Reddit CEO Steve Huffman on surviving the AI slop-pocalypse

Support for this show comes from Odoo. Running a business is hard enough, so why make it harder With a dozen different apps that don't talk to each other? Introducing Odoo. It's the only business software you'll ever need. It's an all in one, fully integrated platform that makes your work easier. CRM, accounting, inventory, E commerce, and more. And the best part? Odoo replaces multiple expensive platforms for a fraction of the cost. That's why over thousands of businesses have made the switch. So why not you try Odoo for free@odoo.com that's o d o o.com

0:00

this

0:41

episode is brought to you by Indeed. Stop waiting around for the perfect candidate. Instead, use Indeed sponsored Jobs to find the right people with the right skills fast. It's a simple way to make sure your listing is the first candidate. C According to Indeed data, sponsored jobs have four times more applicants than non sponsored jobs. So go build your dream team today with Indeed. Get a $75 sponsored job credit@ Indeed.com podcast. Terms and conditions apply.

0:42

Can you defend what you're talking about from AI, Though? I imagine that is a big threat.

1:13

Look, before there was AI slop, there was slop.

1:17

There was human slop.

1:19

Yeah, human slop. And Reddit's got plenty of it. What the Reddit platform is good at is pulling the content out of the slop. The AI version of this problem. It's just an extension of the same challenge we've been facing for 20 years.

1:20

It's access time.

1:39

Is that what we say now?

1:40

I don't know.

1:42

Was that premeditated, Alex?

1:43

It was not premeditated, obviously.

1:45

It just came right out. Huh?

1:47

It would have been a lot better if it had been.

1:48

Well, let's roll with that.

1:51

How are you doing?

1:53

I'm good. This is the first day in a while I haven't dramatically overbooked myself.

1:53

Great.

1:59

So I could actually serve myself up some watermelon hi Chews as a reward for making it to this point and actually read my emails for once. And your newsletter. Can you tell what I have and haven't read your newsletter in the lead up to the episode?

2:00

For sure. But I'm surprised you're not already like Ivan, our guest last week and just consuming email through a chatbot.

2:19

Not yet, man. Notion is a nice company or it's just being back in media again and everybody treats me nice. But I'm currently on a slack room with one of their engineers working through some bugs with custom agents which as a product nerd Makes me feel very, very special. And I do miss that. I mean, I remember when I left the Verge, I was like, are people still going to want to talk to me? And the answer was no.

2:25

But now you're back in media, baby.

2:54

Why is everybody so nice to me? Oh, yeah. I mean, it was always just very ironic to me, like, how nice people were to me given how little I was paid. I'm like, there is no one making $40,000 a year writing media that should have this much influence with that type of compensation.

2:56

It is the only job in the world where you simultaneously make so much less than everyone else around you. You also do not ask the other people around you for money. It's a very unique job in that sense. Like, I'll often be in the room with whatever, you know, masters of the universe type folks, and it's like, oh, literally everyone here just wants each other's money. And even though I'm the brokest of all of them, like, they still want to talk to me because I'm not just with my hand out, you know?

3:15

Please, sir, may I have some dollar pizza? That was me in New York City in 2011.

3:45

Yeah. I mean, I will caveat that unless we're talking about how premium brands can align with the Access podcast, then my hand is certainly out. But otherwise, yeah, man, it's good to have you back in the media world. And we did get great feedback on the episode with Ivan, CEO of Notion, last week. Please check that out. If you're new to the show and leave us reviews, please help us with those. Those podcast charts and watch us on YouTube. I feel like I have to be so relentlessly self promotional now that it kind of kills me inside. But I guess that's the name of the game, right?

3:53

Indeed. And getting into this week's news of interest, it's honestly, it's tough to tell. I mean, there's this post by Citrini, research about the future of AI that's been going viral.

4:28

Citrini.

4:39

There we go.

4:42

I'm rubbing off on you. But it's honestly, like, tough to know because it is such a monoculture within the X algorithm, where it's like, do you like Knights of the Seven Kingdoms? I shall show you the same gif 18 times. And not just over the course of today, but in a row because you like it that much. And then I saw 100 instances of people talking about this piece, which on the one hand can be helpful to see what everybody's saying, but on the other, Twitter just is not A way to find out what is actually happening anymore. With the exception of one topic per

4:43

day, I think it has broken out from the echo chamber. It's got 20 million views on X in less than a day. The substack post is going mega VI, as they say. Cintrini research.com the title of this is the 2028 Global Intelligence Crisis. By the time this episode comes out, you'll have definitely heard, uh, it also moves stocks. DoorDash stonk went down. A bunch of other stocks went down. And the entire premise of this essay, it's not like they have some crazy new insight, right? That's like we found out that these companies are going to zero because of AI. It's purely a. I don't know what you would describe this. You're more of a writer, Ellis, but it's. It's an imagined world in 2028 kind of writing as if you know you're able. You're peering into the future where this financial firm is explaining the global economic fallout of agentic AI that has accelerated on the trajectory it's on right now.

5:19

I think what resonated with everybody is just playing out the scenario of our current system where a lot of people feel like they are already getting squeezed either by corporations or capitalism at large, or just the United States being 300 years old and starting to show its cracks. You could call it inshidification, you could call it extraction. There are a lot of different ways to talk about this, but I think what this piece is showing is that AI is just really only going to accelerate that. And I think one example it talks about is kind of like the entire B2B SaaS sector and technology essentially existing as a way to sell tools to people who can't build them for themselves. What happens when people can build these tools for themselves? The entire sector kind of collapses. And I don't know, man. I was getting bit spooked reading this.

6:18

You sold your portfolio?

7:14

No, not yet, man. I'm on all those Vanguard 2055s, assuming the world still exists.

7:15

Long, baby. Yeah, I mean, I was thinking a lot about actually what Ivan was telling us last week when I was reading this. I think his quote, which has gotten a little bit of pickup and I think other founders I've talked to have gone like, oh wow, he, you know, he articulated that well, was if your product is not set up to be traversed by agents, you're going to be in for a really hard time. Was I think, almost exactly how he put it. And so how notion is opening up as a developer platform and as a way for agents to interact with its data ahead of even an ipo. He knows that this is the eventual end state of companies like notion software businesses. And I think Dylan Field at Figma honestly has this very similar view. I would say two of the more plugged in product minded still, you know, young founder SaaS CEOs that I think are, you know, are understanding the cutting edge saying similar things. And so I think that's advice that's probably worth taking to heart and digesting a bit, especially with an essay like this.

7:21

Well, you like my shower thoughts? Thank you for humoring them. I had a good one this morning. Well, a scary one. I was just thinking what percent of companies, setting aside the founder who sets the direction, what percent of actual white collar work, is trying to come up with strategic solutions to problems and just simply getting this the team aligned around them using best practices and with people without usually a source of truth that people refer to or even remember on any regular basis. It seems to me that a lot of that stuff can be automated by agents that know the best practices that come up with their strategic point of view and can execute it dramatically faster. And I think it's just scary, man. I mean, I think what it leaves and I've had a handful of clients in the creative space trying to talk about like AI and creativity and I've come to believe this idea myself is that one of the things that will be the scarcest will be the outcomes of our own unique lived experiences, whether we turn them into art or products or fashion or services or games. And that's something that like AI almost by definition cannot do, especially when people so genuinely appreciate the story, the personal story behind something. But yeah, man, I mean, this post was all about just like white collar collapse. And a lot of the logic of it seemed kind of sound to me.

8:23

Yeah.

9:53

For a lot of these jobs that just largely boil down to like best practices, execution and coordination.

9:54

Yeah.

10:00

Or does that sound crazy?

10:00

No, I think that's totally right. There was my favorite line was it should have been clear all along that a single GPU cluster in North Dakota generating the output previously attributed to 10,000 white collar workers in midtown Manhattan is more economic pandemic than economic panacea. It's like. Yeah. And also it's just this very, I guess, basic human insight that humans spend a lot on discretionary goods and things that are not quote unquote logical or necessary. And guess what? Robots have zero need for discretionary spending. So what happens to the rest of the economy that is so reliant on up upper middle class, white collar quote unquote knowledge work workers spending on all these things. If their wages plummet because they're out of, you know, work and I, you know, doesn't give a shit about your vacation business, what is that going to do? The ripple effects just seem pretty extreme if you buy the logic of this article.

10:01

I mean we certainly have not felt the impact of agents, even if they are just B or B plus agents working, not just faster. Traditionally they don't need a lunch break to go to the bodega, I guess you said last episode, Alex, you only take a five minute lunch break, so you're still at pace with the agents here. However, you still have to sleep, don't you?

11:01

I go to sleep.

11:22

But you know, you don't have that edge anymore.

11:23

Well, yeah, I mean there's an argument to like, you know, If I have 10 agents at all times or infinite number that can do things for me in parallel, that is going to replace jobs. There's just no way it won't because that's activity that would have either been done by me or I would have hired out to someone else, an assistant or whatever. And if these agents actually work like that, which I don't think right now, they're reliable enough, shout out to the the meta AI executive who accidentally had OpenClaw delete her entire work inbox. That was another viral X post the other day that I saw. So these things aren't quite reliable enough. But you know, when they get there, and it seems like they will at the rate they're going, yeah, man, people are totally going to be out of work.

11:26

One of the quotes I liked from this piece. The economy can find a new equilibrium. Getting there is one of the few tasks left that only humans can do. And then of course the kicker at the end saying it's not too late. Quote, the canary is still alive. Chills.

12:09

Good writing.

12:29

It's weird though, dude. I mean, another quote from this that said every decision was rational but altogether catastrophic. I mean, I work with hundreds of companies and they're all just kind of bright eyed young people trying to say, hey, I worked in this industry and it fricking sucked and I'm going to have AI make it better. But as soon as all this stuff becomes digital, yes, it allows for hopefully more pay, more efficiency. But then down the road, when everything gets marketized, turned into a market where anything can be bought and sold, that's kind of the final outcome and then it bites people back in the butt. And it seems that AI is just going to kind of take that to the next level. And few seem to have the ability, with the exception of maybe Dario and this piece, to actually play some of this stuff out. Have you seen others who've like, written kind of vision Y posts with specificity?

12:30

No. And I think what it showed me is that it is really worth sitting with the potential ramifications of the things we understand. Even just you and I, we're not like experts on this stuff by any means, but I think we do have an unusual level of access, no pun intended, to the people building this stuff. And I am so just kind of tunnel vision on what I do day to day. I don't know how you feel about this, but I don't really spend a lot of time actually abstracting out the implications of what I'm seeing up close, you know, a year from now.

13:26

Yeah. You don't have kids yet.

13:59

Well, yes, that. There's that. But just even the technology, I'm like, selfishly, it's like, damn, like, we could probably make some money if we actually thought this through, right?

14:01

Oh, yeah. I mean, we're certainly the lucky ones who have some scarce asset, whether it is, you know, experience or connections that we can find more ways to monetize in a way that's good for us and our families, but for the vast majority of people is just quite a different equation, I think.

14:11

Anything else going on for you this week? I was going to go to New York and I canceled that trip due to blizzard. And I'm very glad that I did because the weather is glorious this week, as it always should be in la. But, yeah, Anything else top of mind for you, man.

14:35

I'm just glad I'm not dealing with the New York weather hysteria. I found that just so grating

14:50

every

15:00

other week in New York City. I guess that was the old Twitter days as well.

15:00

It's kind of fun. I kind of miss, like, there are a couple blizzards. I remember when I lived there and yeah, it's insane and intense, but it's definitely a memory, you know?

15:04

Well, that's because you're a true reporter, whereas I am not. I think media people like to catastrophize. Yeah, I just like to yap.

15:12

Otherwise, I'm speaking of yapping. I'll be at the upfront summit in LA this week, yapping away with all the VCs and tech people and talking to some potential future guests for access.

15:23

What's on your Yap gender.

15:35

My yap gender. I'm chasing some scoops for sources. I can't reveal the moment, but I think one will be out, one is in the chamber. I was actually on TBPN last Friday talking about some of this. Thanks shout out to John and Jordi for having me on at the Ultradome. We were talking about the art of scooping and as I said then, I've always got something in the scoop chamber. It's just a matter of timing, you

15:36

know, have you ever double scooped? Take my source to Jenny's and gotten the scoop while getting the scoop?

16:02

Yeah. Or you, you just do a scoop at a time, but the cone is full and it's like, you know, I've got a, you know, we can keep this, this metaphor going. But yeah, I'm kind of in that position right now, which is just a rare treat. You know, it's like when you get the triple stacked cone. It's like, man, I really am treating myself. I guess at this point, since we haven't already, we should say who our guest is. This week we have Steve Huffman, the CEO and co founder of Reddit. Reddit is obviously a very important company. Ellis is a DAU and I am a probably weekly user. But I think Reddit has played a big part in both of our Internet lives over the last 15 or so years. I've definitely fixed a router or two through it, I think, as we find out in the interview, maybe thanks to you, without knowing that. And I like Steve, I've known Steve for a while. You'll hear in the interview. We were just at a dinner together a few weeks ago and Steve is always good at actually sharing his opinion, which is increasingly rare as a public company tech CEO and not being afraid to give his take. And I think he did that with us. But yeah, I really appreciated him coming on. What do you think of Steve?

16:08

Yeah, well, I think what's interesting and kind of lovable to me is that they've just kind of found this niche. I mean, they're making a lot of money, they're a public company now, but they are by no means a 3 billion user company like some of the others out there. And I think for a long time people have had this idea that as a social or community platform, you either go boom or you go bust. And it seems like they've carved out a really nice place for themselves where they're making good money and able to actually, certainly in part thanks to having the founder still there, keep some ethics and morals going of some kind, even if you don't agree with them as they go about their daily lives making this thing. So yeah, it was fun to dig into that and more.

17:22

And a CEO who appreciates shitposting, which is also increasingly rare these days. Gotta love it.

18:07

I don't know if we want to encourage that because people think they're good at it and then they're not, and then it's.

18:12

It is an art form. Yeah, it's an art form.

18:17

Don't do it unless you're really sure you're good.

18:20

All right, we'll kick it to Spaz. How about that?

18:24

Support for this show comes from Odoo. Running a business is hard enough, so why make it harder with a dozen different apps that don't talk to each other. Introducing Odoo. It's the only business software you'll ever need. It's an all in one, fully integrated platform that makes your work easier. CRM, accounting, inventory, E commerce, and more. And the best part, Odoo replaces multiple expensive platforms for a fraction of the cost. That's why over thousands of businesses have made the switch. So why not you try Odoo for free@odoo.com that's o d o o dot com.

18:32

This is advertiser content brought to you by Stonyfield Organic. Our cows, them going out to pasture, they love it. They're so excited to go out every day. They wait right at the door and in fact we milk them and we just open up the laneway and let them just go right out to pasture. I'm Rhonda Miller Goodrich and I'm a dairy farmer in Cabot, Vermont. Our farm is Molly Brook Farm. We're an organic dairy farm and we are a supplier to Stonyfield. Molly Brook Farm has been in my husband's family since 1835. We started our organic transition in 2015. We had 53 acres of corn ground and of course we had to use herbicides and pesticides and the soil was dead. Really. For all intents purposes, we stopped growing corn and stopped using herbicides and pesticides and we seeded that down to perennial grasses. After that we begin to see biodiversity in that soil again. To be organic certified, our cows need to be in pasture at least 120 days. I think the organic practices really benefit our animals. You know, having good feed, good water, a nice light area, that's what's important to us and that's what's important to Stonyfield. Visit stonyfield.com to find Stonyfield Organic Yogurt near you.

19:12

Hey Steve, how you doing? Good, how are you?

20:50

Doing great, doing great.

20:52

What's, what's been top of mind for you since we last saw each other? It was just a few weeks ago at that dinner you guys had, you guys had earnings.

20:54

Do we have earnings? Oh yeah.

21:01

It was the evening before earnings. And you know what, this is actually a great place to start. I've done this long enough to where I can tell how the management team feels about earnings based on the vibe of like the press event. Right before you guys were feeling pretty good.

21:02

I was going to say the quality of the appetizers are usually a reflection of how earnings were.

21:20

Could tell you guys were feeling okay about the quarter and it was a good quarter.

21:26

Yeah, it was a good quarter. Think I'm happy with the results. Honestly, almost every earnings has felt the same. And honestly, almost every one of my board meetings has felt the same. I'd say like 90% of them are feel good about the work we did. There's always more to do.

21:29

So that sounds like neither an upvote or a down vote on board meeting, just kind of neutral.

21:46

I love, actually love my board meetings, love my board meetings. It's just that, you know, our mentality is always, I think it's almost always the same. We did a lot, I'd like to do more. We got a bunch of stuff to do. It's pretty much the same pattern.

21:52

A non stop treadmill.

22:08

It's better than a treadmill. I'm running on treadmill torture. This, this is actually enjoy this.

22:10

You've been at this for a while. I mean 20 ish. What plus years. Do you feel like you have another 20 in you?

22:14

I don't have any other plans. It's hard for me to imagine 20 years in the future, especially these days. But that has nothing to do with Reddit. During my time at Reddit I had 5 years not at Reddit.

22:19

Right.

22:35

So I was with Reddit for the first five years, then I was gone for about five. I've been back for the last 10. So I've had a taste of life not working at Reddit and I prefer working at Reddit. It's a cool job. We get to do a lot of interesting things and you get to do

22:35

press dinners with Alex, which, what does that even entail?

22:50

You know, just kibitzing about our industry. It's not the worst way to spend an evening.

22:54

It was great. Yeah. Also you said something wild about prediction markets. Can you repeat it here on the pod?

23:00

I was wondering what. Okay, what did I say, you think

23:06

the CIA is running prediction markets.

23:09

Did I say that? Doesn't sound like something I'd say sounds like some crazy ass Redditor.

23:12

We were talking about it and you know, it's a thing that it's a kind of conversation I'm having a lot with founders. Everyone is so fascinated by prediction markets and I just interviewed Chris Best at Substack about their integration with polymarket. I'd be curious to actually know. Yeah, how is Reddit thinking about that space? Is that something you guys want to do partnerships with or how do you feel about it?

23:19

Yeah, it's interesting. Before prediction markets, you know, polymarket and Cal Sheet, you know, existed, we thought of Reddit as a prediction market. And I still think of it this way.

23:38

Right.

23:52

You can kind of get a sense of where things are going by looking at Reddit now. Sometimes it's, it's, it's, it's more direct and other times I think it's, you know, interpreting, you know, reading tea leaves a bit. I don't know. To me it feels like we're still figuring out where they fit in society. They are in many cases, I think, better at reporting the news than the news. The best example, right, is like the presidential elections. If you just want to know where things are trending, it's so much faster to just look at the graph than to read some, you know, cope from a reporter on election night. But there's other areas where, you know, things are more complicated. Like sports is one. I guess any area where the better has agency seems to be problematic.

23:53

Or there's the specific, like Maduro gets captured at X time and you know, maybe, maybe the CIA, you know, slips an informant, you know, a little bit of the take on a prediction market and maybe you've got a weird thing happening there.

24:40

There's so many opportunities for insider behavior. Yeah, we've seen this in the prediction markets, we've seen it in crypto and we can come up with less conspiratorial examples.

24:55

Right.

25:09

Let's say somebody's thinking about running for governor in California and so they can just go place a bet on will they run for Gala Vener in California or that super bowl producer or whoever that was who keeps winning bets, the first song is going to be that definitely happened. And if you're telling me like college basketball players aren't missing free throws to make a few bucks, like that's got to be happening.

25:09

Well, it strikes me as interesting though. I feel like, Steve, you're the leader of one of these communities that has probably done the most independent research of any other community. I feel like Reddit was always the place where if you aren't checking consumer Reports or where someone is going to post a leak or where somebody with a million karma is going to share the inside story on something. I feel like it's always been a great place for that. And I guess when we talked to the CEO of Kalshi, he was just talking about this as a way for people to essentially turn that research into something that they can actually make money off of.

25:31

The positive way to describe these things is it's people putting their money where their mouth is.

26:04

Right.

26:10

And so there's a lot of valuable information there. So I mean, if there is separation between the, I don't know what they call them, they probably don't call them betters, but I don't know if you can really call them investors. So I'm going to call them a better. So if there's some separation from the better and the outcome, I think it's really, really powerful. It's when the better as agency over the outcome that it's corrupt or not. I mean, is it, I mean, I think it is, but I mean, but the thing is the platform's revealing the information, right? You can see where the money is. Yeah, like where the money is is actually most likely to happen probably more than any other polling system.

26:11

Yeah, I mean insider trading is a huge thing. And also how disputes get resolved is a big issue and how, you know, if it's not as clear cut as yes or no, what happens. Those two things I don't think these markets have figured out at scale. So it sounds like you're more of like a wait and see on if that makes sense in any kind of product layer for Reddit.

26:51

I think the way we'd approach anything like this, and this is the way we thought about it, is let the users build it. So we have the developer platform on Reddit which lets you build, you know, basically apps within Reddit. Today it's mostly games, plus some moderation tools. But like one of the first ones that was built was polls. And so prediction markets are like super polls, but we don't, we don't have money flowing through it and, and that's, I mean, so that's where these prediction markets need to be. You know, they have to get, they have to get regulated or I, I don't know, skirt the law or whatever. That's, we're much, much more conservative in, in, in those things. Like we don't want to be tied up in those conversations. But I think there's so many interesting, interesting things that can be done on Reddit because Reddit provides distribution and lots of people with lots of viewpoints. That's why we built the developer platform, is to let users do interesting things where there's lots of users. But like the impetus for developer platform. I mean, one of the big ones at least was like Our Place, which was the April Fool's game. We've run it a couple times now, where a bunch of users create a work of art, in my opinion, at the same time collaborative. And so, but anyway, things like where you need lots of people participating at once, that's what Reddit does. It brings lots of people participating in the same area. So you could do predictions, but in a sense all of Reddit's predictions.

27:11

I'd be a little sad as a longtime Reddit dau representing in the room here, if it got too overtly like prediction marketing or monetization. Because I feel like Reddit is one of the only remaining places on the Internet where there's some altruism with why people post. And I feel like that's part of the mistake that every other platform has made in terms of community building, is that they've made everything so likable, monetizable. Everybody becomes their own brand. I mean, obviously it's not fully anonymous, it's about pseudonyms, people get cred within their subs and whatnot. But I really enjoy that aspect and so I'd be a little scared if it got a little too commercial.

28:42

Like one of the, I think important things about Reddit is if you're good at Reddit, it doesn't help you in the real world. You're just good at Reddit. You got a bunch of Reddit karma that you can't do anything with. Nobody gets famous or rich in the real world for being, you know, good at Reddit. And that, in hindsight, I think, is a really important quality of Reddit, which is you're just there because you enjoy doing it or you want to do it, or it's fun or helpful or whatever. So we'll leave the influencers to the other platforms. We don't ever want that. But there's a line, and this is one we're trying to figure out how to walk, which is around creators. And so we spend time thinking about what's the difference between a creator and an influencer. Because Reddit loves original content, but then they Reddit doesn't like self Promotion. And so I, but I always ask like, okay, well, where do you think original content comes from?

29:25

I think they can be the same thing. But influencer, if you want to take it literally, I think maybe you're trying to influence people over things that have already been created, whereas a creator is literally creating from net zero. And maybe the influencers come in around that content or amplify it or maybe that's a way to think about it.

30:24

I guess influencers, maybe their main product was themselves, maybe disproportionately versus creators, I don't know.

30:42

Yeah, I think those are both helpful ways of articulating it. I mean, creators create and so there should be some artifact, right? I made some art, I wrote this thing, I made this video. Now obviously influencers do that as well, but that's, that's, that's good behavior, right? I think if you're contributing, if you're creating content that people like, that's, that's what people want. People want the content. But then the line gets fuzzy, right? You could be an artist who does like a brand, brand promotions, it gets fuzzy pretty quick.

30:49

Well, this is the thing you've talked about a lot recently too is Reddit has this just very strong anti promotional culture, right? And I think publishers have felt this over the years. We were talking about this at dinner. Like, I think publishers are scared of Reddit because they've experimented before and then they just get, you know, absolutely destroyed in the comments for trying to promote their work. And it's just this instinctual kind of anti brand, anti promotional part of Reddit that you have to figure out with that, right?

31:20

And at the same time, a huge amount of content on Reddit is links to publisher content. Like at one point when we started Reddit, that was 100% of the content. So the irony, if you're a publication or you're a journalist or an author of something and you write something that the Reddit community will like and you submit it yourself, you'll get chased out of town, but if somebody else submits it on your behalf, it's welcomed. And then if the author shows up and like talks about it, that's celebrated. It's like, oh my gosh, the author's here. And so it's really important that we, for many reasons, land the plane. Like, figure out how do you get the content, whether it's professional publisher content or artwork or whatever in Reddit, hopefully by the creator, giving the creator credit, letting them sit there and enjoy the conversation with their fans or the consumers in A way that feels great, but there's really, there's a lot of emphasis on like, feels great.

31:51

Hashtag authentic Steve.

32:53

Right. But a lot of Reddit is feelings in that way.

32:55

Yeah, Well, I mean, so Ellis and I both worked at the kind of early, I would say Web 2.0 publishers, BI and the Verge, especially in the kind of mid-2000s where, you know, the chart bead would be up in the office and you know, we would randomly get, you know, a big subreddit would link to one of our stories. And it was just, I don't know if you remember this else, but it was just this like, yeah, Reddit bump, the Reddit bomb. I forget what we called it. It was just like, you know, miracle from on high, just fire hose of eyeballs. But it was so random and so unpredictable because we weren't promoting it ourselves. And it would be like way even earlier than that. I remember when Delicious would link something and it was just like, oh my God, the server's almost coming down on this blog I work for because of that. You all still have that distribution. You still have those eyeballs. How can you fan that out and get creators to see that? I guess it sounds like a huge challenge.

33:01

So we're building a bunch of stuff here. You know, we call it Reddit Pro and within that it's Reddit for publishers, but basically a non community space for creators and professionals to put their content on Reddit and they make it easy for communities to pull it in, for multiple communities to pull it in. Because I think also one of the values that Reddit provides is you can see how particular piece of content is received differently in different communities. And that's one of the ideas of Reddit. Right? It's just these communities have different perspectives. So one might love it, one might dislike it, but just making that more intentional. I think the answer to the, to a lot of these questions on Reddit is often transparency and intentionality. So transparency, yes, this is professional content and intentionality. It was submitted over here, like in this like Reddit Pro area and then pulled in by the community. And everything's kind of labeled as such. And this is the creator. So like nothing sneaky is happening. So the devil's in the details. I think it's really important that we get them right because we want the web to exist, we want creators to thrive. Like, we love linking to things or embedding things. Like really what's important to Reddit is the conversation about the thing. It's like the thing and the conversation about the thing. So we're inching our way there. But the no self promotion rule is like one of the first rules on Reddit. Like, we had that rule probably before we had any other rule.

33:59

I mean, it's frankly amazing that you guys are still flying this flag 20 years later. I mean, as Alex alluded, we have watched as media, and then I was internal at Snapchat for several years. We have watched all of these platforms just give in more and more to the cult of performance and everything becomes media. And before you know it, there are no friends left on platforms like Instagram. And I think it's less than 20% of engagement now is friends. And everybody's just giving in to what's most engaging. And. But at the same time, that's also what's allowed them to become these unbelievable businesses. I'm curious how you feel like investors and advertisers these days are receiving the identity, you know what I mean? Because it has its, it has its trade offs.

35:21

It does. And we are, we are stubbornly principled. So in our history. So we predate social media, right? When we started Reddit, Facebook was like a year old. And for college kids, YouTube was maybe 6 months old.

36:06

And Snap was but a twinkle in Evan's eye.

36:23

No, Evan was like a twinkle in his mom's eye. This was so long ago. We watched all these platforms grow faster than us and I knew what was working. We're not doing those things. We're not building a social graph. We don't have friends on Reddit.

36:26

Really.

36:42

The real world identity thing is like a sacred line we don't want to cross. And so there was a time where we accepted just being smaller, like, we're not going to do that. Like maybe we're this corner of the Internet, this little corner of the Internet, you know, that's more authentic or altruistic or niche or nerdy or whatever it was, I think all applied at the time. And I remember saying, don't people know what these platforms are doing? Like, don't they care about their privacy? And I remember being sad. I was like, they just must not care. It turns out they did care. Like the users, the consumers care. They just didn't know what was happening. And so now when this question comes up, which is like, why aren't you more like social media? You know, I tell this story, but then I also remind like our company and investors, we're one of the few platforms that's still doing this after 20 years. And so, yeah, we took a path because We're Reddit and we made some trade offs that maybe you know, inhibited growth for a while, but we're also here because we're Reddit and we made those trade offs and there are thousands of other platforms that didn't make it here. And now we're big enough. A big difference between now and maybe whatever this is 15 years ago is I used to think we were destined to be small and now we're big enough that like we've got one of everybody on Reddit. At least we've checked the box or answer the question, can Reddit work for anybody? I used to just believe that, oh, Reddit doesn't work for everybody. Like it's just too text based, too something to not social media to work for everybody, but it actually can work for everybody. So we're going to keep doing it our own way and I believe we can grow as big as anybody doing it our way.

36:43

This brings me, I think, to the big can of worms that I think is maybe the biggest question facing you all and just the Internet more broadly is can you defend what you're talking about from AI though? Can you defend it from especially agents at scale, what people are doing with stuff like openclaw and how that naturally will scale over time? I think. How are you thinking about that? You know, the, the proliferation of slop, as they say. And you know, there's these stories now about how Reddit's battling AI slop and I imagine that is a big threat.

38:27

Look, before there was AI slop, there

38:59

was slope, there was human slop. Yeah, human slop.

39:00

And Reddit's got plenty of it. Like what the Reddit platform is good at is pulling the great content out of the slop. The AI version of this problem, it's just an extension of the same challenge we've been facing for 20 years, which is not all content's great and some of it's inauthentic, right? Spam or marketing materials or whatever, some other manipulation. So we've been fighting the manipulation battle for our entire existence and we'll continue doing that. And then through the Reddit voting system, Reddit's really great at surfacing good content. Now does somebody get the best of us from time to time?

39:03

For sure.

39:44

Like it's an arms race and so there's always a new front and so we'll always have to evolve. And so I can get into human verification and how we think about all of this. But I also think one thing that's really interesting is there's A community rejection, like completely non technical. When they suspect content of being AI slop. Right. When a Reddit post sounds like a LinkedIn post, people are like, oh, all right, thanks, bot. Is that a bot or is that a human using ChatGPT and sounding like a bot? More of the latter than the former, but both are issues. But the community, it's interesting. The community just is in the process of rejecting it outright, no matter what we do.

39:44

But you are going to label. I saw in the last earnings call you said you're going to start moving to bot verification and labeling.

40:32

So there's a couple of things that are really important to us. The premise on Reddit, the assumption on Reddit is everybody you're talking to is a person. Like that is important. And so if people perceive that to not be the case, that's a real problem. Perception problems are problems. So in cases where the account is suspicious, okay, there's a couple levels. There's one level which our users never see, which is if we caught the account, right? If it's a bot or a spammer, we just ban it. And there's a huge volume of content on Reddit that just gets removed. Probably like 95% of the content gets removed as spam, which is one of the reasons it's so hard to build something like Reddit today. Like just the. And this is even pre AI, right? This is not even with AI, just pre AI. 95% of content is spam to begin with. So we do that and then if there's stuff that's in the gray area, we're still putting more pressure on, right? Are you a human? Is there a human behind this account? What I'd really like to do is what we're calling ass and seat. So not only was there human at registration time, like is there human sitting in front of the computer right now?

40:39

Ass in seat. I like that. A new metric. Yeah, well, because on the next earnings

41:54

call, people are going to use agents. Like there's no way around it. People are going to use agents to control the computer. And so we want to make sure that there's a person directing that thing.

42:00

And if that's happening, it's okay.

42:09

Well, I don't think we can say no agents. So I think we'll have to kind of take this one step at a time. So I think having human verification is important. Having lightweight human verification or re verification, like face id, for example, or touch id, just as an aside, like the design of face ID or touch id, they're basically part of the passkey ecosystem which is built around human presence. So face ID a touch ID a yubikey. All those things require a human to physically do something. That's why they're so helpful. So doing. I think little checks like that periodically will be helpful to us. This is for like the gray area use cases where it's like we're not quite sure so you should assume it's a humans. We're going to ban everything that we believe is not a human. And if it's gray area then we'll do these little checks.

42:12

Predator Badlands now streaming on Hulu and Hulu on Disney.

43:12

Here you're not the predator, you're the prey.

43:17

Prey, prey, prey, prey, prey.

43:20

Critics are saying it's epic, stunning and breathtaking.

43:25

Many have come here, none have survived.

43:31

Badlands now streaming on Hulu and Hulu on Disney. Rated PG 13.

43:37

In the wake of the release of millions of documents related to the Jeffrey Epstein case, the rich and famous are finally feeling some pain. But even with corporate resignations here and with former Prince Andrew being arrested in the uk, the question remains. How did Jeffrey Epstein remain a thriving member of the elite for decades when everyone seemed to know what he was up to?

43:44

I don't think you could be friends

44:05

with Jeffrey Epstein, whose MO was obviously having sex with young girls, even as Trump said, on the younger side, and

44:07

not know his MO Untangling the Epstein conspiracy. That's this week on Today explained every weekday and now on Saturdays.

44:14

One of the things that I like most about it, both as a contributor and as a reader, is that it feels like there's actually a long tail that isn't just based on like what's most engaging or what's most popular. Like, I mean, you were talking about not getting anything out of it in the real world per se, but I do think that there is something special about like the self esteem that you can feel as someone who helped others. And it's just, it's bizarre to me that Reddit seems to be the only place on the Internet that it's working and thriving. Obviously you're not as big as the others, but like every so often I get an email that my Reddit post about my fucked up Nest WI fi spectrum modem configuration that was totally backwards for some reason has helped someone new.

44:31

Wait, that was you?

45:16

You're just.

45:20

I swear to God, I went through something like this a couple years ago

45:21

and I used a router. What a cute misconnection. You helped me fix my router.

45:25

You're saying Your nest Wi fi after a powder outage was not playing nice with your Spectrum modem. Yeah, dude, that honestly could have been my post.

45:30

That's incredible.

45:38

Anyway, I believe WI fi and I believe routers, and I believe modems are some of the worst technologies ever invented. And so that's all. I don't post much on Reddit, but. And I had to post that, I'm like, this is going to help people. It started slow, and over the course of years I get replies, and it's just so strange to me that this mechanic of the human mind of wanting to be helpful to people is very powerful. And it could have existed elsewhere, but people almost just kind of abandoned it in favor of the instant gratification of likes, comments, and engagement.

45:40

Okay, well, totally appreciate what you're saying. I mean, you're describing fulfillment and you're describing, like, one of the basic qualities of a community. Like, a community, not just on Reddit, community in the real world is a group of people who, among other things, have some care or affection for one another. And I think communities are a natural byproduct of people being together. And so Reddit is the digital version of that. So, yeah, it feels good to help other people. That. That's all of humanity, by the way. That's everything. Like, I always, like, I always look out the window at a city. It's like, why do cities exist? It's because people like to build things and make things nicer for other people. And so I think in that regard, Reddit is natural. It's just allowing humans to do naturally what they like to do. You know, share a few laughs, gossip and kibits and talk about things, help each other out, maybe ask for help if they need it. That's why even in the AI world, or any other version of the world, like, we've talked about this before, like Reddit versus social media and now Reddit versus AI. We always come back to these. This, like, core principle that people want to be connected to other people around shared passions and interests and help each other out and feel good doing. So. That's not going away. That's not going away in any version of the world.

46:16

But you're both discounting, like, a lot of. Yes, Ellis, a lot of people do just have that natural gratification and altruism, but a lot of people are also assholes and grifters and freeloaders.

47:38

Not true. Not true.

47:47

Okay.

47:50

Some people are. I think it's less than 1% of the population.

47:51

You feel like, as a whole, the Community has antibodies against assholes as a

47:55

whole, on the whole, in the same way that society in general does. And I just like, I know it's easy to be pessimistic about the world and our fellow humans, but you can go anywhere and talk to, I think, anybody. And as long as you don't talk about politics, you'll probably actually have a lot in common or a lot of fun, interesting things to talk about. And I bet you will find that about 1% of people are, you know, assholes, and the vast majority of people are just normal people. Now we start talking about politics, it gets a little different.

48:00

That's why we have a no politics rule on the show, Steve.

48:35

So great. That's why we're getting along so great. But I think the challenge is on social media, like it incentivizes people to behave in an unnatural way. Right? You can actually accrue power in the real world for talking about politics in an extreme way. Like that's, that's a problem in a

48:38

way that you would never say it to someone's face in a way you

49:02

probably don't even believe.

49:05

I mean, the platforms literally hand you a megaphone. If you were to be in the digital town square, they say, hey, would you like to yell here? Yell louder. Because people are gonna look at it

49:07

like you get rewarded for saying crazy things.

49:18

Yeah, well, I'm curious. I mean, you know, talking about fulfillment, I mean, it seems like you guys have had a lot of success just kind of bucking some of these best practice metrics that social platforms have used. I mean, do you have any metrics for long term fulfillment or esteem in any way?

49:20

I mean, some of it's intuition just having been using this platform for so long. But we can see there's interesting things. Looking back. The more time people spend on Reddit, the more subreddits they join. And it's just like, so people just kind of go deeper and deeper and broaden their interests and then the more subreddits they join, the more time they spend on Reddit. And so for us, a lot of it's like, how do we like the metrics we care about? How do we help people find on Reddit the things they care about? It's a really long tail of interests. And so it's kind of spreading people across more and more diverse subreddits or just more simply having diverse subreddits is some of the things that we think about. So we don't optimize for time on site. For example, we Measure it like we track it. But what we look at is like, how many subreddit subscriptions do you have? Like, are you actually finding things you enjoy quantifying? Fulfillment's a little harder, but I think we can simplify it to are you following subreds you like, Are your posts successful? Is a big one. We think about right now, like how many people write not just good posts, but good posts that don't get removed. Because that's one of our biggest problems. It's lots of good posts get removed. About half of posts get removed by auto moderator, which is this like ancient piece of technology. It's like this non AI barbaric piece of technology that removes posts by a lot of new users. Like, that's a huge problem.

49:36

Slap an LLM in that thing, Steve. It's time.

51:06

Easier said than done, but yeah, that's basically the plan.

51:09

Well, so talk about, I mean, the experience for new users these days. I mean, I imagine at one point, most people at this point who have apps and use the Internet have tried Reddit at one point or another. I think we all know it's hard to get people back once they've tried something. I mean, how is that new user journey these days? And how are they retaining?

51:11

Okay, so I'll give you my board and earnings answer. It's better, but we got a lot of work to do and that's the truth. So we've gone through a big transition at Reddit. We're really still in the middle of it.

51:30

By the way, we're having a heart to heart here as someone who worked at Snapchat for seven years where retention was always pesky. People don't get it. Why don't they get it?

51:43

Once they get it, we've got them. Everybody can get it. And so. Okay, so how do we do that faster and more efficiently? Well, Reddit, the product has changed. We used to be the front page of the Internet. You know, it started as a joke and then it was a title that we wanted to earn and then I felt like we earned it, but now that that was. So in other words, one front page for everybody. Our popular now it's what we now call our popular. It used to just be Reddit.com?

51:53

yeah, which you just said sucks. And you're getting rid of our popular

52:21

sucks because it's the lowest common denominator now. And as Reddit gets broader and more diverse and bigger, all good things, the lowest common denominator becomes lower and lower. So we call this regression to the Meme to the point where R popular is not actually relevant to anybody. It used to be this amazing magical experience when our user base was homogenous. Now it's probably off putting to most new users. So that's a huge change. We used to have one subreddit, right now we have, you know, many thousands. So the problem, the problem has changed along the way there. So how do we get you into subreddits relevant to you quickly? That's really the question.

52:25

What's the reaction been to getting rid of our popular?

53:09

I haven't heard anybody defend it. You know, if we had done this 10 years ago, I think there's a lot of Reddit users who just only had the front page, but truly it's, it's not working. So that'll be a big part of it. Now to have a, a more personalized front page, you have to know something about the user. And so that is where you know, how do you learn something about the user very quickly. And so there's a few assumptions we can make. You know, I'll give you something that'll probably, that's pretty easy and non controversial, like local. So what's the city you're in, what are the sports teams, what's going on, like that sort of stuff. Local is a piece of information, like we generally know kind of where you are, at least at a high level, like what country you're in or what state or maybe what city that's pretty safe and non controversial and relevant. So I think we can go heavier that direction.

53:12

I always feel like Reddit has been good for local and it's always been so underappreciated as the other platforms have kind of vacated this space. Is that something that's personally important to you, Steve, like creating spaces for people to talk about local news? Because it really, really bothers me and I feel like it's in some sense has destroyed the fabric of society that everybody in their one feed is seeing global news all day, every day. That's all ranked by engagement.

54:03

Yeah, and also it's probably irrelevant to like 99% of people's 99 percentile experience. Like the San Francisco subreddit, you know, really kind of started to come to its own during COVID So that's kind of my direct experience seeing this play out on Reddit. I think you just kind of need a critical mass and our recommendations have to be good enough to get it in front of you. But I think local is great, right? Communities. There's this general idea I think about the size and health of communities. And so in the real world, a community can only be so big and still feel like a community. And on Reddit we have this interesting quality where a big community still feels small. You can go to the largest subreddit, which on any day of the week might be Ask Reddit. And it's still. There's something about it that causes it to still feel small.

54:30

Right.

55:16

When you read a conversation that might be between 10 people talking to each other, right? User one to user two to user three, whatever, it actually sounds like two people going back and forth. And so that it's big, but it feels small. Quality, I think, is really important for maintaining that sense of community. But the nice thing about local is the more local you get, it just is small. And you start to have shared values and shared vernacular and an affection for one another.

55:17

All these shared favorite bagel shop.

55:45

Shared favorite bagel shop.

55:47

Always controversial here in la, or at

55:49

least strong opinions about these things. And so local is, I think, a gateway to relevance, as are shared experiences. So parenting is another big one. Like, who would have thought Reddit would be one of the best platforms for parents? Certainly not me when I was 21, when that was maybe more of a nightmare than an aspiration. But those shared experiences, I think are really powerful, creating that sense of belonging.

55:52

Speaking of the back and forth, you are, I would say, infamous. As for a textio shitposter, you recently I saw, posted that you love shitposting on Reddit. You made that known again recently. What have you had success with on the shitposting front recently? What alts are you using for that too? That's. Yeah, no, you're not doing that.

56:23

I don't use alts anymore.

56:44

John Baron.

56:45

No, there's no way. I just. I'm just too old to juggle multiple accounts. And then if I made a mistake, so. So I'm only. I'm only spez. There are no other alts. I'm trying to find this balance. Yes, I love shitposting. I'm also the CEO of a public company and if I say the wrong thing, it could be a problem for folks. So I'm really trying to get better at walking that line. But I think being present is important. I'm trying to get back in it, but it's been a journey, to be honest with you.

56:47

I was just gonna say. You can't dunk on the haters like you used to. Is that the problem?

57:18

I think there's an art to doing that and not punching down you Know, but it's. It is possible, but there's. I think some finesse is required.

57:24

I mean, I just think it's a good practice to be available and fallible and not a big dick as a CEO. Like, people are fine with people that make mistakes. You know, the problem is when you aren't available to be accountable to it or anything like that, especially for these social platforms. I mean, that's one of the things that's interested me about Reddit this entire time, is that no one expects the rules to be perfect for a community, but they do expect to be there to be some local accountability for it. And that's, in my view, what's been the problem the whole time with these global platforms is that they try and make rules for everybody and say it's impossible. People don't actually need that. They just want there to be local emissaries that they could work with.

57:35

I think there's a couple dimensions to this. Like, the first step is acknowledging the problem, whatever it is. People get really upset when they say, I think this is a problem. And then you say, actually, no, it's not. It doesn't exist. Right. So that's kind of the gaslighting behavior. People really hate that. If somebody says, I think this is a problem, and you say, yeah, I agree, or at the minimum, I agree this is happening and I think it's a problem, or I don't. And then you're talking about next steps, you're on much, much firmer footing. And so I think this is a mistake that leaders make a lot of is they just deny that there's an issue to talk about at all that gets people really riled up. But if you say, hey, I agree, and then you're arguing about solutions, you're on much firmer footing. Now, the local laws thing is an interesting one. As a matter of principle, we comply with local laws mostly. But there's a line, and we wrote this in our content policy, Right. There's also kind of values around free expression and free speech that we think are really important. And so it might be the law in some dictatorship to ban political dissent, so we don't do that. Drugs and porn aren't the free speech hill that I'm going to die on.

58:17

Right?

59:32

So again, some. Some consideration required in all of these things, but we're a global platform and we realize that, you know, cultures around the world are wildly different. But I also am not going to tell you I know all the answers

59:32

in all of these. And these pressures ebb and Flow. Right. I'd actually be curious to hear from you. Not even just in the U.S. but globally and in the U.S. i guess, especially poignant right now. How do you feel that is flowing right now? Do you feel like there's more pressure than ever on you to censor, or do you feel like it's not as bad as it's been in the past?

59:46

It comes and goes. I'd say if you ask my censorship barometer, it's lower than it's been in the past. But I think that the argument has changed. So, you know, during the. The 2020 era, there was an immense amount of pressure to censor content that went against the orthodoxy around. You know, Covid was an easy one, but just a lot of the political stuff, we tried to stay out of it. Right, we tried to stay out of it, which meant by and large, like, not censoring. Although there's a few decisions I'd like

1:00:09

to have back

1:00:46

these days, I'd say that the conversation is more nuanced. I mean, there are some bad laws out there about, you know, misinformation or this or that. The parental controls or child safety is a big one.

1:00:49

You're suing Australia over that? Yeah. They called you big tobacco.

1:01:02

Yeah, well, they made this under 16 band. We're totally for child safety. Now, Reddit was specifically mentioned in there. First of all, Reddit's not. It's not for kids. It's not relevant to kids. We don't market it to kids. We don't have a lot of young people on the platform. But we do pick people up, users up. We call them emerging adults. And so, you know, this might be late stage teenagers. These are people who are developing their opinions, including political opinions. And so for the government to say you can't consume this source of news, in fact, you can only consume sources of news that the government approves, is effectively what they're saying. That's a problem. Even if your parents disagree, by the way. Even like the Australia E. Safety board, they don't even like this law. And so I think there are better ways of achieving more safety online than the government choosing which sources of news are better than others. Keep in mind that not just Reddit, but social media broadly is for many people, the number one source of news and information, including political news and information. And these kids are barely kids. They'll be voting in like two years. And so you're basically saying this new group of voters can't get information from the most popular sources of information. They can only get it from these other sources. I don't think that's right. Even if their parents disagree, I don't think that's right. That has a completely different conversation from how do we protect kids online? And I think we spend so much time talking about what kids can't see and not keeping kids out of the adult areas and not nearly as much time talking about how to keep adults out of the kids areas. And that I think is much firmer footing, that separation of kids and let's call them like I separate kids from maybe like young people, late stage teenagers, emerging adults and actual adults. I think separation is really important and I think doing that at the OS level or the browser level would be much more effective. And I think parental controls.

1:01:06

You agree with Zuck on that. That's the thing that a lot of the platform companies are saying is that Apple and Google need to be doing the age verification.

1:03:18

It would be a lot better and it'd be a lot more safe and private than every. Every app re implement it using whatever third party source or not. Like a big part of Reddit is we don't know your identity. Like we don't want to know your identity. Right. Anonymity is safety. So we don't want to see your id. We don't want to have that exist in a database that could be hacked or leaked or, or subpoenaed. It's just the best logs are no logs. And so I think the OS level is a great place to do it because the phone already knows your birthday, it already knows what you look like, scans your face a thousand times a day. And the phones are far more secure than like they have secure chips in

1:03:24

them, hashtag secure enclave, but for real.

1:04:07

So I just think it's a much better starting point for, for us and everybody else to have it built into the phone.

1:04:11

Ellis, you agree with that? As a parent, you endorse that?

1:04:19

Definitely. I certainly haven't heard a better idea. I mean, I have been fully leaked a number of times now. I feel like I get those notices more like every other week at this point.

1:04:22

It's like it feels like it's experience marketing strategy. Oh, we just got breached. Here's your weekly reminder.

1:04:36

Shell company. Who owns the shell company of some place I got my car rental eight years ago has leaked my social again. Whoopsies.

1:04:41

This is actually a nice tie in to the next thing I want to talk about before you have to go, which is AI scraping and your AI deals. Speaking of people middlemen reselling data, something you've been dealing with and combating, especially in the run up to going public. And you've got the Google and OpenAI deal. I've actually been surprised that you haven't announced other big deals like that. And I know you're in some lawsuits with other players. But is the AI, you know, data business where you had hoped it would be by now? Had you hoped to have had other deals, or is this what you always kind of thought it would just be those two big ones?

1:04:52

I'd say it's probably neither. I'm glad we did those deals for a variety of reasons. I think it helped change the narrative about Reddit. I think there's a lot of concerns that, like, AI would kill Reddit, and it's actually no. Reddit fuels AI. I think the partnerships, the partnerships themselves are great. Having a formal partnership allows us to collaborate more on the integration. The integration is inevitable. That's part of my viewpoint on this. We are already deeply indexed by Google and then with OpenAI, it's a helpful way of navigating Reddit and it's a helpful way of discovering Reddit. So I'm glad we have those partnerships. But our thinking has also evolved and I guess you can kind of see it in the answer, which is the relationship started out as basically dollars for data. Like, we have a great business model. Ads is a great business model. Ads will scale on scale, on scale. And so important to us is how do we use these relationships to help make Reddit better? So specifically, how do we help, how do we use these relationships to help make it help people discover Reddit, to find what they're looking for on Reddit, to find their home on Reddit. And if we can do that, then we will help our partners build the best version of their products, which are very searchy. And I think we're seeing how that can go. And that's where these conversations are evolving. So there are other areas where it's maybe more transactional. We give our data away to researchers, so we're not that precious about it. We're actually, all things considered, very open. But our belief is commercial use should require commercial terms, and that allows us to do these things intentionally. What can the data be used for? What are we both getting out of this relationship? What is the data not used for? I think all of these things are important. And so in other conversations, for whatever reason, haven't come to terms, and I think that's okay. You know, we may get there yet, but I'd say the goalposts are moving as well, in terms of what our

1:05:26

expectations are, the goalposts moving from this is a big kind of lump sum in exchange for data to it's more dependent on the actual output and how the product shows up in the experience of who you're partnering with. And so it's more if Reddit's driving, you know, engagement to you, almost like a metered fee, like, I guess, does it just change the way these deals work at a high level?

1:07:35

I think all of these relationships are, you know, the companies are different, right? They have different needs, they're going to build different things. So it's like, okay, at the end of the day, are we helping Reddit advance its mission? If it's not, then what's the point? If you can help us do that, then we'll help you. Do you know, X, Y, Z? And most important, is it not hurting Reddit? Right. We don't want to. We want to help Reddit's flywheel go faster and not make it go slower. If we can't do that, it's not that interesting to us because we have a really healthy business model.

1:08:00

But is the cat out of the bag on all the other companies that have and continue to scrape? Like, that's never going to be fixed or addressed in the way you'd hoped? Like, as the world just moved on?

1:08:30

We will, we will ban and block and continue to advance our defenses. We have a few lawsuits, which is not my preferred way of resolving these things. It is a last resort, but we've gotten to the last resort with a couple of folks and we'll try to come to a reasonable agreement with. Everybody is open to having a reasonable agreement.

1:08:41

What about Reddit buying other companies or scale? Do you think you have enough scale where you're at to continue on in the way you want to continue on, or do you think there's opportunities, especially with this AI freak out, that the markets are having to combine and join forces and maybe maybe link up with other platforms and get more scale?

1:09:00

I'd say most of our focus is internal, right? We have, like, we have the content for everybody. I think we have a unique, differentiated experience. We fulfill a basic human need. And to Ellis's point, we have a lot of people try out Reddit. We have a lot of shots on goal. And so my basically obsession is how do we make those folks who are interested in Reddit, took the time to download the app, have a great experience? Acquisitions are an opportunity for us. We do have a lot of cash. This is a great business. It Throws off a lot of cash. But we're not really looking at how do we use cash to grow. It's more of how do we use acquisitions to bring in great talent or technology that we need. We're not really in the mindset of using our capital to acquire more users at this time. It can be done, and it's been famously done successfully by some of the companies we aspire to. But at the moment, I think it's more of how do we advance Reddit's mission. So first, invest in our business. Second, possible M& A.

1:09:24

Speaking of cash, last question and we'll let you go. I had read you're a newly minted billionaire. You seem like a different type of CEO. Did you buy yourself any nice treats lately? After 20 years working at the same company?

1:10:28

Easy come, easy go.

1:10:40

So give us something. A new scooter, a windsurfing board. What do we got?

1:10:44

Okay, okay. My answer on this a year ago was different. So a year ago I'd say one of the paradoxes of adulthood. Like when I was a kid, I just lust over gaming PCs. Just never had the best. Like I just always wanted to play the latest game and never had the graphics card or PC to support it. And if you asked me a year ago, I'd say I'm an adult. Like one of my funnest things to do is like buy brand new computers and hardware when they come out. But I don't enjoy playing games anymore. But that's changed. I just bought a new gaming PC and I've been playing Arc Raiders and it's so fun.

1:10:51

At all the highest settings.

1:11:28

At all the highest settings. I can't, I can't even get a screen big enough to max out this thing. So would you.

1:11:30

What PC did you get?

1:11:35

Actually got also getting older. A Corsair. Corsair has like a pre built desktop that has like the 5090 in it. You know, I used to build PCs, but it's like what do I like doing better? Like playing games or building PCs? Like playing games.

1:11:37

Right on. He said that was the last. This is actually the last. Do you have a subreddit right now

1:11:51

that sounds like the Arc Raiders subreddit

1:11:55

under discovered that you think people should be checking out.

1:11:58

Well, Arc Raiders is pretty hilarious. Okay. You know, it's a good subreddit when, when, when it's indistinguishable from its own circle jerk subreddit.

1:12:02

So.

1:12:12

So Arc Raiders and Arc Raider, Arc Raiders, circle jerk. Like the game is so masterful. You literally cannot tell the difference between these.

1:12:13

I did a few of those as well and I follow them both and I could never remember which is which.

1:12:21

It's so great. I just. I love that dynamic on Reddit. It's like when the real subreddit and the satire subreddit are interchangeable. Like, you know, you're in a special place.

1:12:25

Hell, yeah. All right, Steve, we appreciate your time. Good to chat with you.

1:12:35

Thanks, guys.

1:12:38

Nice chatting, Steve.

1:12:39

Bye.

1:12:40

And that is it for this week's show. Thanks to Steve for coming on. And don't forget to, like, subscribe everywhere. You get podcasts. We are access show on the Internet and you can find us in video via access pod on YouTube.

1:12:44

If you like this episode, leave a five star review, please, or a thumbs up. Share it with a friend. It really helps. And you can find my newsletter at Sources News.

1:12:58

You can find my newsletter at Nowhere because I don't have time for that. But you can find me at Hamburger, on twitter and@meaning.com Naomi, we'll get you

1:13:08

one one day access is part of the Vox Media podcast network, and this show is produced by Hooked Creators.

1:13:16

No, you won't.

1:13:22

Bye.

1:13:22

Yes, I will. Bye.

1:13:23

It's tax season, and at Lifelock, we know you're tired of numbers, but here's a big one you need to hear. Billions. That's the amount of money in refunds the IRS has flagged for possible identity fraud. Now here's another big number. 100 million. That's how many data points LifeLock monitors every second. If your identity is stolen, we'll fix it. Guaranteed. One last big number. Save up to 40% your first year. Visit lifelock.com podcast for the threats you can't control. Terms apply.

1:13:29