 10 years pioneering the subscription model for books. And it was very similar in that, you know, you have you know, there's a lot of resistance the model at first, then you get kind of a few early adopter publishers on and then you get consumers and you kind of get the flywheel going.
[30:00.9]
And this is kind of similar. We're pioneering the AI licensing model for books. So I think AI licensing model for books is a lot more challenging and more complicated. This is a, it's just a, there's just so much complexity in the situation, but that just makes it fun and interesting. But yeah, I just had a really good experience, pioneering the description model.
[30:18.6]
And, and then also I had a lot of experience with copyright too, with like navigating the dmca. Because we had, you know, people uploading content to SCRIBD and having to deal with takedowns, building a copyright filter, all, all those kinds of things. So yeah, I just had a lot of, of good experience, good relationships to get me ready to work on solving this problem.
[30:37.2]
Yeah. And I notice, one thing you say, often is like, we're still very early. We're a year in a long way to go. I have the same attitude. My last company I was running for 15 years. So I always have the same attitude like, oh yeah, this is very early still. Oh, you're in.
[30:52.3]
But then there's another attitude you can see from founders, like first time founders. They're like, like how quickly can I get to some crazy ARR number within six months or something? That's a very common thing. Y combinator is advertising companies that get to X million ARR within a few months.
[31:09.7]
I look at that as a little bit different. They're playing A very different game in some ways. What are your thoughts on that? Because I worry about people who try to grow too fast and then the metrics are just a mirage people, it doesn't stick around.
[31:25.5]
That happens a lot with AI companies I think. Yeah, we're definitely not being that way. I mean we actually are, are we have deals that would get our ARR up quite a bit faster. We're actively turning them away because we don't think that they are, are the, the right standards that we're setting for, for authors.
[31:44.1]
So, so we're actually, yeah, really taking our time and trying to get the model right. And I think fortunately we have, we have investors who are patient and willing to take that long term perspective. I think if you're a first time founder you kind of have to prove the numbers. But if you have, you know, if you have more patient investors you can, you could take your time.
[32:00.4]
And you have more, more leeway there. Yeah, it's interesting. I'm a partner at a Rebel fund where we you know we invest in YC companies and we, we just had this discussion at a recent session was it was like how much does early traction, correlate with long term success?
[32:18.8]
And the, the founder of Rebel Fund, Jared Heyman he did an analysis of this and he, he found that there was, there was no correlation between traction at Demo day and long term success. That's just based on like 10 years of data. So that's a data point.
[32:36.0]
So at the same time though, I mean early traction is good. I wouldn't say early traction is a bad thing, it's a good thing. But yeah, from what I can tell it's not the most important signal. It's more important to set yourself up to get long term traction and that long term hockey stick growth.
[32:53.8]
Yeah, I mean that's my view as well. I, I, that's why sometimes I look at some of these YC companies and stuff like touting their early growth. It kind of worries me a little bit. And but it's good to know there's no correlation. Doesn't mean there's negative correlation because that could also be the case.
[33:10.1]
Right. Yeah, I didn't, there was a negative correlation. I, I mean the early numbers, I, I mean I'd say early, early numbers is, is, it's better than not. But it's just, it's, it's very, could very easily be a false signal. So I Wouldn't like, you know, in evaluating a company that should not be one of the top criteria.
[33:28.3]
Yeah, yeah. And yeah, we've all been there as first time founders, like you know, desperate to get, you know, the first funding, desperate to get, you know, some validation. You know, if we could get it and now with the AI world you could maybe get some early, you know, revenue and stuff and maybe your peers are getting it so you feel pressure that you also need to get, you know, that early revenue as well.
[33:48.3]
Well but I think maybe both of our messages to founders might be. You don't need to always think of it that way. Right. You can set up a company for the long term as well. I think one way I recommend people do is when they're doing vesting agreements, even do it for longer term like 10 year vesting agreements because that's typically how the journey goes.
[34:09.1]
It's not like, doesn't need to be. Four years is actually a pretty short period of time. Yeah, we did five years. It's not a lot longer than four. But we just wanted to set the tone that this is not your normal startup want people to in the beginnings for, for the longer term. I mean I do, I do like a lot of things about 10 years because yeah, I mean 10 years is kind of like the amount of time it takes to really achieve something with a startup.
[34:30.8]
So I do like 10 years. But yeah, we, we did five. Yeah, yeah, yeah, no, I've done 10 years and it works. Well, I think people what, what the good part about it is that like it sets people up for long term thinking from the very beginning. So it's hard to like think about five years is better than four still.
[34:46.6]
And, and I would just encourage all founders to think about the company arc as long a period of time as possible. Because if you do that it also reduces the stress you get from this peer pressure and stuff like oh, the other companies that ready 3 million ARR and I'm only a 1 or whatever it is.
[35:07.2]
And I think you need to think about things in the longer term. These things can change very quickly over time and, and if you, as long as you build up the right foundation. Yeah, well I think that's, that's also from this being in the whole like YC circle. I mean there's just so many good startups all around you and everyone's got different trajectories.
[35:25.9]
So there's, there's, there's kind of competition. But yeah, I Remember when we did YC back in 06, the, the variation of like who was, who was hot just changed so quickly. It was like one month there's this one company that was going to be the best company and then two months later it was a different company and then two months later is another company and then, and then like five years later it had like no correlation to what happened to yc.
[35:48.3]
So yeah, I think that's right to say long term focused. The other thing about being longer focused though is you have to have investors who are on board for that sort of approach. I, I do see quite a bit of variation in investors. Some investors just want like, they, they really looking for the short term numbers and some investors are just looking for the right long term opportunity.
[36:06.1]
And I, I think if you want to be long term focused, it's important to get that in alignment with, with your investors or with everyone, your employees, your investors, everyone should understand like the long term goal of the company. If you get that could be a very, very good thing for a company. Totally, yeah. Very important to align with your investors on what you see as the trajectory of the company, and make sure they know what they're signing up for.
[36:26.8]
So there's no, you know, there's no misalignment, down the line. So rapid fire questions, what's the worst pitch you've ever made to investor? Do you remember that? I don't know exactly what the worst pitch was, but I'll tell you one pitch story. When we first applied to Y Combinator, we pitched Paul Graham on the idea for Uber and he told us it was a terrible idea.
[36:48.9]
And to be fair, we were pitching this back in 2006 before the iPhone even existed. So it couldn't really be done then. So. So PG was right. Yeah. Well, I don't know if you know this, but I co founded Zimride, which turned out to be Lyft in 2007. That was my first company. So not too far afterwards. Yeah, yeah, yeah, interesting.
[37:06.4]
You did 2007, so we pitched that to PG in 2006. So yeah, the, yeah, we're thinking about that at the same time. Yeah, same time. And it turned out the iPhone was the key thing. So actually Zimmeride struggled for five years before it actually pivoted into the company that became Lyft.
[37:24.9]
So it was a big pivot five years down the line. So that just shows also how long it takes to. That's crazy level. Yeah. All right, next question. What's a licensing clause that you actually want AI engineers to run?
[37:40.9]
Read. Well, when we do these, these deals, we're, we're getting rights holders to proactively license their content, which is a positive thing. But, but rights holders also want restrictions in place for what can and can't be done with the content. So, so we'd like AI developers to, to pay attention to these restrictions.
[37:59.7]
Right? And, and if they, if they're to get the, you know, the, the quality data directly from the source, it also requires just paying attention to the restrictions and making sure those on are focused followed. What's one author request that surprised you? Well, one author, was saying he had a book series where he killed off one of his characters and he thought it'd be pretty cool to bring that character back to life using AI.
[38:23.4]
So that's something that we're thinking about helping him with. But it's cool to see authors being forward looking in that way about AI. One thing you think legacy publishers completely miss. Well, I think legacy publishers, they spend a lot of time kind of like, you know, protecting their past business models versus, you know, embracing innovation and looking forward.
[38:46.5]
So I, I would, I would like to see legacy publishers, you know, really just kind of look forward and embrace innovation more. We're almost out of time here, Tripp. Anything else you want to talk about? Any other topics you want to cover? I think that was pretty, pretty thorough. Yeah, I'd just say we're you know, we're working hard to solve this, this problem of, of copyright and AI and, and, and, and not through the legal path, through through building a marketplace and, and getting everybody to, to agree to terms.
[39:13.1]
And yeah, we're making great progress and yeah, just thanks for having me on. It's been great, great to be on the show and yeah, maybe I come back down the road sometime and give an update on our progress. Absolutely. Would love that. And yeah, thanks for having you on by.
[39:29.7]
By the way, I'm a small investor in Trips company. So I'm very excited to see, to see what you guys do. Thank you everyone for listening to the Founders in Arms podcast. Next week we'll have another show. Please subscribe on Apple, Spotify, all the regular podcast channels and leave us a review and we'll see you next week.
Share this post