You know, welcome everyone to the Deutsche Bank 2025 MIT Conference. Thanks for making your way down here. You know, we're really excited to kick off the conference today with Steve Huffman, co-founder and CEO of Reddit. Thanks for leading us off today.
Thanks, Ben. Good morning. Hi folks. Hope you're doing great.
All right. It's been an exciting ride since you went public, you know, around this time last year. Maybe to kick things off, you know, what have been your key learnings over the 12 months? You know, how have your priorities changed? Maybe, you know, what surprised you the most one way or the other?
Yeah. It'll be a year, I think, in a couple of weeks. I don't think we could have asked for a better year last year. That's for sure. We hit, so I came back to the company in 2015, so almost 10 years. In 2015 we did $12 million in revenue and we had 12 million DAU. And we had a number of long-standing goals that we crossed last year. We got to 100 million DAU, which we've been chasing for a long time. We crossed $1 billion in revenue, which again, we've had our eye on that for a long time. And then we got to profitability for the first time as well. Really fun year. In terms of, you know, what we learned, I think both a story of consistency and evolution.
We're executing, I think, better than ever. We're just building more, shipping faster, higher quality. All of these things make us really happy. The reason for that is we've just been consistent, consistent on the strategy, on the communication, and that's just driving results. There also has been some evolution, which is on the search side. We went so far as to even expand our mission. Our mission now is to empower communities and make their knowledge accessible to everyone. 'Cause what we really saw last year is the emergence of another use case on Reddit, which is users going to the historical Reddit corpus looking for answers to their questions, advice, and so forth. Really embracing that. Our users are already doing it on Reddit, and embracing that as well.
Like, big picture, I don't think we could have asked for more last year.
Yeah. No. Certainly. We'll dive into some of those topics. First, users. It's been a topic over the last quarter. You know, DAUs have been growing at a pretty rapid pace for some time now. Last quarter, U.S. logged- out DAUs stepped down modestly sequentially. Can you just give us a peek behind the curtain and tell us sort of, you know, what happened? Perhaps more importantly, why do you think this is more of an aberration, not potentially a lasting trend?
Sure. If you look over the last year, we grew users 39%. We crossed that 100 million DAU mark. I think lots to be happy with there. Logged- in users on Reddit, we grew over 27% every quarter last year. Also, I think very healthy. We basically have two traffic sources, direct to Reddit and then from Google. We saw some chop from Google, which not seeing chop would actually be more of the aberration, right? They make algorithm changes. They fiddle with their product. Sometimes that causes, you know, downstream effects to platforms like us. What we saw in Q4, which was a little bit of a swing that we've since recovered from, is a corresponding increase in the search term Reddit on Google, which says that internet consumers broadly want Reddit's content.
They actually go to Google with the intention of ending up on Reddit. Last year, Reddit was the sixth most searched word on Google, between news and maps. I think that foundation is there and very strong. Look, we're, you know, what, 48, 49 million users in the U.S., DAU. Weeklies are 170 million DAU, also growing nicely. I think the opportunity there is as large as anybody's.
Right. Maybe doubling down on Google just a little bit, just can you speak to how your relationship has evolved and maybe give us an idea of how it's a mutually beneficial partnership from a user perspective, from an engagement standpoint? I guess relatedly, a question that we get a lot from investors is, you know, how should we think about your dependence on Google as a source of traffic? You know, there are many other companies that fit into that bucket. Typically that's driven some volatility in revenue. That's driven volatility in earnings, which has impacted the way that some of these companies have been valued. Why do you think Reddit's different in that case?
Sure. One of the main differences between us and anybody who gets a lot of traffic from Google is we get more traffic direct straight to Reddit. A lot of companies are wholly dependent on Google, and so that's their business. That's not us. You know, for example, in Q4, where we saw a little Google chop, revenue was the strongest ever. It was one of our best quarters last year, you know, growing over 60% in the quarter. That's basically unaffected. I think the big picture here is internet consumers want what Reddit has. Whether they come to us direct, obviously that means they want Reddit. Or if they're going to Google and appending the word Reddit to their query, that also means they want Reddit.
Within the Google traffic that we see, about half of that on any given day is logged in. Those are our core users. We're gonna see them one way or the other. It's there can be kinda shocks to the ecosystem here or there if Google makes a big product change or an algorithm update. Those happen pretty consistently. We are very experienced in responding to these things. Big picture, our relationship with Google is great. We collaborate with them. On the search side, obviously we have a ton of content in their index that makes their search product better. It's an amazing channel for us, particularly, you know, those logged- out users coming from Google. Though it's volatile, it's a great opportunity for us to teach internet consumers broadly that Reddit has the answer to their questions.
It also happens to be our least valuable cohort of users from a monetization point of view. So it doesn't really affect revenue. That's why you didn't see any revenue movement, you know, related to anything that Google does. And then we collaborate with Google on the AI side. They're a customer for our data. We're a big cloud customer. We're mutual advertisers. We collaborate on safety. It's a really deep and healthy partnership.
Right. Right. I guess one of the ways that Google benefits from you guys is their ability to launch a new product, Google, through the AI Overviews. A question that we get from investors is, you know, how have these potentially impacted your engagement trends? Just conceptually, if a user goes onto Google and finds, you know, what they're looking for on an AI overview, there's conceptually no reason to click into Reddit. Is that something you've seen? Is that a concern? Is that potentially evolving how you think about the partnership and Google's product set?
Reddit, the Reddit product is a product for community and conversation. That's what people are coming to Reddit for, is to be connected with other humans. AI will never replace that if people want to have that experience talking to other humans about what's going on in their lives. What Google does is it helps users, kinda navigate Reddit, right? We have this 20-year corpus. To the extent that, you know, a user has a general question, what Google teaches them is that Reddit has the answer. Whenever they're showing Reddit content, they're also linking to Google. We have confidence in this relationship because we've been, you know, in Google search for a long time. Although I will point out, not for our entire history, you can't get traffic from Google without content in the index.
By and large, that's we grew without that. Internationally, we're also growing without that because, again, the product we're building is a community and conversations product. Now, AI summaries that you see in search is really an evolution of the search product. We've been a beneficiary of search on the internet for a long time. We expect to continue to do so because, again, Reddit is where internet consumers broadly want to be. I think we also have opportunities to build an amazing search product of our own. That's one of the things that we're really taking on in 2025. Our first version of this is Reddit Answers.
This is an AI-powered search results on Reddit itself, which, even in the first, it's only been online for about a month and a half, is, I think, a really fascinating glimpse of what's to come.
Right. We'll get into Reddit Answers in a bit. Just to wrap up the discussion on perhaps U.S. user trends, longer term, how big do you think Reddit can become from a scale perspective in terms of users? You know, you zoom out 10 years, like, how big do you think you'll be? Who would be your closest peer?
Reddit is universal. We have communities and conversations for literally any topic that would be of interest to you, whatever you're into, whatever you're going through, right? Every hobby, every passion, every interest, every major life decision, every vulnerability. Reddit is one of the few platforms on the internet that can claim universal appeal. That's how we think about it. We're at 100 million DAU today. Now we're shifting our sights to how do we get to a billion DAU. In the U.S., you know, let's call it around 50 million DAU. We have about 170 million people in the U.S. visit Reddit every week. That's as big as anybody. From a product point of view, how do we convert those weeklies into dailies? That's product quality, make it stickier, make search better, make onboarding better.
The potential is there in the weeklies, 170 million in the US. Outside the US, it's, you know, we're also about, call it, 50 million DAU. We've got a long way to go there. Like I said, I think we can be as big as anybody 'cause Reddit is one of the few platforms that can claim pretty much anybody as a user. If you're, call it, 17 to 70, men, women, nerds, normies, you know, whatever it is, there's a community for you on Reddit. That's how we think about it over the long term.
Interesting. Yeah. You mentioned product. At least from our vantage point, you seem to be one of the CEOs that's really hands-on in terms of the product side of the business. You know, and I think, Steve, you mentioned product velocity is pretty high. Reddit Answers is another thing that you mentioned, that sort of emerged last year. Can you just shed some light on the early learnings of Reddit Answers? You know, when we look ahead, you mentioned it's sort of a prototype, is what you mentioned, I think, on the earnings call. You know, if we look ahead over the next few quarters, how do we expect the product to evolve?
Sure. Look, so fundamentally at Reddit, we're builders. We get to work on this really fun product. It's a super fascinating thing to do because Reddit, sometimes it's less a technology product and more feels like this living organism that we're studying and understanding and trying to facilitate its growth. One of the big things we've seen develop over the last few years is our core product is community and conversations. If you're in the app or you go to reddit.com, you're primarily interacting with content, with posts that were created within the last 24 hours. It's very recent.
In conversations on Reddit over the last 20 years, it'll be 20 this summer, our users have left this massive trail of information about every facet of the human experience, everything they're going through, thinking about, purchasing decision, all of these things, in this massive 20-year corpus, which has been more or less unavailable to our users and internet consumers more broadly. Google actually only indexes a fraction of Reddit's corpus. When we talk about search or that second clause in our mission to make community's knowledge accessible to everybody, it's really unlocking that corpus. We have traditional search on Reddit, which itself is getting a big investment. Then we have a product we whipped together. We did it in like 90 days.
We put it online in December of last year, really to answer the question, is there something worthwhile here? What Reddit Answers does, you can type in a question on Reddit. Subjective questions work best, right? What's the best way to get started with this author? What's the best travel backpack, right? What's the tipping etiquette in Thailand? You know, just like things like that that don't have clear answers. How do I make my toddler sleep through the night, whatever?
Don't think there's a good answer.
I guess I just kind of outed what I'm going through in life. What we do is we run a basic search on Reddit. You know, maybe we get 100, 100 posts or something like that. And then we use an LLM to summarize the results and to just give you, this is what the communities across Reddit think about that question. It's really, really powerful. We're learning for ourselves how interesting this is because there's nothing quite like it on the internet. You know, when we talk about AI summaries, AI can't tell you what it's like to use this backpack. AI doesn't have kids, you know, yet. There are these really important, interesting things about the human experience that only humans experience kind of by definition. That's what Reddit reveals.
I think this product will be unique on the internet in its ability to provide perspectives that are otherwise unattainable online. Anyway, Reddit Answers is a prototype. It's in the app. It's online. You can go to reddit.com/answers or it's in the app if you're logged in. Over this year, we'll be merging that with traditional search on Reddit 'cause I think there's a really good opportunity here, I think, to make our product just much more accessible and easy to use.
Mm-hmm.
'Cause if we think about the kinda ways people come to Reddit, broadly, we look at our user base as, we call them Scrollers and Seekers. Scrollers is the core Reddit product. They're in the app. They're on the front page. They're there for the community and conversation. That's the product we've been working on for 20 years. Scrollers need search as much as anybody because especially let's say you're a new Reddit user. You're coming to Reddit. Our goal is to help you find your home on Reddit. The best way to do that is for you to type into a text box what you're interested in or what you're thinking about, what you're looking for. I think search can be really powerful for new users. Then we have the seekers. Today, these are users coming from Google, right? They had a question.
They go to Google. They end up on Reddit for the answer. They're obviously searchers. Right now, the seekers coming from Google are interacting. They're hitting this community product. They're not hitting a page tailored designed to give them the right answer. It works pretty good, right? You can see that in our growth on Google over the last decade. I think we can make that product much, much better itself, which also leads us to a search investment. Search and those different entry points have been really important to us, this year.
Right. And just, you know, hearing you talk about that, is there right now, the products are embedded within Reddit. Is there a chance to potentially have these as standalone products, reaching back into the Reddit corpus and not necessarily being part of the sort of, you know, a bit of a breakout app?
That is an interesting question, and it's a debate that we have pretty much consistently, constantly. Would we ever have another app? Okay. Technically, yes, we are gonna have another app. I'll tell you about that in 30 seconds. But would we ever break a use case out of Reddit into a standalone product? The answer is maybe. I think that what would have to happen is we'd wanna see enough heat in a use case on Reddit that it would warrant doing that. There's some precedent for that, you know, elsewhere. You know, Facebook pulled out Messenger out of the Facebook app, things like that.
I think from a product point of view, we'd be looking for is there so much tension in the app or the app became confusing because there's, like, radically different ways of using it that we would need to split it apart. We're not there yet. I would call that a first-class problem to have. That said, we are building another version of our app called Reddit Lite, which we'll be testing internationally. We are doing two things here. One, we're just building a lightweight version of the app to use internationally, which is a kind of tried-and-true practice. Also, we're going to be experimenting with pretty big changes to the UI, making Reddit, in my opinion, much easier on the eyes, much easier to use, making a home feed that better accommodates all the different content types we have.
'Cause one of the things about Reddit, if you go into the home feed, you see text and images and memes and gifs and video and video ads. It's all there. It's a lot. From my point of view, when we talk about new users and converting weeklies to dailies, the path there is making the app more accessible and easy to use. We're gonna try basically a whole new version of Reddit and see if we can accomplish that. We'll be testing that outside the U.S. this year as well.
Excellent. All right. You mentioned on-platform search. I'd just love to hear sort of the, like, a status update on where that is right now, you know, what incremental investments you need to sort of improve that product. You know, and could you see sort of a meaningful impact to DAU growth through improved on-platform search? How do you think about the pathway and maybe even the timeline to monetizing some of those creates?
Okay. We've talked a lot about search. What are we doing? There's traditional search on Reddit. That's getting better as we speak. A lot of quality-of-life improvements coming online right now, things like smarter autocomplete, spell checking, just the basics, right? As a user, you expect these things. When we turn them on, you may not notice. It's just the search is getting better, improving performance and ranking. All of that is a work in progress. Traditional search on Reddit should get much, much better over the course of this year. We're shipping stuff rapidly there. Then folding in answers. When I say traditional search, I mean you're searching for a topic. Maybe you're searching for a specific post, kind of like keyword search, let's call it that, whereas answers, longer queries, subjective answers.
We don't wanna have two search products on Reddit. Bringing those things together. I talked about the entry points. Are you coming from Google? That's a different entry point than, you know, arriving on Reddit's front page or opening Reddit for the first time. What the reason I think search is so interesting and such valuable technology for us is it makes the seeker use case better. It helps new users find their home on Reddit, and it helps core users discover new content on Reddit and answer their questions. It's one of the few products that actually hits every cohort of users. It's search. It's ads on search result pages is literally the best business model on the internet.
Now, we may start scratching the surface of monetization this year, but we do need some consistency in the product before we start selling it. And so that, you know, that'll be that'll come later, once we've got the product, I think, in a really good place, in a more consistent place. Right now, it's under heavy construction. That's fun, but it's gonna be moving around a little bit.
Right. Understood. All right. Let's move on to international users. You obviously leveraged some of your LLM partnerships to launch machine translation across several new languages. Number one, can you talk about the early learnings there, you know, which languages you plan on launching next? Maybe a bigger picture question, you know, other than simply translating the content, what else do you sort of need to kickstart the community flywheel in in some of the non-English-speaking countries?
Yeah. Growing outside the U.S., big priority for Reddit. You know, I was talking before about the kind of long-term growth potential for Reddit. How do we get to a billion users? If we're 50 in the U.S. and 50 outside the U.S., and let's say we turn all of our weeklies in the U.S. into dailies, so that goes from 50 to 170, the next 850 to get to a billion have to come outside the U.S. A lot of wood to chop there. An incredible amount of potential. Let me just tell you why I think that potential is not that far-fetched. If 50% of our users today are outside the U.S., just in other parts of social media, X is about 80% international. That's on the low end.
Facebook, 95% international. Call it the high end. I think naturally, being between 80% and 95% outside the U.S. is where we'll land. Now we just have to get there. You need Reddit as a community platform, a conversation platform. You need content to attract users. You need users to create content. There's a bit of a, you know, cold start challenge there. One of the ways we're getting over the cold start challenge right now is with machine translation. We can use large language models to translate the English corpus of Reddit into other languages. We can do so at human quality. This was not an opportunity, an option available to us a couple of years ago, but LLMs have gotten the quality bar high enough.
They can actually translate humor and sarcasm at a human level. The French was France French in France was the first country that we started this in last year. It's one of our fastest-growing countries. In general, countries where we started using machine translation, they're growing 30-40% faster than other countries. It's a really nice opportunity to lay that kind of foundation. Ultimately, what we want is for every country to be able to stand on its own, right? Native language content created by native users in countries around the world. Machine translation is a really nice way to jumpstart that. We're also, in our focus markets, we have about half a dozen of them this year. We're in-country meeting with moderators, helping identify content areas, doing meetups.
It's very kind of high-touch work because, again, it's a human-driven process. We're right in the thick of it right now. I think the opportunity, as I said, is very large.
Right. Right. Absolutely. All right. Let's shift to revenue.
Great.
Yeah. I mean, it's been a powerful story. Your ad revenue growth last quarter, 60%, with another quarter, you know, year-on-year growth acceleration. Can you maybe talk about the driver that acceleration has been on users and delve in a little bit deeper into what was specific to Reddit versus what was a general sort of broader, you know, stronger digital ad market?
Sure. So look, great, great year last year, I think, across the board on revenue growth. Last quarter, you know, over 60%. We had a couple quarters like that last year. You know, outgrowing the market, outgrowing our peers in the space by quite a wide margin. So what's going on there? Our ads are working. You know, that's the long and short of it. Our ads are, the targeting is getting better. The performance is getting better. We saw particular, particular progress in the performance side of the business, which is really where we wanna be. So mid-sized businesses, small businesses. You know, we had in Q4 record number of advertisers and advertiser growth. But as a platform like ours gets better at performance, you just unlock more and more customers.
You know, generally, what happens is a new ad platform starts with large customers, brand advertising. Of course, we have, I think, great relationships there that are getting deeper and more valuable over time. For the mid and small businesses, they do not care about relationships. They care about, you know, clicks, conversions, objectives. The ads just have to work. For us, it has been a story of investing in the ad technology, and making that as performant as possible. I love the trend that we are on. The good news here is I think we have years of roadmap in front of us of basically known winning ideas, of things that advertisers expect that work very well on other platforms. We build it, ship it, and we just see the numbers get better and better.
I mean, I think what's fun about Reddit is we're growing fast, but we're still, I think, on the ad tech side early in our journey. There's just so much opportunity.
Yeah. You know, one of the new ad formats that came out or that's been evolving were Conversation Ads, right? You know, from our vantage point, from the outside looking in, they appear to be a pretty decent hit, actually. Can you talk about the learnings there? Perhaps relatedly, is this one of the initiatives that you have up your sleeve to, you know, better monetize logged-out users? If that's the case, you know, what else are you working on to narrow that monetization gap between the logged-in cohort and the logged-out cohort?
Great question. Basically, there's kinda two main surface areas on Reddit for ads right now. One is in the feed, which, you know, if you've been on the internet, you're familiar with that ad type. The other is on our conversation pages. If you click or tap on any post on Reddit, you get into the conversation. This is really the heart and soul of the Reddit product, right? This is where our users are talking about things in a way they don't communicate elsewhere online. They're often talking about a specific question, a specific decision, often a specific company or their product or vertical. As an ad surface, it's really powerful, and the opportunity there is great. We put ads in that surface last year, and it's performing very well because the pages are so intentional.
Of course, our visitors from Google are landing on those conversation pages. They are basically those pages that are one click more intentional than what you'd see on Google, again, talking about a specific company or product or purchasing decision fairly often. It is a huge opportunity there. To grow that business, it is better targeting, more advertisers, more advertiser density, making sure that we have the perfect ad to put on that page every single time. When we talk about, you know, logged-out users and, you know, what's the best way to monetize them for the external traffic, yeah, some of them do log in. In that moment, in their session, they ran a search, right? They are looking for an answer to a question.
We expect them to be sort of transient because it's a short internet session in general, right? They're doing something else with their time, but making sure when they land on Reddit for the answer to their question that there's the most relevant ad possible. I think that's a really lucrative opportunity for us if we make that connection. I think it'll be really beneficial to the user as well, again, because they're in a transactional mindset. How do we do that? It's the basics: more advertisers, better targeting, targeting in this case on context, right? What is on the page? Just continuing to chip away at that, I think it's a massive opportunity.
Right. All right. We're gonna switch things up a little bit here. In preparation for this, for this chat, you know, the Reddit team solicited some questions from the community, in the RDDT subreddit. The question is on video experiences, right? Reddit is great at catering to users who wanna dive deep into communities or casual users, you know, or, you know, just looking for information. With the surge of platforms like TikTok and Instagram Reels, which obviously capture audiences who want quick, engaging video content, do you see opportunities for Reddit to evolve in the future, to develop features that would appeal to this user segment that favors, you know, the more fast-paced, video-centric experience?
First, thank you for asking the question. This was a question from our user base, and we try to include our users in these things. One of the reasons we went public was so that our users who feel a deep sense of ownership over Reddit could actually be owners of Reddit. Thank you for doing that. The question is, do we see more video and short-form video on Reddit? How are we thinking about that from a product point of view? Look, Reddit is a conversation platform. Whatever it is that people want to talk about, they can talk about on Reddit. This means from a content format point of view, we try to be agnostic, whether it's text or links or images or video. The real magic of Reddit is talking about the thing.
Reddit has lots of video, as it happens. It's one of our fastest-growing content sites. It's a little bit different than what you see on social media. We're on social media. It's often camera in, right? Look at me. On Reddit, it's more camera out, right? Let's talk about this thing. Important nevertheless. What we try to think about is how do we make Reddit a comfortable and effective platform for talking about any content type? Having video and text in the same feed, that's both the magic and, I think, some of the challenge of Reddit. I was talking earlier about Reddit Lite. Can we reimagine Reddit to make it more effective at kind of juggling these different content types? That's one of the big things on our mind: make it a platform that's as equally hospitable to text and video.
I expect video will continue to grow just because it's such an important content type to the internet. I think it will manifest on Reddit as it is in a little bit of a different way just because the way people talk on Reddit is a little bit different.
Right. Right. Okay. Moving on to a topic that I think is near and dear to your heart but has not gotten a lot of limelight recently, the Developer Platform, right? You know, perhaps for those who are not as close, can you explain your ambitions with the Developer Platform and then maybe talk a little bit about the results of the public beta? You know, what surprised you? Where do you need to do a little bit more work, for instance?
Sure. Developer Platform is near and dear to my heart. Reddit, the first thing to understand about Reddit is everything interesting about Reddit was created by our users. It's a relatively simple platform. It's communities. It's posts. Saying communities kinda masks all of the interesting things our users do with communities on Reddit, right? Some of them are for advice. Some of them are basically games. There are places like Ask Reddit or WatchE xchange where people are buying and selling luxury watches. There are sports and TV-based ones where people are talking about, you know, things that are happening in real time. What we are trying to do with the Developer Platform is allow any user to expand how Reddit is used so they can write code that will run on our clients to make Reddit more interactive.
I'll give you an example of kind of the high watermark for this. On April Fool's Day, a couple of times, we've run these interactive games. One of them was called Place. By the way, if you've never used Reddit, you should go to YouTube and watch a video of Place. It was basically an interactive game where users had a shared canvas of 1,000 by 1,000 pixels. A user could place a pixel every 10 minutes. Over the course of a week, it turned into this dynamic, living work of art, if you will. 10 million people played it in one canvas. It's really, really insane scale. We know these kind of game-like social experiments work really well on Reddit. With the Developer Platform, I wanna do two things. One, let users make Reddit more interesting.
We're seeing this already, like the scoreboards. If you go to Wall Street Bets, they have a widget at the top. That's the tickers most talked about on that subreddit today. That's all Developer Platform. What we'd really love to see, like our dream, is for users to be able to create kind of interactive experiences and games like we've done, with Place. We haven't had a hit like that yet, but that's the sort of thing we're trying to unlock.
Right. Let's move on to data licensing. It's a big story, at least last year, certainly a big contributor to growth. Can you just talk us through how your thinking has evolved there as it pertains to data, data licensing and how should we anticipate that line item to move forward?
Yeah. I think one of the biggest learnings last year is the value of Reddit's corpus, of the nearly two decades of conversation about anything and everything. We did a couple of deals last year, one with Google, one with OpenAI, licensing our corpus for training for their LLMs. We also have started investing in our own work to kind of reveal the value of that corpus. That's, you know, that's where our head is at right now. This is search. This is Reddit Answers. This is training our own models. We're still, you know, in the market. We're open and open for business. We've done a number of smaller deals. I mentioned the big ones.
We've done a number of smaller ones to kinda like social listening services or financial institutions, anybody who wants a real-time feed of what people on the internet or people in the world are talking about and thinking about. You know, we'll kinda see where this goes in terms of is it smaller customers? Is it bigger customers for training? What I know for sure is that there's an incredible amount of value there. I think we're actually in the best position to kind of capture it.
Right. Speaking of LLMs, and obviously the DeepSeek news recently, I'd be curious to hear how you think about, you know, moderating costs of AI and how that could be specifically beneficial to you at Reddit.
Sure. I mean, look, this is with DeepSeek and some of the other open-source models, we're seeing happen exactly what we've been saying would happen, which is the actual like LLM technology will become commoditized. The price will come down, and the open-source versions of these things will be, you know, at any point in time, let's say 90% as good as the major commercial closed-source versions because that's the story of, you know, internet, software for the last 20, 30 years. It is playing out, I think, exactly as we would have expected. I think all companies can benefit from AI across their business, but Reddit in particular, right? Reddit is text and words and content evaluation and translation and, you know, moderation, safety, ad targeting. All of these things are things that AI and LLMs specifically can dramatically advance.
I see just across the board so many ways for our product to be better, safer, more interesting. Whether we use a commercial model or we make our own, both of which we do now, I think there's just opportunities across the board. I mean, it's crazy. You can take DeepSeek's model and run it on like four GPUs, right? You can run that under your desk. Obviously, that's a huge opportunity for us and I think every company in this space.
All right. Parting question again from the Reddit community.
Great. Ben just wants to make it clear it's not his question.
No. It's a great question. Sorry. You know, where do you envision Reddit's product and community evolving over the next 10 years? I know you can spend as much time and as little time as you want on this, but yeah, it's interesting.
Okay. I've got 17 seconds.
The shot clock's down.
Look, we talked about these things. Our four priorities: improve the core product, grow the ads business, search, internationalization. You know, I think about how do we get to a billion users? I believe Reddit's product is universal. So how do we make that a true statement, convert our weeklies into dailies? Product quality, performance, search. It's the basics, but I think the path to where we wanna go is just through doing the basics at a very high level.
Awesome. Thank you, Steve. I look forward to seeing how things progress over the next year and maybe you give us a status update in 2026.
Only if you do this in Palm Springs, which is where I thought we were going. Thanks, Ben. Thanks, folks.