Exclusive: Jonah Peretti explains why he sold BuzzFeed

Today, I’m talking with Jonah Peretti, who is, technically, the CEO of BuzzFeed — although that will be coming to an end very soon.

Just days before we spoke, Jonah agreed to sell 52 percent of BuzzFeed for a total of $120 million to Byron Allen, who owns The Weather Channel, a number of broadcast stations, and several other websites. The deal is a bit of a life raft for BuzzFeed — the company was once valued at $1.6 billion dollars, but just last quarter, the company had told investors it was at risk of running out of cash. Now there’s a new lease on life — and new leadership. As part of the deal, Jonah himself is stepping down as CEO and taking on a new role as president of BuzzFeed AI, and Byron Allen himself will become CEO of BuzzFeed. 

That’s obviously a huge structural and organizational change and a really big decision — prime Decoder bait if there ever was any. And, of course, I’m very interested in what digital media companies are doing to adapt and survive in an information landscape dominated by algorithmic social platforms.

Verge subscribers, don’t forget you get exclusive access to ad-free Decoder wherever you get your podcasts. Head here. Not a subscriber? You can sign up here.

After all, I’ve been saying for a long time now that the original sin of digital media was Jonah and BuzzFeed betting they could so consistently go viral that platforms like Facebook would pay them for content — the way cable companies pay carriage fees for channels like ESPN. This was the big bet for a lot of companies all chasing BuzzFeed’s influence and valuation, and it’s mostly all come crashing down. So I really wanted to know if Jonah had reflected on that and how he saw the work of building audiences and influence now.

I also really wanted to talk to him about his new role leading AI. In the press release announcing the sale, Byron Allen says BuzzFeed will now compete with YouTube through the power of AI. That’s quite an ambition, and I was very curious about what it specifically means. Jonah is also making lots of games and apps with AI, and we talked about some of them, including his new hybrid of a meme generator and social network, BFIsland.

Of course, we also talked about where it all went wrong and what Jonah might have done differently. There’s a lot going on in this one, and Jonah was pretty open about it all. But I’m still not sure I know what’s going on with that YouTube plan. You tell me.

Okay: Jonah Peretti, still — for the moment — CEO of BuzzFeed. Here we go

This interview has been lightly edited for length and clarity.

Jonah Peretti, you’re the co-founder, and as of today, still technically the CEO of BuzzFeed. Welcome to Decoder.

Thanks for having me.

I’m excited to talk to you. It feels like BuzzFeed, Vox Media, and The Verge have all come up and weathered all kinds of storms together in digital media. I say that you are still, as of today, the CEO of BuzzFeed because just a few days before we’re speaking here, you announced a huge deal to sell 52 percent of BuzzFeed to Byron Allen, and you are going to take a new role as president of BuzzFeed AI. Explain what’s going on there.

Yeah. We were looking at a few possible deals that were transformative for the company, and I’m very excited we ended up with Byron. He is a force of nature, an incredible media mogul. He owns all kinds of different assets. His skills are very complementary to mine. I mean, he’s in the mix with advertisers and with partners and with sources of capital in a way that I never really have been. So it’s super exciting to have him come in. It also provides liquidity for the company and resources.

And then the thing that I’ve been most excited about working on is really trying to reimagine how a company should operate in a world where these new AI technologies have become so much more advanced. So I’m going to have the opportunity to spend more time on that, which is the thing that I’m most passionate about. So it’s been a really great week, and we’re happy we were able to close this deal. It’s going to be transformative for the company.

One of the reasons that I originally scheduled this interview with you was because there were some quarterly earnings, there were some statements in your financials about BuzzFeed’s ability to continue operating, whether or not you had enough cash. There was some speculation that the company was pretty close to bankruptcy, but the brand is strong. I mean, BuzzFeed is one of the most famous brands in digital media, if not all of media. Did you have incoming suitors after that last set of financials? Or was this something you were talking about for a long time?

I mean, it was frustrating. We had a going concern statement, which means essentially, as a technical accounting term, that we didn’t have enough capital to cover our expenses for the year. Which, obviously, is a serious thing that we wanted to and needed to disclose to investors. But simultaneously, we had a lot of inbound interest in our assets and in partnerships. We were talking to creditors, we’re talking to all the people you would imagine about the injection of capital and things that could be really transformative for the company. But, of course, you can’t talk about things that aren’t signed, that aren’t done.

So during that period, it was quite frustrating because I saw a lot of exciting prospects that we had. This deal with Byron is one of those. It’s really exciting to be able to move forward with more swagger and more confidence and be able to start to reimagine the company and really do a more fulsome turnaround to create a business model and an approach that is for the next five years instead of an earlier era of digital media.

Take me inside Byron’s pitch. Was it “I’m just going to give you money, and you can do what you want”? Is it “I have a plan to reboot the entire company”? Compared to the other folks you were talking to, what made his pitch seem like the winner?

I mean, to me, what was most exciting is that he’s a better media executive than I am, and he could be a media executive, and that could allow me to be more of a tech executive. So that’s always what I’ve loved the most, the intersection of tech and media. Really being able to think about what these new technologies enable us to do in the future that wasn’t possible. That, plus the capital to actually fund a turnaround and to restructure and reimagine the company, those two things together were very enticing. His ability to bring things to the company that we were just lacking was a huge part of it.

I’ve known a lot of media executives; you’ve known a lot of media executives. You’ve just evaluated two of them. You’ve said you’re not a good media executive, and he’s a good media executive. What in 2026 separates a good media executive from a bad one?

I mean, I think we’re in an era right now where deals really matter a lot, and so deal-focused executives are having a lot more success than trying to just build organically. I think the connections with advertisers and the connections with partners and marketers, and being able to see from their perspective and sell them things that they get excited about and inspired about, is really important.

I think in the kind of peak growth period of BuzzFeed, 2013, 2014, in that era, really understanding the technology was the key to being able to build a good media company because the technology was changing so quickly with the rise of social and social platforms, that you really needed to be deep on the tech side. And there was tons of organic growth available, and the platforms were very open in a way that allowed you to drive tons of organic growth.

I think today, if you look around the landscape, it’s a lot more about deals, partnerships, business development, and things like that, and that can be really transformative for a company, whereas the organic growth on the big platforms is really quite anemic. There’s not as much there in terms of sending traffic to publishers, platforms, and content.

Let me dig into that. When you say the technology at that time, 2013, ’14, ’15, what you mean is that social media platforms were growing, they were aggregating a bunch of users, and then they would send traffic out. What specifically was the technology that you needed to understand at that time that made BuzzFeed successful?

Well, I think if you look at pre-social media and what media was about, there were newspapers, magazines, and broadcasts. None of it was really social. You’d watch a TV show, maybe there’d be some word of mouth where you’d talk about it with your friends. You’d read a newspaper article or magazine article, but you really couldn’t share content. I think the world didn’t quite realize that once everyone was connected together by these social platforms, and once you could connect, when you could use media as a way of connecting with other people and sharing with other people, that the kind of media that would thrive and succeed would change. We saw that really early, and that was what allowed us to grow so rapidly and build such a big brand. So I think the technology drives a change in the medium, and so social media became a new medium that wasn’t possible before.

I would say there was also the convergence of social and mobile. Mobile was the perfect personal device that was also a social device, plus the social media, and those trends converging really changed what media became. I think we’re seeing the beginnings of a similar shift now, where AI is going to start creating a new medium for content that wasn’t really possible before. So there’s part of the media industry now where understanding tech is really important, and that’s the part that I’m most excited to focus on.

Yeah. There’s the new thing you’re going to build. I want to talk about that. I do want to talk about the structure. I just want to stay focused on the 2013, ’14 era for one more turn here. Because I think a lot of media companies have made all the same mistakes, and BuzzFeed is like the vanguard of making mistakes. 

I have often described the original sin of digital media as you, Jonah Peretti, betting that you could go so viral so often that Facebook would pay you money, that the social platforms would pay for the programming. In particular, BuzzFeed was great at programming the social networks of that era. I will never forget watching people put rubber bands around a watermelon with like 90 million other people. There was something about your understanding of the dynamics of those platforms at that time that made a lot of people bet that Mark Zuckerberg would pay you a carriage fee the way that the cable networks paid ESPN a carriage fee, that your content would make their platforms compelling, such that they owed you money.

At some point, I think they all woke up and realized there was an army of teenagers who would work for free and that they could get the same dynamics without paying anyone any money, and then all of our businesses changed, and BuzzFeed’s business changed maybe the most dramatically at that specific moment. What was that experience like for you when you went from the top of the world, where you thought you had enough leverage to get Mark Zucker to pay you money, to “Oh no, there’s an army of teenagers who will work for free”?

Yeah. I mean, first, just as a factual matter, we did get paid millions of dollars by Mark Zuckerberg. The prediction that they would pay for content was accurate, but it was just short-lived. So we got paid for that exploding watermelon. 

But those were all those pilot programs, like, “Stand up your Facebook live team, we’ll give you the seed funding to buy cameras.” There was never actually payment for content. There was no ongoing revenue relationship.

Oh, there absolutely was. We got paid rev share for videos on Facebook. The short Tasty videos that initially blew up — there were no payments for that. But then, when Facebook realized they needed to make longer content so they could put in mid-roll ads, there were programs that generated millions of dollars for our company, where we were making longer-form video for Facebook. We were getting rev share payments on that, and millions of dollars from YouTube. So it’s not that it didn’t happen. It did happen. The news tab was another example of it. We got paid millions of dollars for putting news on Facebook.

I think what you’re saying is accurate in the sense that it went from free distribution and sending traffic to our site, to hosting content on their platform and getting paid for it and a rev share for it, to an explosion of creators, which made the value of content and the platform’s willingness to pay much lower. It was happening, and it was working, and it was exciting. I actually believe that it was a mistake for Facebook and Mark Zuckerberg to not continue with the news tab, to not continue to pay professional content creators. I think having a diversity of content creators, professional entertainment, and professional news… These platforms have tons of resources and could have spent a couple of billion dollars a year on that, sustained a vibrant media ecosystem, really owned media in a way that would have been incredibly powerful for them and given them charisma, relevance, and authority that they just don’t have today.

Like now, you look at the lawsuits about addictive behavior, you look at all the kinds of toxic content on the platforms. They didn’t have to do that. They could have just continued to pay and even ramped up the amount they paid to professional content and to news content, and they would have owned the world. It was actually a mistake that they didn’t continue it. They did it, and they should have continued it. 

The fact that they didn’t, I think, shows a huge blind spot that the tech industry has, having a small percentage of your revenue going to underwriting a robust, diverse, trusted ecosystem for content and news and other things would have been a great decision for them and would have been much better long-term. So I think there is some parallel universe where they don’t get this wrong. They started on a good path. They were paying content creators, it was expanding, and then it was like, “Oh, let’s get rid of the news tab. Let’s stop doing rev share, or let’s show vertical video where we don’t have rev share instead of longer-form video where we do.” Over time, it just really messed up the digital media ecosystem in a serious way. It led to the world we have now, which, although I love a lot of creator content, the sort of battle to get into the feed, being the most extreme, saying things that are just completely false and outrageous, or whatever, has been a huge missed opportunity for those platforms.

It’s not that I don’t agree with you, but they won’t anyway, right? They stopped investing in news. They piddled away however many billions of dollars on VR headsets, and they are, in fact, now the gatekeepers to all of media. It didn’t matter whether or not they paid for news. A bunch of Congresspeople yelling at them has had no meaningful impact on their business whatsoever.

I’m certainly not arguing that they didn’t have tremendous financial success, and maybe a couple of billion extra in profits is worth it to them. But I think that being an influential company with real relevance, more charisma, and cultural muscle is something that a big tech monopoly could certainly afford, and ultimately would have been better for them in a bunch of different ways. I think the biggest risk of the big platforms right now is essentially a PR risk. The public turning against tech, the public thinking tech is making the world worse, the feeling that any new technology is suspect, because look what happened in the last round of technology. So I don’t think it’s good for them, even if they’re maybe a little bit more profitable.

I think being marginally less profitable and having a buffer against this major backlash that is brewing, where kids are trying to get off their phones and delete their apps and things like that, and feel like it’s an addictive product that they’re trying to battle with. I mean, I think they could have avoided a lot of that just by operating differently. So they have won, I’m not arguing that, but I think that they could have won in a deeper sense if they had continued that basic process of helping to underwrite quality content, news, information, and entertainment, which they certainly could have done.

I’m only sitting on this because, honestly, how many other people have lived through all of the twists and turns and have made the decisions? It’s been a real journey. I guess I’m looking at it now with a little bit of distance, talking to you. I’m so nuclearly opposed to being dependent on the platforms. The Verge has always wanted to be its own thing, especially because we cover these companies, and you just have a different perspective. I think my response to all this is, should you build a business on this handful of billion dollars, will it make people like us more in the long term? Or was there ever a point where you felt like you actually had the leverage to make the social platforms pay you real money in a sustainable way?

If you look at the cable industry… Now, Byron Allen owns The Weather Channel, and The Weather Channel was part of this amazing history of cable, where these media companies were paying for carriage. They were like, “I’ll pay you a dollar per subscriber so I can get The Weather Channel in homes,” when cable was new.

The cable operators, I think, at a certain point realized, “If they’re all paying us a dollar to reach the home, they’re going to have a really bad business, and they won’t be able to invest in content, and then people aren’t going to sign up for cable.” So they’re like, “Why don’t we switch it? Why don’t we pay you a dollar?” Now, they didn’t have to do that. They could have just said, “We’re only going to carry channels that pay us.” But instead they said, “Why don’t we pay you a dollar per customer and make your programming better?” When you make your programming better, more people will be like, “Oh, I really want to watch cable.”

So The Weather Channel switched from paying a dollar to getting paid a dollar, and that wasn’t because the cable operators didn’t have leverage; they’re the only way that you could even get into the home. They did have leverage. They just saw, on a deeper level, that having better content would, in the long run, be good for everyone, and then the whole pie grew because of it. So I think that the answer isn’t, did individual content companies have the leverage to dictate terms to Facebook or to YouTube? We certainly never did. The question was, are there leaders who have enough market power and concentration to see that it’s in everyone’s interests, it’s in society’s interests, and it’s in their interest to grow the pie and pay for content to have different kinds of content?

I think early on, talking to the tech leaders, they didn’t really even understand or think about the cost of producing content. So I would say, “Hey, news costs a lot more, doing fact-checking, calling people up, doing interviews. If every piece of content is competing against every other piece of content, there’s no incentive to ever make any news. You should just make the cheapest, most popular types of clickbait content, or something like that. That’s not good, that’s a race to the bottom, that makes your platform worse.” And I think they started to get it, like, “Oh, let’s add a news tab,” or, “Let’s do some programs to make sure that there can be different kinds of content.”

So I think it was less about having the power to dictate terms and more about looking at the history of media and saying there are ways that the pie can be grown for everyone, and the economics needs to work for everyone, and the incentives need to be right for everyone. When that’s the case, the whole market gets bigger. That’s something that I think they kind of wanted to do at some point, but ultimately, when things got tough, they decided to just focus on their own profit growth and not necessarily-

They’re paying less and less for content today. I mean, TikTok pays nothing effectively. The creators have to become little ad agencies themselves, which is another thing you pioneered, the idea of the branded content studio that you would distribute advertising onto these platforms in a way that went viral. Even that got outsourced to an army of influencers and creators who are now all standing up businesses with the same kind of dependencies that you experienced.

I consume content on Instagram, but you kind of feel like you’re watching… The actual editorial content is ads, where creators are promoting themselves or products they’re selling. And then in between that, you have ads. So you’re having a billion or more people just hanging out on an ad network all day. I think there’s a better vision for media than that which we could have all built together. But now, since that didn’t happen, we have pivoted really hard to direct traffic.

I mean, I’m so happy that HuffPost has so many people coming directly to HuffPost’s front page every day, and getting the news from an independent source that tells it like it is. It’s something that people are really hungry for. More of BuzzFeed’s traffic is now direct. So we’re really realizing you can’t have someone between you and your customer. And that’s a huge thing, that Byron Allen also, one of his rules of business is, “Don’t let anyone get between you and the customer.” And that’s part of the reason I think he was interested in BuzzFeed. It gives him a way to go directly to audiences. And for all of the ambitious things we want to do in the future, we want to have that ability to go to tens of millions of people directly from our platform.

Let’s talk about the deal. Under the terms that are announced, Byron Allen, through his company, Allen Family Digital, they’re going to acquire 40 million shares of BuzzFeed for a total purchase price of $120 million, which will lead them to owning 52 percent of the company’s outstanding shares. It’s $20 million at closing, and then $100 million that will be due five years from closing. How does that work structurally? At the end of all that, how is BuzzFeed going to be structured?

We’re still a public company. It used to be we were a public company that I controlled through super-voting stock, and now we’re a public company that Byron controls through the 52 percent ownership.

So you’ve given up your super-voting stock in this?

Yes.

That was actually my next question: Where are the 40 million shares coming from? Are they coming from you?

No, we’re issuing new shares.

Okay. But you’re dissolving your super-voting shares?

I am just converting my shares to be low-vote shares or normal common stock.

Great. And then of the $20 million, you said this is going to give you some operating cash. How much runway do you get out of $20 million?

I mean, I think infinite runway is the plan. I mean, if you look at our cash burn, it’s been pretty insignificant. I think we also announced on our earnings that we plan to do a restructuring. So I think setting the company up to operate profitably and to do that for the long term, and then have a strong foundation that we can build on top of, and new initiatives are coming on top of that profitable platform of our core businesses.

Let me just ask the follow-up question here. So you’re issuing new stock. Are any of your existing investors or employees who had equity getting paid out as part of this deal?

No, no. This was about bringing capital into the company so that we can have a stronger balance sheet and be able to start going on offense again.

You’ve mentioned a restructure. I noticed there’s a restructure in the press materials. This is a show where I ask everybody about structure. What does restructure mean for BuzzFeed after this, too?

Well, I think it means a few things. It means looking at our strategy and where we’re headed in the future. Some of it is around understanding how we’re going to operate with new technologies that can change the way the company operates. I think some of it is about just making sure that we have enough buffer to weather ups and downs in this industry. So being profitable enough so that if traffic goes down 20 percent, or if revenue or ad markets shift or change, then we’re in a strong position so that we can continue to operate.

Be more specific. New technologies, you have a new title of president of BuzzFeed AI. There’s something called Branch Office, which was your app skunkworks inspired by Nintendo, I think, which is what you told The New York Times. Is that going to get set up and more capitalized and become less skunkworks? How do you see this working out?

Part of it is, how do we use AI across the business as a whole? I can share some things, but not everything. Part of what I think AI is really good for is that it can almost be like a nervous system that is able to understand and detect what’s working and what’s not working, and how content is being shared and engaged with. Some of the early stuff that BuzzFeed did about social content, I think you could do on this whole other level when AI is able to actually understand the content. And then push challenges to creative employees to make new things, but they’re making new things with a lot more information about how people are engaged with the content. So just to put it in simpler terms, using AI to help our core business and help our people be more creative. That’s one part of it.

Another part of it is building entirely new apps. Branch Office is an incubator that creates new apps. We launched BF Island today, which is a new kind of messaging app that allows you to play, message, and talk with your friends using GenAI. So it’s kind of like an inside joke engine. We found that making AI content and posting it on Instagram or TikTok just kind of feels like slop, but making funny things with your friends that are about the specific things that you’re joking about and talking about is just a lot more fun. So we built BF Island for that use case.

We also launched an app called Conjure, which is a camera app where you get a challenge every day to take a picture and discover a mystery every day. You can, for example, take pictures to see if there might be UFs hiding in the sky near where you live, or get messages, fortunes, or other things on Conjure. So I think when you think about BuzzFeed AI, it’s a combination of things that integrate across the entire company and help our teams do their jobs better, and then it’s also new apps and new experiments that allow us to test AI as a new medium that could potentially lead to new businesses or new apps that break out.

There’s what you’re saying here, and then there’s this quote from Byron Allen about what BuzzFeed is going to be. I’m just going to read it to you, and I’m hoping you can explain to me how this will work. Here’s Byron Allen describing why he’s buying BuzzFeed: “As of this moment, with the power of AI, BuzzFeed is officially chasing YouTube to become another premier free video streaming service.” How are you going to do that?

I think one thing that is pretty exciting about working with Byron is that there are more assets, more resources, and more capabilities to do things that we just couldn’t do on our own. I don’t want to give away too much, but he’s doing a lot of things in these wholly-owned businesses that he has. He is in the home with Local Now, a connected TV app. He’s doing a huge amount of production with his studios for making content. There are conversations, deals, partnerships, and things that he’s able to access. So I think what that means is you’ll have to wait and see. But putting together pieces and combining that with new technology and making something that doesn’t exist is why I’m so excited about this deal.

What would YouTube be if it started today? What would YouTube be if it were something that was created in the world of GenAI? I think it would be pretty different. It might not be what every listener is imagining. It’s certainly not just making a bunch of AI slop videos or something like that, which I think is the thing that people jump to. But there are different ways that creators could make, play, and share. We have a lot of stuff to work out and a lot of stuff to build, but I’m excited about combining a lot of new capacity that we didn’t have before to build things that seemed out of reach in the past.

You know the folks at YouTube. I know the folks at YouTube. I have my own complaints about YouTube. They’ve heard them all endlessly. But if I had to put myself in the shoes of Neal Mohan, the guy who runs YouTube, I would read this and say, “Good luck.” It’s very hard to chase YouTube. TikTok had to spend a billion dollars a minute on Facebook ads to get enough users to chase YouTube. I understand that you can’t give it all away. I’m just saying, “With the power of AI, BuzzFeed is officially chasing YouTube” is a pretty big ambition. I’m curious what the power of AI means there. Is it just operating leverage? You do more with less. Is it that you can vibe code a YouTube app? How does that work?

I think you’ll have to wait and see. Neal probably listens to this podcast, so I don’t want to… Hey Neal, if you’re listening, let’s go to a Warriors game next year, and we can share an app.

There’s a distribution challenge that all these platforms have, which is somewhat driven by consumption. Vertical short-form video became a dominant way people wanted to consume. So then Instagram turned into TikTok, and then YouTube turned into TikTok. And then maybe a bunch of people on YouTube didn’t want YouTube Shorts, and now you have the option to turn YouTube Shorts off. What they really want is long-form creator video, and the economics are better there. I’m just saying, what is the opening that you see that this will be a competitor in this way?

It’s going to be a lot easier to put your content in more places and to have content take different forms and different shapes because of these AI tools. So a simple example would be, if you have a video, having that turn into a BuzzFeed-shaped object that can live on our platform and be optimized for our platform feels like something that would have been a lot of work and challenging a few years ago, but it’s kind of trivial now. That opens up some possibilities in the beginning.

My wife reads BuzzFeed every night. I watch her wind down by reading BuzzFeed roundups of social media posts. That, to me, right now is the BuzzFeed-shaped object. Look at what these people are saying on Reddit. Look at what these people are saying on Twitter. There’s some value there. I think curation and taste, and being able to see the trends and bring them out together in some kind of synthesis has value. Is that-

Has she played Word Chain?

I think so, but there are so many word games going around our house at once that it’s hard for me to determine which ones are in style at the moment.

I mean, the BuzzFeed games have really exploded recently, and that’s another great example of the fact that writers can vibe code games. The New York Times launches one game every year, and we’ve done hundreds of games this year with creative people who are writers and not programmers, by having a toolkit that allows them to build in that space. We’ve had a bunch of breakouts. It’s become a big percentage of growth in time spent, and the comments from the audience are just that they love them. It allows us to iterate and evolve them more quickly and change them more quickly. That’s another example of the AI acceleration of what is creating new BuzzFeed-shaped objects. So I think that level of being able to iterate more quickly and mutate and evolve content is something that allows us to play in new spaces, whether that’s user-generated video, interactive games, or new post formats, quizzes, and things like that.

The challenge there is always distribution. I mean, this is where we started talking about Facebook and the other platforms. How are you going to get new users to consume the stuff you make? Ten years ago, 15 years ago, it was, “We’re going to put up links in the open web, and they’re going to drive a bunch of traffic to links. Eventually, we’re going to do a llama in a dress in the same 48-hour span, and that’s the future of media.” And that was great. Then, the open web collapsed. There are bigger publishers every day. I think Roger Lynch was literally saying today that Google Zero is there for Conde Nast, and they are just betting that there’s no Google search traffic anymore. Roger should give me credit for that phrase, by the way, but we’ll come to that in a different episode. They do all listen.

The question I have for you is… You can invest more in BuzzFeed-shaped objects on BuzzFeed, but where do you get new readers, new game players from? Because that seems like the hardest problem now for digital media.

We have a majority direct audience now. Some of that is because of declines in other audiences, you know? But tens of millions of people visit every month. You can look at our Comscore; you don’t have to trust me on it. It’s like, we’re the biggest publisher in our competitive set. We’re bigger to people. We have a lot of people coming. When we see new formats released, like, for example, these BuzzFeed games that we have been creating, they come back more frequently. They spend more time, they engage more. They send the games using iMessage to friends, so you have this new kind of social that’s more personal, that’s more private. So we have distribution that if someone were to create a new company and try to do any of these things, it’d be very hard to achieve. We haven’t really started promoting our Branch Office apps, these new AI apps. But soon, Conjure will be all across BuzzFeed, and BF Island will be all across BuzzFeed to market it and let people play it.

So we built a lot of distribution in a different era, and we have millions of people consuming our content. That is the distribution that can help launch these new things and help us grow, and launch something that could be a competitor to YouTube or a competitor to some of the other big platforms for people who are getting sick of these big platforms and want smaller places that feel more at home, that they can go directly to and spend their time on.

Back to our earlier part of the conversation about whether it was a mistake for Facebook, YouTube, or others to sort of pay less for content? I think that it did result in a lot more people going direct to other platforms and smaller platforms that they probably never would have if they could have gotten quality news on Facebook, Instagram, or other places. But now people do go to those places. It hurt to have a lot of that referral traffic taken away, but now we have the direct traffic, and that direct traffic is people wanting something different that the platforms aren’t giving, and that’s a big opportunity to build on top of.

That’s the new kind of distribution. It’s more fragmented. It’s not everyone on the internet, it’s not the dress, but it’s millions of people who really, truly value the content. They’re coming directly, saying, “I get my news every day from going to the front page of HuffPost, and they’re not trying to kiss up to the Trump administration and worrying about mergers or whatever. They’re just reporting the crazy stuff that’s going on in the world. And I go directly to BuzzFeed and know what’s going on across social media,” because as your wife has realized, it’s a lot more satisfying to see roundups of what’s happening across social media on BuzzFeed than to be in these toxic environments for hours doomscrolling through the sort of raw sludge that’s on these platforms.

You were always really against display ads, programmatic ads, banners, and boxes. I always thought you were correct about that. I dislike them. I think the vast majority of the audience dislikes them. But that’s how you monetize a bunch of webpages to this day. It’s still the thing you do. You have a bunch of direct traffic, and it’s growing. You think you can create more direct traffic. You think there’s an opportunity there. But the monetization was lagging such that, last quarter, you were saying there was substantial doubt about your ability to continue, and you needed to take investment. How are you going to monetize all that direct traffic in the future in a way that’s sustainable?

I think it’s a combination. There’s sort of a floor that is driven by programmatic advertising. Then there are transactional types of revenue. Transactional revenue, I kind of mean that abstractly. So it might mean commerce, where people are spending hundreds of millions of dollars on BuzzFeed Shopping, where they click through and discover products and buy them, and we’re getting the affiliate fee, so we can actually show that we’re driving direct value.

And then there’s the other kind of transactional, which is paying to be a member, subscribing. HuffPost has a membership program that has been growing really nicely. I think that there’s a lot of headroom there where we can grow more with reader revenue and direct revenue. Branch Office apps are really natural for freemium models, where you can have fun on the apps to a point, but if you want to go deeper, and with Conjure, if you want to start exploring and photographing more weird things in the world, you’ve got to become a paid member or transact with us. So I think that’s a good combination.

We were playing with BuzzFeed Island today. I generated a bunch of pictures. Our producers put a clown nose on me. That was delightful. I thank them for that. It just occurred to me that I could burn enough tokens to make my use of the app unprofitable today. I could just sit here and generate images all day until I had definitely cost you some meaningful amount of money. How does that work before you monetize it? Who are your model providers? What’s your rate of spend on letting people use the app versus when you think that will return in any kind of monetization?

Yeah. I mean, I think this is pretty established that the freemium model for these kinds of apps is pretty natural. People have shown a willingness to pay for AI services. So I think-

Wait, have they? I mean, ChatGPT has what, 900 million users, and very, very few of them pay anything.

I mean, how much revenue is ChatGPT generating from subscribers? I mean, it depends on how you look at it.

Not enough to pay their bills, I would say.

But there are a lot of smaller freemium apps and a lot of companies that have five million, 10 million in revenue, and are profitable. You might not hear about those because they’re not big venture-backed companies, but I do think that these mid-size and smaller types of apps can be really good businesses. In some ways, it’s a little bit like a new kind of media business. The economics might be more similar to a YouTube show, a series, or a podcast, where the cost of the talent and the show is a hurdle that the advertising or the subscription needs to clear. But I think there are a lot of examples of that that I’ve seen, talking to other founders and getting a sense of that market. So I feel like there’s a bunch of different ways to monetize.

It’s also reminiscent a little bit of the earlier era when you had Moore’s Law making everything cheaper. You could have a product that lost a little bit of money, but if you wait a year, the processing is cheaper, the costs are cheaper, and AI is progressing incredibly quickly. So anytime we use a new model, six months later, the cost of it is like a 10th of what it was. So if we could lock in a bunch of users to an experience that’s not about creating the most beautiful images, it’s about joking with your friends, getting ideas for what to make… BF Island is kind of an idea engine where there are all these cool contexts and things that you might not have thought of that you can then use to delight your friends, or play. The cost of that is… Making those images is going down a lot. So, a breakeven subscriber six months later is a very profitable subscriber.

Are there going to be ads there? Or are you just hoping for freemium? Are people going to pay to unlock different prompts and whatnot?

I think there could be some pretty interesting native advertising solutions there, but we’re just very early, and we’re just opening this up to the public now. We want to get real people using it, and then iterating, changing, and evolving the app very quickly. I mean, this is the other thing that AI enables, which is the ability to iterate and change a product are just so much easier. So getting to product market fit becomes a different endeavor if you can get data more quickly and analyze it and understand it, and then modify and test features much more quickly. And that applies to the monetization side as well.

It strikes me in this conversation, “We’re going to compete with YouTube with the power of AI,” and then playing with BuzzFeed Island — this thing is a social network, right? You make images, you share them with your friends. There are incentives to invite your friends. And after all of this time, what Jonah Peretti wants most of all is to just run the damn social networks himself. That’s what this looks like to me: you’re building social products because the ones you were relying on screwed you over so badly.

There’s a little bit of that for sure, that I feel like I gave them advice from the perspective of someone who cared about content and culture and news, but also spoke their language in terms of tech and social platforms. They took it for a while, but then they kind of went in another direction. Building really good community, building really good social platforms, building really good content networks where people can share content with each other and consume content and have great things to talk about, I think there are openings to do that. And probably they’re, in some ways, smaller businesses. We might only be a 10th the size of YouTube. So maybe you’re right that we can’t get to 100 percent the size of YouTube, but-

Neal is shaking in his boots.

Maybe we’ll get to 10 percent so he can still have 90 percent of the market.

The turn of the modern social network is creators. That’s the other theme that’s been showing up in this conversation over and over again, right? There was professional content, and there was the idea that maybe we’d pay for it. The creators showed up, and they upended the content economy in a million different ways. 

BuzzFeed, in many ways, was also at the vanguard of that problem, that situation. The Why I Left BuzzFeed video launched a million YouTube careers. You just sold Hot Ones, which you had purchased and then you sold. There’s something about BuzzFeed’s relationship to the creator economy that at different times was really rich and really rewarding, really, really complicated, and then for many, at the end, really poisonous. 

What have you learned from that? If you’re going to stand up new social networks, how do you bring new creators along with you? And was there ever a time when you could have salvaged the creator relationships you had at BuzzFeed?

When you think about BuzzFeed video, it was really awesome for the team that was there because they had all these collaborators. A lot of them came in as interns or fellows. The Try Guys came in that way. A bunch of the talent came in, almost like interns that got fellowships and then became junior producers, and then kind of worked their way up. 

The challenge, I think, became when the platforms really started to disaggregate the media companies, like essentially having individual creators create their own channels without media companies. It meant that the very top creators could make more money if they left, and it kind of started to break the cycle of development. Because essentially, almost every media company, if you look over the last hundred years, has some stars, and they get paid the most, but they get paid twice or three times as much, but they’re producing 10 times as much value. And then you have a lot of new talent coming in, and they’re learning, and maybe they’re working their way up, but they’re getting kind of subsidized by the bigger talent.

So it creates this natural tension in these businesses where the whole process of having a community, an organization, professional development, and new talent coming in, all of that kind of breaks if the top talent is like, “We’ll just do a Substack,” or, “We’ll just do a podcast that we own and take 90 percent of the profits from it,” or, “We’ll just make our own YouTube channel.” Then the process kind of breaks where you no longer have that redistribution that helps the next generation learn — the young reporter, the new video producer, or whatever. So I think early BuzzFeed was trying to build a collective and an organization, and at a certain point, some of the talent were like, “I’m not making enough money. I’m going to go create my own thing.” And that worked for some of them.

A lot of them found that it really sucked to try to be a YouTube creator, and having to call in friends and favors, and not having resources or collaborators, the loneliness of it, and the burnout. YouTube and TikTok love to give the trending designation to new talent, but then a year or two later, it’s like, they don’t get that anymore. So it really is kind of broken for a lot of the talent that was frustrated and left. It didn’t work out that well. But for some talent, it works great.

I don’t know, I think that’s one of the core challenges with media companies and these platforms that are trying to just completely blow up the media companies. It’s one thing if you’re a news organization with a union. It’s one thing if the top talent is stuck to that news organization. It’s another thing if you collectively bargain, and the people who want to make the most money just go to Substack. You know, Substack is probably the biggest union-busting development that exists, even though that’s not its intention.

But there’s this contract of the top talent only capturing part of the value to help support the up-and-coming talent. I think that is poorly understood and is one of the reasons why the media environment has been somewhat impoverished. You have a lot of creators who’ve gotten big and podcasters who’ve gotten big, but never came up through a system where they had smart peers, editors, and people who could help them get better at their job. They just came up through a system where they’re trying to game the algorithm to get attention and get big enough that they can then capture that value for themselves. 

You run HuffPost. HuffPost has this problem and has this dynamic. You’re talking about AI. Are you going to start to publish AI journalism on HuffPost to solve this problem?

No, no, we’re not. Absolutely not. The reason people come to HuffPost is that they can get trusted content that they can’t get on the platforms that are full of AI slop. So HuffPost is a place where talented human journalists who actually do have a sense of collective endeavor and who believe in the mission of HuffPost work together, and they make content that the audience knows is trusted.

Are they receptive to using AI in the news-gathering process? Is that something that you want them to start doing? You’re taking on a new role as president of AI. This is the conversation in journalism.

I think that there are things that AI can do in journalism. I think the example of being able to vibe code and create new games and new formats is one example. I don’t think a HuffPost journalist would… You remember back when “Snow Fall” was this big story?

We did tons of “snowfalls” on The Verge. The big complicated layouts, we love them.

Yeah, yeah. That kind of thing is trivial to do now, right? You can design interfaces that are… It’s trivial from a technical standpoint. You don’t necessarily need an engineer to do that. You need someone who has a good sense of design and good vision, or whatever, and you need the journalism to be good and the story to be good. So I think a lot of it depends. It’s like, how can you use AI to elevate journalism, to promote it better, to connect people to it, to research and find… Crawling the web is something that… How many times have journalists spent three days going around to websites or whatever?

Now, with Claude Code, you can basically say, “I need this data source. I need this data source. I need this data source. I want them all in folders. Okay, I want to cross-reference them.” The kind of fact-gathering and things like that is something that could help elevate journalism and make journalism better. But having AI write journalism would just undermine the core trust that is the main reason that people are coming directly to HuffPost to get their news every day.

I guess I’m just curious how much of the talent business you still want to be in this new world. You’re apparently going to run a thing that competes with YouTube. I assume that means user-generated content. BuzzFeed Island is a bunch of user-generated content, but the BuzzFeed-shaped objects today are roundups of user-generated content on other platforms. How much of the talent business do you want to be in in the future?

Well, I think what’s kind of amazing about YouTube and some of these other platforms is that they’re not really in the talent business.

This is specifically what I’m asking. Do you want to run platforms for users?

BuzzFeed was in the talent business, providing health care and benefits, professional development, a context, real estate, and all these things. When creators would leave to go directly to YouTube, they got a rev share. So I think, essentially, some of what we will do in the future will be more user-generated, like where you look at users making quizzes and creating things for themselves. You’ve played with BF Island; that’s something where the user creation is actually the entertainment. It’s not like they’re creating content, not for the joy of creation.

So I think there are a lot more things like that that are possible. But then, when we do have talent that works for us, I think it’s important to find people who want to be part of a collective, who want to be part of something that they’re doing together, who want the professional development, the resources, the peers, the health insurance, and all of those things. So it just depends on the business.

But I think if you look at HuffPost, it makes a lot of sense to have journalists who have worked together in some cases for over a decade, who really know what they’re doing and are professionals, and that shows in the work that they’re doing. The same is true at BuzzFeed. Figuring out how to do even more to elevate the talented writers, creators, and people making things at BuzzFeed, I think, is part of what I want to do with my increased focus on AI.

You’re at the end of one kind of road here. You’re stepping away from being CEO, you’re going to become the president of BuzzFeed AI. Byron is going to be the new CEO. Again, in this conversation, it strikes me as you were the right CEO for what you would call the open distribution era, and he might be the right CEO for what you’re calling the deals era, where you have to show up, and you have to negotiate rights. Maybe the AI companies are going to pay you for stuff. There are just a lot of deals to be made in a way that, quite honestly, it feels bittersweet for me. The open web era is also where I came up, and I feel very strongly about that thing in a lot of different ways.

At the end of this road, you’re taking on a new role, and there’s going to be a new CEO. There are two moments I want to ask you about that could have been previous endpoints. In 2013, Disney offered to buy BuzzFeed for what Ben Smith has reported to be $650 million. Do you think you should have sold back then?

I mean, from a financial standpoint, it’s hard to say that that wouldn’t have been a good deal to take. It was, I think, $450 million with an earnout that could get to that $650 million number. We were just hiring Mark Schoofs to build our investigative team. We were just starting with BuzzFeedVideo, and that whole cultural phenomenon of BuzzFeedVideo was kind of just about to take off. There were a bunch of things that I’m really glad we got to do as an independent company, and I wouldn’t really trade those years where we were able to build, explode, hack culture, and influence the internet in a massive way. I think it would have been harder to do that inside Disney. It would have been a great financial deal, and I do wish that my partners, investors, and others had been able to have that financial windfall, which it was unfortunate that we weren’t able to sort of live up to that.

The other thing I would say is that I don’t think selling the company at that moment would have been the maximization of value. I think selling it two years later would have been, and there was a lot of interest that people don’t know about in those subsequent years. So the Disney deal was the first time someone came, and I think Iger was ahead of the curve and saw that there was a lot of growth and excitement there, but I think we could have done a transaction for more than that even two years later.

There was a weird thing that was this view, I think, from the board, that the fact that I turned down that Disney deal meant that I would never sell for any price or whatever. So that also partly led to never really exploring any of those other types of deals. But in retrospect, if you had perfect hindsight, there’s probably some kind of deal we could have made in 2015 or ’16 that would have also given us the right partner and resources to fare the next challenging time. 

You’re stepping down as CEO. I’ve tried to ask about the future, but it feels important to check in on these moments in the past. Especially now, Disney, in particular, has Josh D’Amaro, the new CEO, saying Disney Plus is going to be an everything app, and that very much implies user-generated content in a totally different way. It feels like those things would have been aligned. The other moment I want to ask about is when you went public in 2021, which was a SPAC. It was very fashionable at the time. Do you feel like that was a miss? Do you think that was the right decision at that time?

Well, I think the biggest issue was that we missed the window. Again, if you had perfect hindsight, if you could do it over again, I think we would have gone public via a SPAC, and we would have done it without buying Complex, and we would have just done it right away. Because the market was super hot, we could have gotten a lot of capital into the business. We didn’t need Complex. We were bigger with Complex, but our business was probably better. Our numbers were probably better without Complex than with. So we could have just done the deal more quickly and had more cash on the balance sheet. And then we would have been able to have a lot more leeway and latitude to reinvent the company when the market got really tough.

The problem is, when we started to explore this Complex acquisition, it slowed down our ability to complete the SPAC transaction because we needed to agree on a sale and agree on all these terms and negotiate with Hearst and Verizon, who co-owned it. And then we needed to, once we went public, figure out how to integrate and work with cultures that were not really well suited to each other. The Complex culture was just very different from the BuzzFeed culture.

So in retrospect, I think hitting the SPAC market while it was hot, bringing in a bunch of capital, doing it with just BuzzFeed, and then using that to reinvent the company as a public company, we would have been in a really exciting position, and we would have had capital to do acquisitions and things like that. Instead, we missed that window. The SPAC market got ice-cold. We were already kind of far along on the Complex deal, and so we ended up with debt instead of cash. So the debt to buy Complex basically has been a burden that we’ve slowly been able to alleviate, first by selling Complex, then by selling First We Feast, and now with this partnership with Byron Allen and that investment. But it’s taken a bunch of time to kind of get back to where we have the balance sheet we need to innovate and operate in a way that’s more confident and more deliberate.

Why did you buy Complex if you didn’t think it was the right choice?

At the time, the strategy was, “Let’s consolidate the digital media industry. Let’s give advertisers something for every demographic.” We were stronger with female audiences, and they were stronger with male audiences, and they had a lot of great properties, and so we thought this would be a one-stop shop for advertisers.

What ended up happening was that both the value of digital media and the traffic that we’ve been talking about were challenging. But in addition, advertisers started to spend more and more with the platforms directly, and the programmatic buying became a bigger and bigger thing. Complex just doesn’t have that much distribution. They were premium but smaller audiences. And when more shifted to programmatic, that really helped BuzzFeed and HuffPost, but it meant that Complex was really dependent on this direct sales channel, which was becoming weaker for all digital media companies.

So I think that things didn’t go the way we had hoped with Complex, but we were excited at the time, and that’s partly why. And then it was one of those things where we’re like, “Okay, we now have to take on this debt to buy them, but if things go well, the debt won’t be an issue.” But when things were tougher than expected, the debt burden was really challenging. 

It’s one of the reasons why I think people don’t realize that the underlying BuzzFeed assets are better than people think. Because they look at the whole thing and it’s like, “But you start backing out pre-COVID real estate, and you start backing out the debt burden, and you start backing out some restructuring costs.” You kind of look at it like, what is the actual business and the pieces of that? I think it’s doing better than people think.

I’m glad that Byron was able to see that, and some others were able to see that. So we had an opportunity to do a deal that makes a big difference and allows us to unlock that value and not have it just be kind of captured by our current structure.

BuzzFeed’s greatest strength, I’ll call it the millennial era of digital media, was that it was a cultural phenomenon in millennial digital media. I think that’s the best way to describe it. There were all those posts about Why I Left BuzzFeed, but before that, I will never forget the letter to the editor at The Awl, where it was just the poor digital journalist who was like, “I don’t work at BuzzFeed, and it kills me every day.” This thing rocketed around, and I think Choire Sicha wrote back to them. It was like a great digital media moment that I will never forget. It had totemic power in the media ecosystem at that time. There’s a reason Disney wanted to buy it.

You’re talking a lot about AI. I think my last question for you is, how do you get back to that cultural relevance when I look at the polling from Gen Z in particular, and they dislike AI? If AI is the thing that’s going to accelerate you back into relevance, but the younger demographic dislikes it so much, how do you connect those dots and solve that problem?

I think that a lot of what we do does not read or appear as AI. I think people don’t like certain things about AI. If you just ask them in general, “What do you think of AI?” There are a lot of pretty negative responses. I think people love playing Word Chain, and I don’t think they mind that AI coding tools were used in creating Word Chain. The human agency and human intention are in the product. The person who designed the game is obsessed with making a really fun game. People playing it really like it. The creator of the game is in the comments, talking to them, and they’re like, “When are you going to make the next one? I love this.” So I just think that saying, “Do you love AI?” is kind of like saying back in the day, “Do you love mobile?” 

But a lot of people would say yes. My career is based on “Do you love mobile?” The answer was millions of people being like, “Yeah, I want a whole website about that.”

Yeah. I mean, there were industry people who liked mobile, but I think most people liked the things they could do with it, and they liked the way that they used it. It’s like, “Do you like microcomputers?” It’s like, “Well, no, I like that I can do spreadsheets now, and I don’t have to have…” So I think if you think of AI as a computing platform, the question is, what are the apps built on top of that computing platform that Gen Z will like? And there are already many of them. Whether it’s using them to do their homework, helping solve problems that they have, or personal challenges they have, and talking to ChatGPT or Claude or whatever, I think there is a lot of love for AI and a lot of hate for AI. I think we just live in a world where everything is super polarized, and everything is kind of love and hate.

Fighting for a new way of using AI and a new way of thinking of AI that puts people first, that is creative, magnifies people’s agency, is not about feeding everyone the same slop, but is about helping people have a more personalized and connected experience, and connect with their friends — I think that’s going to be something that you will have to fight for because there’ll be all kinds of dystopian uses of AI. But I think we can build something that is special, that makes people see the power of this new computing platform, but not through the platform itself; it will be through the applications and through the fun things that they can do with it, and the fun ways they can connect with their friends.

I’m excited to see how that all plays out. When does the deal close? What’s the timeline?

Hopefully, we’ll get it closed by the end of the month.

All right. And then you and Byron are going to have to come back, and you have to show me this AI-powered YouTube that’s going to make Google shake in its boots.

Yeah, you’ll have more fun talking to him than me.

I’ve talked to him before. He is quite a character. I’m honestly curious. You’ve described some big products and some big visions. I’m eager to see them come to fruition. You’re going to have to come back soon.

Jonah, thank you so much for being on Decoder.

Thanks for having me.

_{Questions or comments? Hit us up at decoder@theverge.com. We really do read every email!}