Subscribe
Share
Share
Embed
AI is the biggest technology shift of our lifetime. This show is about how to profit from it together. Each week I talk with the founders and CEOs closest to AI and Content, the ones figuring this out in real time. I’m also building an AI content business myself and share the lessons I learn along the way.
WHAT WE COVER
THE TITANS: How companies like OpenAI, Anthropic, Meta, and xAI are moving, and why their decisions matter.
THE INCUMBENTS: How content giants like Disney, News Corp, Universal Music Group, and Reddit are responding to AI, and what it means for creators and publishers.
THE PLAYBOOK: Real lessons on AI business models, content strategy, IP licensing, distribution, and getting paid.
ABOUT YOUR HOST: Rob Kelly has interviewed Steve Jobs and Bill Gates, helped pioneer early web content licensing, and built multiple companies with more than $100 million in total sales. His work has appeared on CNBC, CNN, TIME, and Entrepreneur.
Beyond business, every episode explores what AI means for jobs, creativity, families, and the next generation.
If you want clear thinking based on real experience in AI and media, Media and the Machine is your guide
Thanks! -Rob
I'm Rob Kelly, this is Media in the Machine, a show about the biggest technology shift of our lifetime and how to profit from it. Each week, I talk with the founders and CEOs closest to AI and content, the ones figuring this out in real time. I'm also building an AI content business myself and share lessons of what I learned along the way. You know, life's funny. I began my career lucky enough to interview leaders like Steve Jobs and Bill Gates.
Rob Kelly:I went on to be a three time founder and CEO, driving a $100,000,000 plus in revenue and some failures too. And now I'm back at the table, interviewing this new world's current and future leaders. This isn't only a business story, it's a human one. So every episode ends with me asking my guest what AI means for our jobs, our families, and the next generation. We'll figure this out together from the inside.
Rob Kelly:Welcome to Media in the Machine. My guest is Doug Leeds, cofounder of RSL. He's tackling the question every media CEO and content owner is asking right now, how do you get paid by AI companies, including even if you're small? He's already working with Reddit, People Inc, USA Today, and others, a collective representing nearly half the content AI models train on. That gives RSL real leverage across the table from the AI giants.
Rob Kelly:RSL stands for really simple licensing. Doug's co founder, Ecker Walther, co created RSS over twenty five years ago. That's a standard that powers this very podcast feed. It feels like they were built for this moment. Doug explains why the ship has not sailed on monetizing your content with OpenAI, Anthropic, at Google and others, and how even small publishers can land real licensing deals.
Rob Kelly:We go deep on Reddit where Doug coached CEO Steve Puffman and how Reddit is making real money licensing content to AI while also still growing their traffic. He breaks down how RSL differs from pro rata and toll bit, what the first real AI licensing deals will look like, and gives us hot takes on all the AI frontier models, including one, which frustrates him a lot. He also talks about competing with Google. When Doug was CEO of ask.com, he and his boss Barry Dillard turned Google from a threat into a profit driving partner. He teaches at UC Berkeley, is a long time CEO, has spent decades navigating the space between content and search engines.
Rob Kelly:And we close with how AI touches his mom, his dad, and his daughters. You get to see the human behind the CEO. Please enjoy my conversation with Doug Leeds.
Rob Kelly:Can you share what RSL stands for?
Doug Leeds:Oh, yeah. It's from the heritage of RSS. It's really simple licensing.
Rob Kelly:So my understanding is there's really two main components. One is it allows a creator of content to set their rules.
Doug Leeds:It sets their license terms.
Rob Kelly:To set the licensing terms for their content on the Internet specifically.
Doug Leeds:Correct.
Rob Kelly:And then part two is this more sort of those familiar with the music industry kind of ASCAP like rights organization where you allow content creators who commit to the RSL standard to collectively negotiate with companies that wanna use more than just a one off content site and want to perhaps license a bunch of content at once. Did I get that right?
Doug Leeds:That's exactly right. Yeah. And your analogy to ASCAP is is spot on. The problem that we are trying to address with the RSL collective, which is very much modeled after ASCAP, that the number of creators of relevant content to AI is so high that it's literally impossible to do one off deals with every content rights owner that you possibly could use or would want to use for your AI product. It's nearly impossible.
Doug Leeds:They were talking about hundreds of thousands of potential deals. This is exactly what happened a hundred years ago or more now with the onset of radio. Radio want to play music and how are they gonna play music? They would have to do a deal with every musician, everyone who's published music to get their rights to publish music and Collective Rights Organization and ASCAP is the one that we know most of in The US said, oh, we will collect all those rights and we'll create a blanket license so that anyone who wants to consume all this content and use it can do one license agreement and cover hundreds of thousands of artists and get it all at one time so it becomes a much much more efficient way of licensing. And we saw a need for that in digital content.
Doug Leeds:Same idea, same model, but it didn't exist. And one of Eckert's real breakthroughs in thinking was that it couldn't exist without some infrastructure layer. And that's why we started with the RSL standard. It needed some infrastructure layer to be able to say, okay, I'm able to tell you what my license terms are in a machine readable way. That's the first step.
Doug Leeds:And then you can have one let's call that the language. If you create a language, then you can create one statement using that language that covers a bunch of people.
Rob Kelly:Can you talk about the origin story of RSL and how much curiosity played in that?
Doug Leeds:Yeah. It absolutely did in two important ways. The first one is my partner and I getting back together after twenty years later since we worked together. This is Eckhart? This is Eckhart.
Doug Leeds:Yeah. He and I worked together at Yahoo back in the early two thousands. He was running web search and I was in my last ever, didn't know at the time, my last ever legal job as an associate general counsel at Yahoo. He was creating something called Yahoo Answers, which was gonna be like the Wikipedia that Yahoo built and we worked on that a little bit together. But the real way it came about was because I teach a class at UC Berkeley and my TA, a wonderful undergraduate who has her own company, she asked if she could invite someone to come speak and I said, yeah, course, if it's someone who's interesting and relevant.
Doug Leeds:And she said Eckert and I was like, that's amazing because I know Eckert but I haven't talked to him in fifteen years. And then we started talking about what we're seeing in the world. And particularly, RSL was really Eckert's idea at the beginning. And he had co created RSS, the standard that sort of underlies like content syndication on the web. And he had been thinking about how some of those things that he had solved for with RSS could be applied to licensing of content.
Doug Leeds:And we just started talking and spitballing ideas, and it kept going, and it ultimately resulted in RSL.
Rob Kelly:I've got a couple of Internet content properties. So Yep. I wanted to give you an example and then ask you how it would benefit me as, you know, just a content creator, content owner. So I've got a newsletter where I recommend documentaries called Daily Doc, and it's got maybe six, seven hundred articles created over the last couple years. How would that benefit me, the creator, in those cases?
Rob Kelly:What would I even do?
Doug Leeds:So right now, I would argue unless you tell me that I'm wrong and I would love to hear that. You have no particular way of monetizing that content when it comes to its use in AI. Correct. It's not like you're getting a phone call from OpenAI or from Anthropic saying, hey, we wanna license your newsletter for use in our LLM. It's just not happening.
Doug Leeds:That's not to say it's not being used, they're just not licensing to use.
Rob Kelly:Yep. It's being scraped and I get some level of traffic and I'm allowing it to be scraped because I'm leaning into the learning of it. Right. I'm okay with that. But thinking down the road, let's say I wanna turn into a business, obviously, that changes.
Rob Kelly:So, yeah, walk me through the step by step process.
Doug Leeds:So it's very simple and it's getting simpler. So right now, all you would do is you would sign up for the collective. Now we haven't concluded our first licensing deal yet, but let's talk about when that happens. It would be very much like registering a song with ASCAP. You would register your content with RSL and then whenever your content is used by an AI company, then you would get paid.
Doug Leeds:The way that would happen was just like ASCAP. If you write a song, you can register with ASCAP and then that song can get distributed. So for example, on Spotify. When some user of Spotify plays your song, Spotify knows that. It tells ASCAP that your song was played.
Doug Leeds:It gives ASCAP a pool of money which ASCAP then has its own rules for distributing, but it says, oh, if the musician is what's called Machine Gun Kelly, just because I'm looking at your name, then he says, okay, Machine Gun Kelly gets paid according to our rules from this pot of money every time their song is played. If their song isn't played, they don't get paid. That keeps the alignment of incentives really nice. So Spotify is only paying for music that its customers uses, but it's paying for customers music that its customers uses, so that music can continue to get made. In our case RSL, same exact thing.
Doug Leeds:You would register with us, you would say, here is my content, here's my newsletters, I would like to join the collective, so when some AI company is using my content, I get paid. And then we would pay you. We would give you the money on a regular basis, call it quarterly.
Rob Kelly:So it sounds like a two step process for me at least. First, I have to adopt the standard, put in some I assume this is some code I put into Daily Doc Yeah. Website on WordPress. And do I set any particular rules or is it sort of all or nothing at this point?
Doug Leeds:So since it's on WordPress, WordPress has created and there's a GitHub link that allows you to do this through WordPress. So it's very simple. And other platforms whether that be Substack, Medium is one of our partners. If you are on these partners, they can give you a one button access to be able to implement this very simply. So you don't have to do any coding or anything like that.
Doug Leeds:Although it's very simple, but even simple coding for me is not simple. So it could be as simple as one button through your WordPress dashboard to say, I wanna use RSL. That's the first step.
Rob Kelly:But in that first step, let's just say whether it's WordPress or me doing it, is it sort of all in or all out or No. Are there particular rules that I'm setting on how I want my content to be used or optimized or is it just Yeah. Yes or no?
Doug Leeds:It is up to you. So RSL is a standard which is open and freely available and you can use it starting right now, allows you to articulate any of the rules that you want. Yes, you can use me for Rag for example, but you can't use me for foundational training. Any rules, they're up to you and you can determine them and describe them using RSL and those are your terms.
Rob Kelly:Is there a difference other than training versus RAC or those the two main
Doug Leeds:Oh yeah. There's there's it's an extensive protocol that allows you to do a bunch of things. For example, we just announced today that Creative Commons is supporting us. So that would be, for example, a attribution and maybe a contribution license. In other words saying, you don't have to pay me but you have to give me attribution and say where the content comes from, and or maybe you want to contribute to a nonprofit or a fund, and that can be part of your RSL.
Doug Leeds:Basically, think of RSL as a machine readable terms of service, and it can be any terms of service that you want. But it's now in a machine readable format, so crawlers, bots, AI companies can determine and can see what it is before they access your content and determine if they wanna continue to access your content based on those rules that you set.
Rob Kelly:K. So it seems like in the simplest terms, I could say, I'd like to participate in monetization of it and sort of be done with it, you know. That's it. I could do something more detailed like say, you know what? I want attribution and links.
Rob Kelly:Is that Yeah. I heard you right? Which is sort of like the contract we set up with Google years ago many we meaning, many websites that Google gave attribution and links and thus traffic.
Doug Leeds:And you could also say for example, hey Google, it's okay to put me in search results but it's not okay to put me in AI abstracts where I don't get the traffic. You can absolutely use RSL to do that. Just this week the European Commission said that they're investigating Google for antitrust or anti competitive practices because they don't allow you to opt into search but opt out of AI abstracts. And the good news for the world, for Google, for publishers is RSL now is a method for you to be able to do that.
Rob Kelly:So if part one is me getting the RSL code in, whether it's me personally or through some sort of platform partner like WordPress and Medium and so forth, and the second part being joining the collective, that's optional?
Doug Leeds:Join the
Rob Kelly:collective or you don't join the collective?
Doug Leeds:It's optional on a licensee basis. So for example, you can say, I would like to license my content as part of the collective to OpenAI, but I don't wanna do that to Anthropic. You can do that. Everything is non exclusive
Rob Kelly:and opt in. So it sounds like to me the RSL collective, your first phase has really been to set up both obviously the standard and then you're now forming this collective. Right? You can't do the collective without standards. That makes sense.
Rob Kelly:That's the sequence. Okay.
Doug Leeds:Yeah. And then the collective would be we would negotiate as you suggested a deal with AI companies and then you could join that. You could we would say, which we haven't said yet, hey, we have a deal with OpenAI for example. And anyone who wants to join is free to join. And then when it when OpenAI uses your content, you'll get paid.
Doug Leeds:And it's as simple as saying, I agree to participate in this. Very much like I register my song, hey, ASCAP, go ahead and license it. When you do, give me my money.
Rob Kelly:I know you can't tell me who the First AI deal will be with, but can you tell me what it looks like in the normal world of doing this? There might be a one time payment for training, and then there's a payment for usage. What are gonna be the components of those first AI deals?
Doug Leeds:This is why we have a publisher steering committee where they give us advice on the structure, but we have all aligned pretty strongly along the idea that what we're going to be using is a paper inference model. So when your content is used, you get paid. Again, until we do the first deal this is all subject to change, but this is certainly our current thinking and I think it's well thought out at this point, which is when your content is incorporated into training, a lot of people including publishers we've talked to have said, okay, the the ship has sailed on monetizing that content because it's already been used to train. And we have said and believe and will likely structure these deals on the idea that that's not true. That the content, while maybe part of training, is still monetizable when it gets used.
Doug Leeds:And that's when it has value to the AI companies, and that's when that value should be shared with rights holders. So again, this is very similar to the model of Spotify. You don't get paid by Spotify when your song gets uploaded. And when they say, oh, we have a 100,000,000 songs on Spotify. They're not paying for every one of those songs, they're paying for them when they get interactive, when they get played specifically by a user.
Doug Leeds:That's the model that we're talking about. That's the fundamental idea, which is whether it's rag, whether it's grounding, whether it's from a foundational model, when your content is used, you get paid for. And that aligns the incentives between the AI companies and the
Rob Kelly:rights holders. What's the timing of starting to approach the frontier models for actual deals for the collective?
Doug Leeds:I'm confident that those deals will be happening next year. We have started preliminary and productive conversations, but I can't really say more than that.
Rob Kelly:2026, we're talking about. And is it as simple as you'll go in order of the largest ones first? Start with OpenAI and then just hit up all the big guys?
Doug Leeds:Can't really comment on that one, if that's okay.
Rob Kelly:What have you learned so far from the media companies making the RSL commitments? And what surprised you the most?
Doug Leeds:I think what surprised and encouraged me and excited me is that some of our partners, many of our partners have already done deals with the AI companies. So Reddit is a partner and People Inc. Is a partner. We have lots of companies who have already working with us who are on our publisher steering committee which is the group that helps gives us advice about how licenses should work and what the terms would be, who are already doing deals. Reddit for example has a deal with OpenAI.
Doug Leeds:Reddit has a deal with Google. So why are they participating in us? One, and I think this is the most important, is they understand that whether or they have a deal, the open web and ecosystem of the open web is in danger. And without a structure like this, the infrastructure of the web is in trouble. And that's important to them to be able to ensure viability of, even if they're getting their money already.
Doug Leeds:So that's one of the nicest things that I've seen and, you know, hats off to our partners Reddit, PeopleLink, others who have said, we have deals, but this is important and we want this to work both for ourselves, but also really for the broader web and for the ecosystem of the entire Internet.
Rob Kelly:Who is the first content company to join?
Doug Leeds:Well, when we launched in September, we launched with Reddit and Yahoo and People Inc and Ziff Davis and Internet Brands, which is WebMD and other sites, Cars Direct, and Medium, MIT Press, O'Reilly, some of the groups, some smaller ones, which are really important to us. Company called Ranker, an amazing company called Raptive, which actually represents creators. I don't know if you know Raptive, Rob, but you probably should Yep. Because they're great. So those were all our initial launch partners when we left the gates three months ago today.
Rob Kelly:What was the most important way to get it to critical mass?
Doug Leeds:I think a few things. One is I think the background of myself and Eckert having both been in search, run search engines. Eckert ran web search at Yahoo. I obviously ran ask.com. And then in content and media, the company that is PeopleLINK now is something that I started.
Doug Leeds:So understanding the ecosystem I think was really important. We have a long background, we understand what has worked, what is failing now, that's an important part of creating the design of this. And then I think the next piece of it and maybe and all fairness probably even more important than that, is that we decide to do this as a nonprofit. Because I don't know of any other nonprofit doing what we're doing and I don't know if anyone doing trying to do or talking about monetization that's doing it as a nonprofit. And we really are doing this because we believe honestly Rob, you and in journalists and in people who are writing content and creating it and publishing it, and they need a way to be participating in in AI, what AI participates in taking their content and using it.
Doug Leeds:And that ecosystem for us is the ultimate reward as opposed to, oh, we have a bunch of outside venture capital investors that are gonna make billions of dollars off of this. Our main thrust is to make the ecosystem sustainable. And so it was very clear to us after our some initial calls that the way to do this was as a nonprofit, some of our earliest conversations we had. It was like, look, I've got people calling me up down sideways, my traffic is declining and you see that as a business opportunity and that kind of upsets me. And as soon as we sort of said, what if we did this as a nonprofit, It was like, okay, tell me more.
Doug Leeds:I'm interested.
Rob Kelly:Are the closest comparables on the for profit side pro rata and toll bit?
Doug Leeds:Pro rata and toll bit are are the ones we hear the most of. They're not doing collective rights organizations and they don't have a standard, but they are talking about monetization. They have each of them have a little bit of a different take on it. And, you know, what I would say and I have said to anyone who's interested in what we're doing is do it all with Tolbit or with Pro rata, with any of the companies that are out there with Microsoft's marketplace that they're launching. There's nothing mutually exclusive.
Doug Leeds:We're in a very early stages of a big disruption, and so the best thing we think of for publishers is to be able to give them the freedom to test and try stuff out. And that's again, part of our nonprofit mission is like that. That's why I think it goes very hand in hand with our values and why we're here.
Rob Kelly:How important was it that Reddit committed in that early set of content companies?
Doug Leeds:Super important.
Rob Kelly:Right? Some say it's what percentage of AI is trained on on Reddit?
Doug Leeds:I saw 40%. Yeah. But I I know that that I'm sure is a point in time and can change but super super valuable, super super important. And I've known Steve Hoffman for a long time and he's the CEO, he's a fantastic guy. The company is incredibly valuable.
Doug Leeds:I use it every day myself. And then them being able to say, this has real value, this has real merit. To them and to the open web was an incredible vote of confidence. I would say them, they first maybe, but it's pretty close across the board. Yahoo, right up there.
Doug Leeds:People Inc, right up there. Internet brands, right up there. Ziff Davis, Medium, all of them. But yeah, Reddit because of the percentage of content that they have and of and the deals that they have. I don't know too many people that have deals with Google and they have one of them.
Doug Leeds:And yet they're still working with us. So that's a big indication to the industry that says, yeah, you should be doing all these things if you can.
Rob Kelly:You mentioned Yahoo. I'm still a big fan and user of Yahoo sports content for fantasy basketball. Just has a certain way to be the best source.
Doug Leeds:I know that Yahoo is a big syndicator of content. And so a lot of content from different places on the web will be distributed through Yahoo. And so as we were putting together RSL, one of the things that we had to solve, and Yahoo very engaged and helped us solve, was how do we have different license rights for different content on a single website. Because if your content is being syndicated on Yahoo, it may have and should have different rights to the content that Yahoo produces themselves. And that was something we originally thought of this as an augmentation to robots.
Doug Leeds:Txt, robots. Txt, where the original sort of like, yes, you can come to my site, no, you can't come to my site signpost is posted. But that is for domain wide and Yahoo very early on and then Medium and then Conde Nast and others have pointed out to us that the set of rights on a domain is very different across the domain, and the type of content that they have is not uniformly licensable. So we had to create RSL not just in robots. Txt but that could go into the HTTP headers so that you could actually control the license terms of each page of your website.
Doug Leeds:And then that wasn't enough. We had to go into basically be able to put the standard into every atomic piece of content so that any image you got would have its own set of RSL rights, which I think was one of the reasons we just announced today that the Associated Press has joined us. Their content is being distributed by their partners all over the web and they want to make sure what you just described where an AP for example, story or a photo lives on one site that gets scraped and the site doesn't have the ability to say you can't touch this photo because it's not mine to give you rights to. RSL allows you to do that.
Rob Kelly:So is RSL now multimodal?
Doug Leeds:Oh, yeah.
Rob Kelly:Can you talk about text? Obviously, does text Yes. Audio? Yes. Images?
Rob Kelly:Yes. Video?
Doug Leeds:Yes. And more. You can put it into DRM. You can put it into like books, for example that you host digitally.
Rob Kelly:So it's media agnostic. It's agnostic to the form of media.
Doug Leeds:It is extremely flexible. I can't say that there won't be a form of media that exists either today or tomorrow that gets invented that we haven't thought of. But our job at RSL is to make sure that the standard can be embedded in any type of digital content.
Rob Kelly:What's the biggest surprise so far in saying this all up?
Doug Leeds:I think in some ways it's been the massive adoption and in support that we've gotten so fast. Something that came to mind, like I said, Steve Huffman, I've read it as a as a friend. Steve is another person that comes in speaks to my UC Berkeley class every year. And he has said when he was starting Reddit that one of the things that that he and his co founder would say is don't screw this up. Like the momentum is unbelievable, the traction is incredible, just don't get in the way of it.
Doug Leeds:That statement has echoed in my head a bit over the last few months of like the interest, the need, the problems that we're solving and doing so in as a nonprofit in a in a unique way. I'm just astounded by the reception that we've received both in The US and globally. We announced this week a deal with Note Inc, which is kind of like the Substack meets Medium meets YouTube of Japan, and they have joined and we're leading our efforts in Japan now. I've briefed the government of Slovenia already and have other governments in the EU particularly that are interested in hearing more about this. I don't know a single person who lives in Slovenia, but Slovenia reached out to me and said, you brief us on this?
Doug Leeds:So that kind of reception, that kind of awareness about what we're doing has been the most surprising thing, I think.
Rob Kelly:So you've teach last I checked two courses at Berkeley. Right? Correct. Both on leadership. What are the names?
Doug Leeds:Creativity and leadership, I teach in the in the fall, doing that just finished this semester. And leadership by persuasion is my spring speaker series class, is where all these guys come in, non gender term guys come in and come speak.
Rob Kelly:I've heard all sorts of fun things about those. I thought it'd be interesting just to get first off, can you define leadership? And then I've got a couple questions, your hot takes on some leaders out there. How do you define leadership?
Doug Leeds:So I define it, and I'll I'll I'll say that I define it this way because I think I have a way of thinking about it that works this way. But I define it as the ability to have folks follow you or subscribe to what you believe in or what you want to get done as if it's their idea and if it's in their interest. So leadership is bringing people on to what you want to accomplish and making it what they want to accomplish.
Rob Kelly:Love it. And I thought I want to get your take on some leaders out there. You know, just a few words about them on how they're doing in this new era of AI. How's Sam Altman doing at OpenAI? What's your take on him as a leader?
Doug Leeds:I I think the disruption is incredible. And they're bringing it to market, which I really think OpenAI deserves a lot of credit for, because really their launch in in November 2022, I think really was a defining moment. And I think we'll look back at this in to accept the AI is influencing everything we're doing these days. So everything I'm doing, but I mean more and more in society. That's a moment in time that I think was really easy to identify a pivot in the world.
Doug Leeds:And you've got to give credit to Sam Altman for that and to OpenAI. They brought to market, they showed it, that product is accessible to everybody. You know, there are people out there who are good people and do good things, and there are people who are not good people and do good things. I don't know Sam. I don't know which one he is, but I'm quite impressed with what he and his team have produced and it's hard to believe you can do all that without being a good leader.
Rob Kelly:How about give me your take on Anthropic?
Doug Leeds:I'm frustrated with Anthropic, Because while I'm sympathetic to the argument that there's no way to license all the content in the world. I don't think that excuses necessarily stealing the content. But at least OpenAI has said we can't license all the content but we can license some of it and we're gonna go try and do that and so they have deals with Reddit and PeopleLink and others. Anthropic to my knowledge doesn't have a single deal. And they talk about being ethical and they talk about doing this for a greater purpose.
Doug Leeds:I find it very hard to reconcile those statements with the grift that's taking place in their product.
Rob Kelly:How about Elon in x AI?
Doug Leeds:The fact that he's repeatedly built companies from zero to one that are so impressive and add so much value, I'm impressed. I'm very impressed. I don't think you can do that without being a good leader at least on some metrics.
Rob Kelly:Have you tried to get x.com, formerly Twitter, to adopt RSL? I have not. They'll be on the list certainly. Right?
Doug Leeds:Yeah. Think so. I mean, you know, it's one of those interesting things, like, we haven't reached out to the companies that are generally being run by companies that are trying to build LLMs themselves. We really think that that's probably more likely to be more in the licensing conversation than the license or conversation. So for example, YouTube would be a perfect participant in us but for the fact that they're owned by Google and Google's probably not probably, Google is certainly on the LLM side of the ledger more than on the content producing side of the ledger.
Doug Leeds:So I put x in in that bucket. But it's interesting because we're having good conversations with others. I don't I can't really name them, but others who are on both sides of that. So as you say the question, it's making me rethink that maybe I should reach out. Maybe I should.
Doug Leeds:But I haven't.
Rob Kelly:What's your take on Meta and what they're doing AI wise?
Doug Leeds:It's certainly interesting. They just announced a bunch of of deals for RAG including with a couple of our partners, USA Today and People Inc. So I'm glad that they're starting to license content. Look, I welcome everybody to the pool of licensing. I think that's so important to preserve what we do, what we value in this world.
Doug Leeds:Like, I've had a conversation with people at the White House who've described the sustainability of independent from government, like voices, as a national security priority. And I agree with that across the board, as well as a human priority. And to the extent that we are seeing companies who like Facebook have been at one point notoriously absent from the licensing space start to license, I'm excited and encouraged.
Rob Kelly:How about Google? And specifically asking, what's your take on their leadership in the AI space so far?
Doug Leeds:They're the leaders. Even Sam Altman acknowledged that recently, that they're the leaders. And, you know there's a question in the EU is raising it about whether they're leaders because they're extending their dominant position in one market to a new market or if they're leaders because of something that they're acting and adding value to it organically. I don't need to take a point on that specifically. I think that what we need is companies that are creating AI and we need competition in the marketplace and they're providing it and so great.
Doug Leeds:Good for them.
Rob Kelly:Yeah. I could tell you, they're neck and neck Gemini versus ChatGPT, with my 10 year old son right now for
Doug Leeds:Oh, is that right? Which one did he start with?
Rob Kelly:He he and he doesn't talk to the EU. Yeah. Of course, you start with ChatGPT and just lately Yeah. Gemini is just out doing it in terms of creating stories specifically and images.
Doug Leeds:I think the images for me, I've seen it too. The images are much better on Gemini. And the speed. When I use it. You know.
Doug Leeds:The speed. Yeah. Exactly.
Rob Kelly:I'm curious on the frontier model part. What's your there are two out liers of major tech companies who are still not considered frontier models, Apple and Amazon. Can you just give me your take on each of those and where you think things are going AI wise for them?
Doug Leeds:I know Amazon and Apple are both working on it because I've read that. It makes a ton of sense. Both of those companies have been counted out in different spaces and then come roaring back. I expect that they'll both be producers of foundational models and not licensees of it. You know, we'd obviously heard a lot about OpenAI licensing to actually both Amazon and Apple, was thinking specifically of Apple for Apple intelligence.
Doug Leeds:But this is the next space and I don't think like Google has showed that if you have some natural advantages and Apple certainly does and Amazon certainly does, both of them do, why would you do anything but use those advantages to try and get ahead in the next big thing, which is AI? I wouldn't wanna be blockbuster video of anything. So, you know, blockbuster video could have been Netflix and they didn't do it. And I I would be thinking about that all the time if I were a leading company that isn't doing AI right now.
Rob Kelly:You've worked directly with a few interesting content founders, CEOs, and I just wanted to get your take on how you think they're doing in this new world of AI. So the first one you mentioned, Reddit, Steve Huffman.
Doug Leeds:I'm so impressed by Steve and how I've seen Reddit. I mean, I was Steve's executive coach for a couple years back in the late twenty teens. We got to know each other and to see his evolution as a CEO, as a leader, as someone who is really positioning his company for the future, I'm so impressed. And the people on his team are incredible, which is a tribute to him and the value that Reddit provides to AI and also directly. I mean, you know, this is one of the things that Steve has said, I think is really important.
Doug Leeds:We talk about Reddit in the AI sphere about being so important to training, but their traffic is not decreasing at all. And so it's a real tribute to Steve, to the company he's built, that what they have is valuable to AI and also valuable directly to humans to go there themselves and experience it. And I'm incredibly impressed by him and a fan.
Rob Kelly:You bring up a good point. He's pulling off the best of both worlds. One of highest amounts of revenue. I track these things in spreadsheets, I know Yeah. Arguably top one or two in terms of AI licensing revenue going into Reddit and still getting the traffic increase.
Rob Kelly:How's he pulling off the traffic increase part?
Doug Leeds:Well, it's incredibly valuable. Look, you can use ChatGPT for example, I think is a better version of what we are building what I've been working my career on in search. Like, I have a question and instead of pointing me to a site or a bunch of sites that might have the answer, I'm actually getting the answer. That's freaking great. Full stop, that's a better experience.
Doug Leeds:It was something we tried to do at Ask and never could do fully at scale like AI has done it. But that is a responsive mechanism as search was. I have a it's something I'm interested in and I have a question about it and you're gonna give me an answer. Or, you know, as as we've used it, it can iterate and we can work together on, you know, as a partner of my ChatGPT is a partner of mine and I can work with it and and we can get to where we're going. But the serendipity of Reddit is not something that is found in AI.
Doug Leeds:The ability to see what other people are thinking about in a topic that you're interested in, and be able to browse it, and to be casually interested in it, I'm super super impressed and would be very upset if I didn't have access to Reddit for that serendipity, for that learning, that information that I didn't even know I wanted. TikTok is somewhat the same way, but TikTok for me is more about entertainment and Reddit is more about information. But both have that value of I don't necessarily know why I'm going there, except that I know when I'm there, I'm being fulfilled.
Rob Kelly:Is it as simple as Reddit is giving human answers, you know, in this new world of AI where AI is giving answers based on a corpus and a conglomerate of content out there, that it's a community at Reddit that is giving real life human answers? Is it that's what makes the content even more valuable now?
Doug Leeds:So you said the community and the answers. And I think it's the community absolutely, but the questions are as valuable as the answers. When I'm on Reddit, I'm seeing questions that I didn't even think to ask, that are super important and valuable to me. So like, okay, I'll disclose one of my hobbies. I love hacking Southwest Airlines.
Doug Leeds:Like not hacking like in a technical sense, but trying to figure out how to get the cheapest flights from point A to point B. It's so much of a passion that like I literally will take a $100 Uber to a nearby airport to save a $20 airfare because I'm like, yes, I got $20. It's like penny, any poker. Like I don't care that we're only playing for nickels, I gotta beat you. Like that's my passion on Southwest Airlines.
Doug Leeds:And so I live on the Southwest Airlines subreddit for all the tips and tricks partially just to know what percentage of what I know does the world know. And so far it's in that order, not what percentage of the world do I know. I'm pretty good on this one. But every once in a while I see something that I didn't even know. I saw a trick that I didn't even know about on Reddit about Southwest Airlines, how to do something.
Doug Leeds:And I couldn't stop thinking about it for three days. I was so happy.
Rob Kelly:What was the thing? So I don't
Doug Leeds:I really don't wanna say because
Rob Kelly:Just give an example. Give a flavor.
Doug Leeds:It was a way of saving money on flight credits and being able to use them that I didn't know existed.
Rob Kelly:Okay. So saving money on flight credits, and the reason I asked is I wanted to specifically then ask, in that case, Reddit can surface that a real human came up with the question, I assume. Right? Yeah. Okay.
Rob Kelly:Real human came up with some question about how to maximize flight credits on Southwest. And that is actually a thing that AI and the LLMs cannot do. They cannot surface a new question. Unless you asked it, they're not going to surface it. So literally, that piece of content is something that you, Doug Leeds, can go to Reddit and get, but you would not go to ChatGPT or I would have thought to ask.
Rob Kelly:I would never
Doug Leeds:have thought to ask this question. But on Reddit, someone else thought to ask. I got to be in that community and see that exchange between people who asked the question, which was a good question, and people who answered that question, which were great answers, and then create value for me that I didn't even know I was missing out on.
Rob Kelly:So what's the lesson there then for other content companies? Is it content plus community is the magic? Well, don't think
Doug Leeds:there's any one rule for this. But yeah, I mean, I think that's content plus community is something that that Reddit nails.
Rob Kelly:Mhmm.
Doug Leeds:I think by the way TikTok does too in a different format. For example, on TikTok there's this one woman who is a beekeeper and she just shows about how she rescues bees from different places. Why is that so interesting to me? I'd never would have known that in a million years that I was interested in watching someone save colonies of bees, but I love it. It's so much fun.
Doug Leeds:It's that serendipity that you don't get from, at least at this point, from AI. You go there with a problem or a need and it does a great job of solving it, but it doesn't yet know you well enough at least right now to say, here are some things that you don't know about that you'd be interested in. And I don't trust it to do that where I do trust TikTok and Reddit. If I'm open AI by the way,
Rob Kelly:what I'm thinking is how do I create a new social media community based product with AI to take care of what you do. Yeah. Yeah. But let's we're
Doug Leeds:going on.
Rob Kelly:Uh-huh. Yeah. Yeah. Bring it on. Okay.
Rob Kelly:So another hot take. How about how's Neil Vogel doing? You hired him back in the day at IAC and he's running People Inc. Now, one of the biggest sort of digital media companies in the world. How's he doing He's age of AI?
Doug Leeds:I think he's a leader in every sense. I'm very proud that that I put him in that role, but it's all been Neil. And he was, you know, he was dealt a surprise, like we all were. Like, he had just bought Meredith with people and travel and leisure and a bunch and paid a lot of money for it. And AI comes around and, you know, it's a punch in the gut when you have a model that you've just spent billions of dollars on and also that model gets disrupted right away, immediately.
Doug Leeds:Can't think of a better response and a better pivot than Neil and his team have made. John Roberts is our chief innovation officer, Alex Ellerson is our COO. I mean across the board, that group, they've made a big bet. The rules changed right after they made a big bet, and they're converting that big bet into a winner.
Rob Kelly:What's the main way you think they're turning it into a winner? They're incredibly ahead of the
Doug Leeds:curve on thinking about how to do licensing. They're talking with everybody and really processing and and smart. I think Neil's amazing. And I want to call out John Roberts and I could talk about how astounding he is all day long. So from the work that they are doing with Microsoft on their marketplaces to the deals that they just signed, the first one to sign the deal with Meta, they're a leader in every sense of the world in the content space.
Rob Kelly:I'd be remiss if I didn't ask about Barry Diller. I was just reading his new biography and can you give me maybe a favorite story or memory about him that maybe is even different than his public persona? A good Barry story that we haven't heard.
Doug Leeds:Look, Barry is one of the smartest business people I've ever had the pleasure to be around and he's also super curious and really interesting, Lee, that when we would have our quarterly business reviews, we come and present what our businesses were doing. And the people whether they were the board or they were just sitting in like, you know, Michael Eisner or Jack Welch, sitting in those conversations, asking questions. And that was because of Barry's leadership. Like, who was at that table was because of Barry. And here's another thing that I didn't I didn't read the biography, so I apologize to BD as we call him.
Doug Leeds:But one of the things I'm most was most appreciative and impressed about I see for my over decade there, was there was no politics whatsoever at that company. It was purely a meritocracy from top to bottom. If you had the best ideas, your ideas would get funded, would see the light. This is why Tinder was hatched there and many other things. It's because it was a pure meritocracy.
Doug Leeds:All they wanted was the best ideas. And if there was anything I would say about IAC was because it was such a meritocracy, sometimes this wasn't a rule, but sometimes maybe the people who weren't the nicest people would win internally. And I don't put Barry in that category, but as a meritocracy you're interested in the idea and not the person who has the idea. And so if the idea was good, it didn't matter necessarily if the person who came up with the idea was good or not. Because I think to a fault they were more concerned with how people felt and whether an idea would make someone feel good or bad, than whether the idea was itself was good or bad.
Doug Leeds:So you didn't get to explore really disruptive or interesting ideas. You didn't get to push the envelope and see where it took you because they were very concerned about, oh how would people feel about this and would they they like it and maybe you shouldn't even bring it up if they don't like it. So to Barry's credit, that organization is driven by what's the best thing for the organization, what's the best ideas. And when you're in that space, it's so liberating because you have the freedom to really explore what you think is gonna be the best thing. And if your idea is the best and you can defend it and you can my definition of leadership, you can make other people agree to it as if it's their idea, then you have the resources and the runway to go pursue it.
Doug Leeds:And I took it for granted a bit when I was at IAC.
Rob Kelly:And it reminds me when you're at IAC and you're CEO of of ask.com and I feel like both you and Barry said this at the same time, around the same time, that you can't compete with Google. This is back in the search days. And then I know you got real creative and figured out a profitable way to run Ask, which actually meant partnering licensing Google's search technology, buying ads from Google. At one point, you were buying one in 12 ads and then selling the traffic or the ads on your own site, and you've made lemonade out of lemons. But what I wanna ask you about specifically was now that you're launching RSL, you know, Google and YouTube in particular has a content compensation system, at least one.
Rob Kelly:They've got this thing, I think it's called content ID in YouTube.
Doug Leeds:Yes. That's right.
Rob Kelly:Why wouldn't Google just launch a new protocol or a new platform? Are they gonna be in this space? Are they gonna be a competitor to RSL?
Doug Leeds:Come well, I mean, like I said, the water's warm. Jump in. I think a couple years ago, certainly before OpenAI launched, if Google had done that, it could have been a real advantage for them. Looking back in hindsight's 2020, and I didn't even I wasn't even thinking about RSL back then, so I'm not one to talk. But Google could have done something like this.
Doug Leeds:But I don't know that they can now because it's not just about Google anymore. It's about all LLMs and all AI uses which is way broader than than Google. But I do think an ASCAP is is a model that we've talked about already, that doing it as a nonprofit provides something that even Google can't do as a captured company. And it's about doing so in a way that preserves the open web, which Google should be the one defending. I think maybe look, I'm I'm speculating here and it's probably not the right place to speculate, but I do think that they were a bit caught off guard by how fast competition in the space.
Doug Leeds:Like, you've mentioned me and at Ask and we couldn't beat Google until we had to pivot our model to fit within Google's model. I don't think Google was in the habit of pivoting their model to fit other people's businesses. They were the king on the mountaintop and what they said everyone else had to deal with and all of a sudden they got disrupted. And they're responding very well to it, but that's a new behavior that they had to learn. And I think that's why in some ways they they've been behind, in some ways they've missed opportunities.
Rob Kelly:What have you learned from your students about AI, your students over at UC Berkeley?
Doug Leeds:Oh, I can't even start. So much that we have to teach differently.
Rob Kelly:How so?
Doug Leeds:Because there was by the way, there was an amazing article on the LA Times from a professor at USC who talked about this, and I'm was hugely excited by the article, which is basically saying, hey, the way we've taught college which is to impart information no longer makes sense because information is easily had. And some of the things that people who teach college students are struggling with is fitting this new world into the old world which is what I give you information and you tell me what the information I give you and I test you and grade on how well you told me what I told you. And that's just stupid in my opinion. And at the same time what we're seeing is that students are in desperate need for community and connection and they're not necessarily finding that. They're spinning on I say I've talked positively about TikTok but there's a downside of that sort of spiraling of scrolling and stuff.
Doug Leeds:And then they're spending less and less time with each other all the time. So I believe that what AI is doing is it's making college more about how do we get students to interact with each other and less about how we get them to interact with information. Interesting. So a little bit of
Rob Kelly:a flip on where a bunch of education went online, is it possible that we might see this flip back to the physical hangout in a learning environment becomes more important than ever because it's got the community?
Doug Leeds:More important than ever. Exactly. Provided that schools understand that and pivot into that need. And if they try to just keep saying, oh, we're here to teach you information that you can get in ChatGPT in five seconds, they're in trouble. But if they lean into their opportunity to be a community, we talked about this with Red Hat, a community where people can interact with each other, then they really have something differentiated.
Doug Leeds:When you think back about your college education or mine, and I think about my college education, my brother was making this point to me the other day because his daughter's couple years away from going to college. It's not what we learned in the classroom, it's what we learned from each other that really mattered and what we keep with us. Do you care if their deliverables were created with AI or not? Absolutely, I care. But let me describe that.
Doug Leeds:I think I'm not doing my job if their deliverables can be created with AI.
Rob Kelly:But you allow their use of AI. There's not some sort of ban of using AI in the class?
Doug Leeds:I allow look. No. I don't have a ban. Berkeley has recently said you have to come up with a policy and describe what that policy is. And my policy is you can use AI provided that you tell me when you're using AI and that it's still your thoughts and you've thought about it.
Doug Leeds:Like, don't like if they put something in AI and they submit it and they haven't thought about it. If they put something in AI and it comes up with an answer and that answer looks like what they would have written but it's faster, then great. I do that too. But even more importantly than that is I've got to make sure that my classroom isn't designed for the kind of information that they can just get from it. Like it's gotta be something that you can't do with AI.
Doug Leeds:Like Yeah. Going out in the community and changing people's minds.
Rob Kelly:Otherwise, it's a bad product.
Doug Leeds:It's a bad product. Exactly. That's exactly right.
Rob Kelly:Do you consider yourself an AI optimist, pessimist, or some other descriptor?
Doug Leeds:I'm an optimist. I'm an optimist.
Rob Kelly:How do you feel about AI replacing your job? Your role, any one of them? You've got at least two jobs I know of.
Doug Leeds:Yeah. I mean, I guess this is where being on the back half of my career makes it a little bit easier to take because I don't have the years in front of me of having to worry about that as much. The Internet disrupted things. Like my dad's age, he grew up without television, that disrupted things. Jobs changed.
Doug Leeds:I believe in humans and I believe in the value of humans. And I don't know if that's anachronistic belief, but I still believe in it. So when we get new tools that we can make humans more productive, it makes me excited. Does that mean there's gonna be disruption and so jobs that used to exist are gonna change? Absolutely.
Doug Leeds:Does that mean I'm pessimistic about how that's gonna work out? I'm not because I believe humans will persevere.
Rob Kelly:How do you feel about universal basic income? I
Doug Leeds:haven't thought too much about it but I like it as a concept. It's again, it's another one of those things that is freeing. If we can allow people to have the confidence to know that they're gonna be able to be fed and that they're only gonna be able to be housed, then where can our economy and our society go with that comfort? It's an exciting thought. Of course, there's gonna be free riders on that.
Doug Leeds:There's gonna be people say I don't want to participate and I don't want to do anything because of it. And I do think that there was a lot of mistakes made with the ideas of communism that had some of those noble ideas in them and then saw some real bad stuff happen out of it. But I believe that there's a path that that can work. It's just it's just a hard path.
Rob Kelly:What are you telling young kids in your life, like your daughters, preparing them for this new world of AI?
Doug Leeds:Well, I mean, I think a couple things. We talked about it already a little bit, which is your curiosity is your strength and be interested and learn and get excited about what you're excited about and lean into that. Going back to universal basic income, it's hard for a lot of people to do that because they're they're not it's I have to pay my bills. I can't lean into my curiosity. I don't have the ability to do that if I have to worry about where my health care is coming from or my housing is coming from or my meals are coming from.
Doug Leeds:But if we could do something where people could actually lean into their curiosity, and this is what I tell my kids, the opportunities for you both in terms of financial opportunities, but in terms of fulfillment opportunities are the greatest. And I guess that's the last thing I tell my kids, which is you get to define your own goalposts, your own scoreboard. The world's gonna tell you it's based on how much money you make, and I think it's based on how happy and fulfilled you are. And those two things, I know plenty of people who have hundreds of millions of dollars or billions of dollars and are not happy at all. I live here in LA, I can go to the beach and look at like the Jeff Spicoli type of people who all they need is a Tasty Wave and Cool Buds and and they're okay and there's nothing wrong with that, man.
Doug Leeds:That's great.
Rob Kelly:If AI ends up doing a lot of the work, your work in life and you have endless free time, what would you do with all your newly found time?
Doug Leeds:Talk to people. I mean, I really enjoy it. I really enjoy learning what's going on in people's minds and being social and learning what motivates people. That's why I teach actually because I I learn more UC Berkeley's School of Business where I teach has something called defining leadership principles and one of them is student always. I would just be a student, always if I could.
Rob Kelly:Will you create or have you created a personal AI avatar for others around you to use to interact with you, once you're gone?
Doug Leeds:I oh, no. Not for the for the post death experience, but I did I did create one for my TA in my class because she was getting bombarded by all these same questions and they were really annoying to her, so I created her name is Hana, so I created Hana GPT to answer all the questions and had a lot of fun doing that. So I can see the value of it and and my mom is sick right now, she's in the hospital, it's one of the things that is gonna cause me to have to leave in a second. And so I'm thinking a lot about that idea of like how do I interact with it. And if it was a true way of interacting with someone, my dad who has been one of my truly best friends and closest advisors, if I could preserve him his thinking for his entire life, yeah, that would be amazing.
Rob Kelly:So will you? Probably not. How come?
Doug Leeds:Maybe because I don't know how to do it, maybe if it was easy as pushing a button, I would do it. Maybe there'll be a product that comes out and says, here's how you do it and I would experiment with it. He wouldn't like it and I wouldn't do without his permission. I guess the better question is, would I do it for myself, for my kids? I would.
Doug Leeds:Would you?
Rob Kelly:Would I? You know, I've been asking this question a lot and someone the other day said had the best answer I've heard so far. I don't wanna bias future people from answer the way they answer, but their answer was really interesting. It was if my kids asked for it or wife. Yeah.
Rob Kelly:And that actually felt organic and sort of natural, like if they wanted it, then I would. So I'm open to anything, but I like that answer in particular.
Doug Leeds:I love that answer. Although, the question then becomes what if they don't know they want it until after you're gone? Do create it and then put it in a safe deposit box or a metaphorical safe deposit box with a note that says it's here if you ever want it.
Rob Kelly:I mean, I'm laughing, but stranger things have happened. Yeah. This might this might be, you know, the digital PO box or safe is gonna have a little Right. Avatar of Doug Leeds and Rob Kelly. Hey, Doug.
Rob Kelly:I've really enjoyed this. Anything that I didn't ask you that I've I should
Doug Leeds:been told to always say yes to that, but I really think you did a great job across the board. I can't think of anything. So I'm gonna say no. Someone who ever told me I should always say yes is gonna be upset with me, but I
Rob Kelly:think no. I liked it. Well, this case, I'm glad because, that gets you to your mom all the sooner. So Thank you. Thanks.
Rob Kelly:I really enjoyed the conversation. Thanks so much. Likewise, Rob. Well, this is Media and the Machine. A few things about you and me.
Rob Kelly:If you wanna hear about the next new episode, make sure you hit follow on the show in your podcast app. If you wanna go a little deeper, head to mediaandthemachine.com and subscribe. When you share your email with me, you can see handcrafted transcripts, read the essays in my newsletter, and be the first to hear about who the guest is on the next show. You can also email me directly from there. Maybe you wanna recommend a guest.
Rob Kelly:I'll give you a shout out if you do. I love paying it forward. From time to time, I also open up office hours and host small meetups for subscribers, just to meet, talk, and build things together. If you're creating something of your own or thinking about it, I'd love to help. Maybe you've got a podcast in you.
Rob Kelly:Finally, I don't have a marketing budget for this show. So if it's finding you and others, it's because someone like you passed it along. I'm genuinely grateful. If you have a moment, an honest rating helps me make this better for you. You can just go to the show page and click one of the stars.
Rob Kelly:And I'd rather you give me a low rating than no rating at all. I mean it. It pushes me to get better. Thanks again, and see you next time.