Watching Reddit “eat itself”

STEVE HUFFMAN: There were things happening on Reddit that I thought were bad for business, bad for morale, and bad for the world. And so that’s when we had to start wrestling with hard decisions. We couldn’t say, “Hey, just everything goes. I was watching it die. Like, we were there watching it eat itself.”

REID HOFFMAN: This was 2015. Steve Huffman, one of Reddit’s co-founders, had just returned as CEO after watching the site he built from scratch struggle to survive. 

HUFFMAN: When I came back to the company, employees, they wouldn’t wear our swag in public. Cause they didn’t like, you know, what was going on and what was being written about Reddit. But then I would ask them. I was like, “Okay, if this is so difficult for you, why are you here? Like, why do you work here?” And they’d say, “Because I love Reddit, because I love what it does for people, because I know that 99 percent of Reddit is amazing, and there’s nothing else like it. And Reddit has changed my life, and that’s what I’m fighting to protect.” And I was like, great, then let’s protect that. And so that’s what we got to work on. 

HOFFMAN: Steve Huffman co-founded Reddit with his college roommate when he was just 21 years old. In the nearly two decades since he and the site have both grown up a lot. Earlier this year, Reddit — often referred to as the front page of the Internet — had its IPO. It was valued at 6.5 billion dollars. 

I’m Reid Hoffman, your host. 

I was eager to talk to Steve Huffman about Reddit’s unconventional business, but first I wanted to talk to him about the scale of Reddit’s impact on our digital culture. 

Steve, I’ve been looking forward to this for months. Welcome to Masters of Scale.

HUFFMAN: Thank you. I’m excited and honored to be here. 

The evolution of Reddit’s content policies

HOFFMAN: How do you think about Reddit’s role on the internet and the online ecosystem? And what are the things that it’s for and part of the internet treasure of, and what is it not for?

HUFFMAN: A saying I say a lot is that like — Reddit is people. The idea of Reddit is that it both reveals people and empowers people. And that was one of the important ideas that was in our mind, you know, back in 2005 when we started Reddit, and it’s still one of the ideas we come back to often.

You know, we start from a principle of free expression or free speech. But there’s, I think, also, there’s a parallel principle, which I think is really important, which is this kind of idea of decorum, right? How are you speaking? How are you treating others? Separating behavior and beliefs.

And so we try to spend a lot more time on behavior than we do on beliefs. And so that gets us into our content policy. So when we started right in 2005, there basically was no content policy. You know, we thought anything, everything should be allowed. Like who are we to say what’s not allowed? Easy to take a position like that when, you know, this really wasn’t any bad behavior on the platform.

You know, maybe some spam here or there, but we just didn’t have difficult decisions to make. Now our policy is much more robust and constantly evolving. So, no hate, harassment, bullying, no spam, nothing illegal, no inciting or glorifying violence, no involuntary sexualization, nothing involving kids.

There’s a lot of, kind of, sub-rules and things like that. But we’ve, I think, had to adapt as we look at, like, how are people using the platform, maybe in ways we didn’t intend, and adapting to that while still trying to preserve the core idea that people should be able to converse and debate, and argue, and talk about more or less whatever they’d like.

How Reddit incentives healthy interactions

HOFFMAN: I think, you know, probably the most common brand for internet discourse is it’s uncivil. You know, that there’s a lot of polarization, you know, hate speech, etc. What are the things you do to try to encourage the positive aspirations of human discourse?

HUFFMAN: Okay. It’s a great question. And I think I have a couple of ways of thinking about it. So first is the general way, when we think about people and how people behave. 

When two people are together in person, they generally get along. Even the far left and the far right like to take extremes in the United States right now, chances are in their day-to-day life, the vast majority of what they’re going through or would talk about, they have in common, right?

Maybe they’re parents, maybe they are sports fans. Maybe they watch the same TV shows or movies. That’s this one, I think, just fact: it’s like people generally get along when they’re together in person.

So on Reddit, every conversation you have is in the context of a community, you’re not just yelling into the ether — communities have rules. They have written rules. They have unwritten rules. So this kind of idea of decorum, there’s a social pressure to behave according to the rules of that community, right?

If you don’t, you’ll be downvoted and you’ll be silenced by your peers, which by the way, getting feedback from your peers like that is way more powerful than getting feedback from, you know, the company or the man. When your friends tell you, Hey, you’re being rude. That’s very different from your parents telling you being rude or, you know, the government, right?

HOFFMAN: What things would you kind of suggest as things to do to kind of improve online interaction culture?

HUFFMAN: I think a lot of it comes down to incentives. And so if you look at the incentives of social media, a lot of it is about, there’s a strong attachment to your real-world identity and status. And so people are saying things, I think, not necessarily because they believe them, but because they want to be perceived as believing them in the real world. And they want to be accepted by other people who they perceive to believe these things.

There’s a kind of a showmanship or performative aspect to it. And I think what happens is there’s this kind of drift to the extremes.

Reddit has the upvote. Reddit also has the downvote. And so that’s where a community can reject behaviors or whatever they don’t like. I think this is really powerful because if you’re extreme or polarizing, chances are, you’ll get just as many downs as you do ups. And so it kind of cancels out.

Whereas, I think, elsewhere on the internet, you just get a lot of ups because it’s, it can be, you know, more attention-grabbing or emotional, or whatever. And so on Reddit, the controversial stuff sits at zero points, right? Ups minus downs. And so we get a kind of a better outcome there.

It’s humans behaving more humanly. I think that’s really powerful. And they don’t get credit for it. The only thing they get is a little bit of karma on Reddit, which, I guess, is worth more than zero, but you can’t take it with you.

Measuring success with both data and intuition

HOFFMAN: As CEO, what do you look at as a dashboard for what’s going on with Reddit? I mean, obviously, there are questions around revenue and engagement, a lot of that kind of stuff. And that’s interesting too, but I’m curious if there’s anything else that’s kind of very Reddit-ish about your dashboard?

HUFFMAN: Oh, that’s interesting. So look, there are dashboards with, you know, revenue and users. And of course, there’s another layer of metrics that we really care about. You know, number of advertisements from these companies and retention, you know, customer attention, and new user attention, all these things.

But for better or for worse, I use the Reddit app a lot. I have been working on Reddit for 19 years. A lot of how I think about Reddit, it comes not from the numbers, but from the feel of using the app. 

I feel like my intuition is constantly being refined. And that’s what I take to the team. It’s like, “Hey, I think I see an issue here.” And so from a product development point of view, I think we went through a phase at Reddit, which I think a lot of start-ups go through, where they actually put their intuition aside. And they only look at numbers.

I think it’s really hard, if not impossible, to run any company or system from a spreadsheet. And I think there’s a little bit of this kind of false ideal about what we’re going to be so scientific and disciplined, you know, put metrics on everything, and test everything, but you lose some of the most important information, which is, is it working, is it good, how does it feel, do I enjoy this, do I see other people enjoying it?

And so for the last couple of years, I’ve actually been working really hard to give the team permission to listen to their gut more. Honestly, I think the company is getting much better at it, but that has been a real, you know, piece of work over the last couple of years.

HOFFMAN: No, I couldn’t agree more that it’s kind of a combination of art and science, combination of intuition, and data, you know, and hitting both is what’s actually in fact very important. 

HUFFMAN: Right. Looking solely at data, I think, at best you can optimize to a local maximum. To do something new, you have to have some vision for the future. And then you have to take a few swings at it. Of course, you’ll be wrong along the way but to get there, you have to make a few leaps. Now, and then there’s the other side of that, which is you do want to test your intuition and see, okay, I have this belief, this hypothesis.

If we build this this way, users will like it. And so we build it. And then you have to wait a little bit to see, do they like it? And there’s a really important step, I think, in there, which is, do I like it? And I know we’ve built stuff at Reddit over the years where we didn’t like it. But we pressed on.

And so like, for example, we used to be very aggressive on mobile web when you come from search, external search about getting you to like log in, and like, yeah, I made the numbers work, but we all hated it. And then our users hated it. Even our investors hated it. And so it didn’t pass the first test, which is: we didn’t like it.

Resisting the urge of controlling users

HOFFMAN: How do we build better towards a shared context where we have increased trust in information, in, kind of like, what is the state of the world, where we can build bridges versus polarization. What do you think the techniques we should do to try to build in that direction?

HUFFMAN: The mistake I think we continue to make is at all different levels. I see people do it. I see companies do it. I see the media do it. I see government do it — \ they keep trying to take more control. I don’t like what people are saying, or I don’t like what I’m hearing, or I don’t like the way this conversation is going.

Therefore, I need to take more control. Hold on even tighter. And I think the solution is to resist that urge because that control, that manicuring the message, that kind of pushing people into the desired outcome is what creates that sense of distrust. That sense of I am being manipulated. I am being controlled. I am being constrained. 

HOFFMAN: You know, I’m thinking around the discourse that’s had around, you know, vaccines, which, you know, frequently got idiotic. You know, where some online discourse, QAnon, et cetera, what are the things that you think are key for making those approaches that aren’t exerting control, but are building trust and shared reality?

HUFFMAN: Transparency. I think, in that era, and honestly, we still do, I think, we lacked transparency. I think, broadly, for a leader to be credible they have to share the bad news along with the good news. Because anytime somebody is telling you only good news, you know there’s something missing because there are always trade-offs. And I think as a general rule of thumb, the more emotional they are, the more suspect the message should be. 

But I sense there’s always this fear, and this is one of the things that drives me nuts, is this distrust of people. ‘Oh, people can’t handle the truth.’ That creates more distrust.

HOFFMAN: Yeah, no, you have to give trust in order to get it. 

HUFFMAN: That structure of that sentence, I think is really important. You have to give trust to get it. If you want people to behave like adults, you have to treat them like adults. If you want to have good friends, you have to be a good friend. That sentence structure is just, I just glued onto it because it’s just one of the things I use a lot in my day-to-day life. 

The opportunities & threats of generative AI

HOFFMAN: Reddit is not a generative AI company, but Reddit’s name does come up a lot in conversations about the business of AI. That’s because it’s such a rich and dynamic trove of data — so many human interactions that could be used to train large language models. And indeed Reddit has struck deals with Google and OpenAI. So I wanted to talk to Steve about both the opportunities and threats to his business from generative AI.

HUFFMAN: At the minimum, I think we can all agree large language models are incredible at text and words. So I think LLMs across the board are going to revolutionize and hopefully reduce the amount of time we spend doing this drudgery, like word processing. So that’s very exciting. 

So I think there is a tremendous amount of opportunity there. The other interesting dynamic with Reddit is of course, our corpus has been used as a training set for all of the major LLMs. We’re getting closer to the ‘with our permission’ part. That’s been a little bit of a journey, but I’d say taking the business relationships out of it, or, you know, the IP conversation out of it. I’m very proud that Reddit has played a role in the development of these technologies. That’s part of our validation of our belief in the open internet, which is we’d like these things to be accessible and used to advance the state of the art. And so that part is interesting. Now, of course, the part that I think is general across IP holders is we can’t give it away for free because that undermines the IP creation incentives, the IP creation business.

So whether you’re talking about content on Reddit or newspapers, or books, or music, or movies, there’s a lot of human thought and work that went into the creation of that IP. And A. I., one of the things we’ve learned with these models, they need to be continually training, kind of absorbing new human intelligence, right?

You can’t have artificial intelligence without human intelligence. And so I think practically there’s an ecosystem, there’s a balance here that’s really important, which is: if human intelligence isn’t compensated fairly, then AI will eventually lose access to it. It’ll dry up. There’ll be no incentive to create new art, music, words, whatever it is.

And so I do think there needs to be a balance there. I guess that’s a big picture. Small picture, or much easier to explain, which is, Hey, you’re building these huge for-profit companies, you are enriching yourself. In what world is it fair to basically take or steal resources for somebody from somebody else to enrich yourself with no compensation?

And so that is an issue. And I think that’s probably those, those two aspects of are you eliminating human creativity and are you committing business malpractice by just taking all of the content you can find without permission and using it to enrich yourselves. Now there’s another aspect of it for Reddit, which is, let’s say we answer the business questions, which we have with a couple of folks, you know, Google and OpenAI being the two big ones. We still have other constraints, which is protecting user privacy.

Deletions, right? You can delete stuff off of Reddit, you know, we want to make sure those requests are forwarded on. We don’t want any company to use Reddit content to reverse-engineer the identity of our users or to use it to better target ads against them. That’s not what we’re trying to do.

So again, that’s kind of like the business thing, which is like, can you use Reddit data to, you know, make search engines better and make search experience better, but continue, for example, to send traffic to Reddit? Which has been the balance or kind of the relationship between us and search engines for a very long time.

So, you know, I think we’re getting there. Certainly, I’ve gotten there with a couple and I think we’re in a good place. And we’ll kind of see where things land with everybody else. 

HOFFMAN: Still ahead, I talk with Steve about Reddit’s scale journey and how becoming a public company has changed things.

[AD Break]

HOFFMAN: Welcome back to Masters of Scale. You can find this conversation and more at our YouTube channel. 

HOFFMAN: All right. So let’s go to the Reddit scale story. Let’s start with the origin story. 

Reddit’s origin story

HUFFMAN: So, Alexis, my co-founder, college roommate at the time, he and I applied to Y Combinator. 

HOFFMAN: Y Combinator is an incubator for start-ups that’s been the starting point for many successful companies.

HUFFMAN: This was the first batch of Y Combinator in 2005. We applied with a different idea. It was ordering food from cell phones. A good idea, I think.

Would we have made it successful? Probably not. Probably a little ahead of our times and probably a little ahead of our skills. But Paul and Jessica, so Paul Graham and Jessica Livingston at Y Combinator liked us. So after rejecting us on the food idea, they basically invited us back to work on something else.

Paul had half this idea, which is that he really liked Delicious. Delicious was just a website back then, social bookmarking. And he liked the dynamic, there was this kind of side effect on Delicious where if people were bookmarking the same page over and over, you could see what were the most popular things bookmarked on Delicious. It was called Delicious slash popular. 

And then I had the other half of the idea, which was Slashdot. I loved Slashdot. Slashdot’s still around, Slashdot.org., it was kind of headlines on the surface, chosen by editors though, but what was really amazing about Slashdot was the community that existed in the comments about any particular headline.

And Slashdot’s constrained to tech. What if there were no editors, and it wasn’t constrained to tech? And so, it was kind of a Delicious plus Slashdot, but make both of them better. That was the idea for Reddit. 

Honestly, I think that’s pretty much what we built. But for 19 years, we’ve been iterating on this and tweaking it, and kind of following our users and adding features, like for example, the most important feature we ever built was for users to create their own communities. We didn’t build that until 2008, you know, almost three years later.

Inside Reddit’s sale to Condé Nast

HOFFMAN: And then, there was a sale to Condé Nast. What happened there?

HUFFMAN: Condé Nast or Advance. Advance is really the parent company. I use those words interchangeably. When they came along wanting to buy Reddit, you know, they’re kind of scratching a couple itches. One was I felt like it was, well, that’s fulfilling the prophecy. Like that was what was supposed to happen. We built this thing, and now we get to sell it. 

And then two, like we were a really small team in that era, and I mean, it was all I could do to keep the site online. It felt like the bottom could fall out any minute.

Now that was incorrect. I just didn’t know what product market fit looked like, the fact that this thing was growing despite us hanging on by our fingernails, the thing was actually probably far more powerful than I realized, but I was like, Well, we better find safety before we run out of a runway here.

That said, Advance were really great owners. They never really meddled. The powers that be at Advance, who did the acquisition, who are still involved with Reddit today, like, they love Reddit. And so, you know, we got to continue working on Reddit, continue to evolve it, continue to evolve it, kind of following our own intuition.

I and Alexis, and some others would end up leaving when our acquisition contracts expired. But Conde Naste were great stewards of Reddit, and then they eventually realized that Reddit needs to be independent. So they spun it out. And so, you know, all things considered, how can I say different from where I’m sitting today, things worked out really well. But, I think as with many decisions, we were very much living kind of in the present in that era.

HOFFMAN: So, as part of the kind of Steve narrative-hero journey, you kind of covered a little bit of why you left, what’s the return look like?

Lessons from Steve’s failed start-up Hipmunk

HUFFMAN: Yeah. So I left because I wanted to run my own company. And so I no longer owned Reddit, I was an employee. So I started a different company, Hipmunk, with another friend, Adam Goldstein. Fun company, fun product, bad business.

HOFFMAN: Hipmunk was a travel-focused search tool Steve started in 2010. 

HOFFMAN: It was a great product, by the way.

HUFFMAN: You know, if we had just built that product, we built a product in the summer, if we had just built that product and run it as a small company, and then maybe sold that or licensed that. That would have been a good business. Instead, you know, raised a bunch of money, raised a bunch more money, this and that.

Anyway, lessons learned. It’s hard to be small in Silicon Valley. So while I was gone, like I still used Reddit every day, still thought about Reddit, could never turn that part of my brain off. Reddit spun out, became independent, but also started going through crises. And what I later learned when I came back is the strategy at Reddit was — ‘don’t change anything.’

But I mean I can see from the outside it’s like Reddit’s got to change a few things, right? 

They definitely need a content policy. We’ve got these troll subreddits who are just wreaking havoc. And the product’s not evolving. And so, ultimately I came back to the company with a mission to help Reddit live up to its potential.

Envisioning Reddit’s next phase

HOFFMAN: Are there any kind of set of principles that, or ways of thinking about that or what the lessons of scale were there?

HUFFMAN: Oh, gosh. Look, I could start with quickly, you know, what didn’t change. We believed in community. We still do. We believe in the agency of people. We are aspirational about people. We believe we should follow our users. They create the content, they rank the content, they create the subreddits.

They self-organize, they set the tone. But then we also learned that you can’t just completely, you can’t sit on your hands. I talk about free speech as the founding principle of this country, and for a country, it’s essential. Democracy cannot exist without free speech.

And I think that’s a great place to start, right? A lot of our thinking about how Reddit works. Again, we go to the real world, and so a lot of things we look to democracies, in particular, the democracy in the U.S., and federalism is another important aspect of, like, we have our rules but the subreddits have their own rules, and separation of powers — all these things. But then, like, one of the big learnings is we are not a country.

You know, we are a community platform, and we’re also a business. This is a place that people have to want to be. By and large, you don’t choose where you live on Earth, but you can choose whether to be on Reddit or not, versus something else. And you can choose whether to work at Reddit or work somewhere else.

And so this kind of comes back to the idea of decorum. What do we want Reddit to be used for? Or what do we want Reddit to be not used for? Do we want to be proud of our work? Do we want to be proud of the role that Reddit plays in the role world and what it does for people? 

And so I think just bringing some of that common sense and practicality, and humanity into the company was really important in this kind of more modern era of Reddit.

Running a public company post IPO

HOFFMAN: Yeah, totally. Since we’re on, kind of, the story of Reddit to scale. So, you know, initial public offering — that’s obviously a huge step. How has that changed anything at Reddit? How you operate or how you present, or anything else? And how are things going?

HUFFMAN: Well, so we went out in March. We’ve had two quarters, two good quarters. I think if you ask me again if we have a bad quarter. Right now I feel pretty good. The thing is, though, we didn’t just go out in March. We thought we were going to go out in April 2022, but we didn’t because the world and market really changed in that era, which would have made it very difficult. But since April 2022, we’ve been running the company as a public company. So close the books fast, no mistakes, do our board meeting, do earnings calls with the public company analysts, they just weren’t public but we’re able to kind of just get into that rhythm. And so by the time we actually got out earlier this year, I think we were really well practiced, probably more so than the average company who IPOs.

We got really lucky. But now when I talk to any CEO who’s thinking about going public, I tell them: start doing earnings calls now. Like just get into that rhythm, learn what data you can and can’t share internally, and on what timeline. And there’s just kind of a little bit of a way of working and get your costs in order, level up the discipline and all these things. And so we did all those. And so, the company is immeasurably better than we were two years ago. Just so much more disciplined. Execution is as good as it’s ever been. We’ve been able to grow revenue and users basically without growing the team size, which basically means costs have stayed level while we scaled the rest of the company.

We had to learn how to do that, but we did. And I’m very, very proud of the work the team has done there. 

Now I think Reddit’s a little bit different in that prior to going out, we already had a bright spotlight on us. So we’d been in the press, you know, we read things we don’t like, or, you know, I say something silly and it turns into a headline or, you know, this or that.

So we’ve kind of had those experiences that kind of come with being a public company. So two quarters in, you know, I’m enjoying life.

HOFFMAN: That’s great.

HUFFMAN: Ask me again when the wheels fall off, but I feel great right now.

HOFFMAN: Well, hopefully, it’s only like one wheel gets a little low pressure.

HUFFMAN: Yeah, it gets a little wobbly, but like, I think, well, the other thing that worked out really well is we also did all of those things in choppy conditions. In choppy market conditions. So, find me a company that’s worse run today than it was two years ago, that’s gonna be hard.

But everybody had to really like, you know, the difficulty level went up. And so I think everybody got more disciplined.

Lightning Round: Innovative funding strategies?

HOFFMAN: No, that’s part of the good thing about, if you learn well from challenges, you can definitely get more robust and strong. All right, so lightning round. What’s the most innovative funding source or strategy you’ve used?

HUFFMAN: Reddit, it’s pretty much always done kind of traditional financing, though I’d say our investors, our largest investors, the ones who have worked out the best, are themselves nontraditional.

Folks who are really kind of true believers in Reddit, in the internet, in people. It’s really hard to see the future through spreadsheets. And so both, I think with finding investors and hiring people, the strategy that I found works really well is I keep meeting people until I find that values chemistry. And you can tell in almost like 10 minutes, it’s like, Oh, these are my people. And then I find deals come together really fast, hiring, we don’t just get them in the door. They end up staying with us for a long time. 

Lightning Round: Successful habits?

HOFFMAN: What is the habit that has helped you succeed the most?

HUFFMAN: I have a mantra that I use a lot now. Every day. The thought is, you know, you aren’t who you are, you are what you do. So, if you want to be somebody who works really hard, you have to actually work really hard.

If you want to be somebody who goes to the gym, you actually have to go to the gym. If you want to be a good friend, good spouse, good colleague, you have to actually do those things. 

Lightning Round: Who do you talk to when you feel stuck?

HOFFMAN: Yeah. It makes you philosophically very Aristotelian. You are what you do. When you’re feeling stuck about a big decision, who do you talk to?

HUFFMAN: Probably Jen. Jen Wong, our COO. So she joined us in 2017, on paper to run the business side of Reddit, so ads and sales. But, you know, she’s really my partner in running Reddit. And so we’re in near-constant communication. And, you know, I was talking before about that values alignment. You know, I had that with her instantly, but we have completely different backgrounds, right?

She came up kind of through the consulting media business world. And I, you know, came up as an engineer product person. And so, I mean, there are a couple of people like her at Reddit, but she’s probably the first one that I call.

Lightning Round: Favorite way to unplug?

HOFFMAN: No, it sounds like a great collaboration and partnership. Your work and your life and work is so online. What’s your favorite way to unplug and recharge?

HUFFMAN: I read a lot. I love my Kindle. Both because it’s a convenient way to carry on a bunch of books, but it’s also not online. So I read a lot. I’ve got kids now, play with my kids, and I enjoy going to the gym. So life is pretty simple. 

HOFFMAN: Yes, exactly. 

HUFFMAN: And I love it.

Lightning Round: Book recommendation?

HOFFMAN: What’s one book you think everyone should read?

HUFFMAN: Oh goodness, there’s Man’s Search for Meaning.

HOFFMAN: Frankl? 

HUFFMAN: Victor Frankl. Super famous book, he was a concentration camp survivor. And surviving a situation like that, you know, you have to have a reason to live. And so kind of the thesis of the book is you need to find a reason to live. And it doesn’t even matter what it is, but you do need to find it.

And make that your reason for living, that one really helped me get through a difficult period of my life where I was very stressed. And I was constantly chasing this idea of like, how do I eliminate stress? And I think it’s impossible to eliminate stress. And what that book kind of flipped around for me is the sort of stress that I have in my life is a result of the work that I do and the job that I have. Which is truly a privilege. I have a special job. I have my dream job and it’s hard. And so if I didn’t have that stress, what’s the alternative? I’d be bored. And so that book helped me basically say, thank you that I get to be stressed. As a result, I felt a lot less stressed. You know, it’s kind of funny how those things work out. That’s a big one.

Lightning Round: AI fantasy?

HOFFMAN: Last lightning round question. How would you like AI to change your future?

HUFFMAN: I just want to acknowledge I’ve given five-minute answers to every lightning-round question. 

HOFFMAN: No, it’s all good. Actually, they were just meant to be kind of a provocation. Not, not necessarily like 42 is the answer.

HUFFMAN: The stuff that I want AI to do, AI can do today. So I’m just waiting for Apple or Google to connect the dots, and I think we’re so close. I just wanna add items to my to-do list or take a thought and organize it. By speaking. Because, you know, I’m constantly going around through my day. I’ve got my phone in my pocket and I’d love to say, Remind me to do this thing or I have this idea.

I want to just you know, it’s kind of that whole getting-things-done idea of like get things out of your head into your system. I think AI can get things out of my head into the right place much more conveniently, right? The best UI for computers is now speaking. And that’ll be, when we’re actually there, we’re like a year away. 

HOFFMAN: Yeah, it’s very close. 

HUFFMAN: I don’t want to ask for much. I just want to put things on my to-do list.

HOFFMAN: All right. Well, it’s been a great pleasure and honor, Steve. Thanks for joining me on Masters of Scale.

HUFFMAN: It’s really been my pleasure and honor, Reid. Thank you so much for having me.

HOFFMAN: Steve Huffman’s scale story is a good reminder that not every unicorn IPO is some sort of hyperscaling, overnight success. It took Reddit almost two decades, multiple ownership structures, and a whole lot of sweat equity to get to where it is today. 

And all that hard work has given Steve a practiced hand for dealing with the site’s next chapter of challenges — from navigating AI to an increasingly fractured information ecosystem. 

I’m inspired by his unwavering belief in people and the meaningful connections that can form when the internet is at its best. I hope he can keep scaling those values too. 

I’m Reid Hoffman, thanks for listening.