The Exchanges

Q Not at all, but I do want to start staying on you, and it's easy to get lost in the incredible journey that is Reddit, but I do want to get back to the very beginning. How did you make your way into the world of startups, and what was that founding moment with Reddit, Steve?

A I don't know if there was a founding moment. It's more of a founding really long story, but I'll give you the, give you the high points, which is, Alexis and I were college roommates, and we were going to start a company together called My Mobile Menu, which was, it would allow people to order food from their cell phones, which seemed novel at the time, but it was really, I think, an idea ahead of its time because the smartphones didn't exist, and most restaurants were not online, and so there's a lot of reasons why it wouldn't have worked then, but we applied the Y Combinator with that idea because we had met Paul Graham, and we were optimistic because Paul had given us initial feedback that was really positive, but ultimately, We were not accepted into Y Combinator. And the way Reddit came about was basically Paul told us, hey, we really like you guys. We just don't like that idea. So if you're willing to work on something else, you can be in YC. And so through the ensuing conversations about what else we could work on, the idea of Reddit was born. That is, we'd build a website where people could find new and interesting content.

AI assessment note: “through the ensuing conversations about what else we could work on, the idea of Reddit was born.”

Q What was that like? I mean, we see this kind of historic monument in YC State. What was that first batch like?

A It was like the early days of any startup. YC barely knew what they were doing. You know, they had maybe a vision. I think Paul even described it then as an experiment to see whether something like this could even work. And there were, I think, eight startups in that batch. About half that batch, about half the people on that batch, I'm still fairly close with. I and Alexis with Reddit. There was Chris Slo and Zach Stone doing Memamp. Chris Slo is our CTO. There was Justin and Emmett, who were the founders of the time Kiko, but then Justin TV and ultimately Twitch. Justin's little brother, Dan Kahn is, is the guy who connected us. Sam Altman is on Reddit's board and led one of our investment rounds, was in that batch. Aaron Swartz was in that batch and was with us at Reddit for about a year. And so, it was a pretty tight-knit group, and looking back on it, a lot of those characters are really significant people in my life, even today.

AI assessment note: “It was like the early days of any startup. YC barely knew what they were doing.”

Q Not at all, but I do want to start staying on you, and it's easy to get lost in the incredible journey that is Reddit, but I do want to get back to the very beginning. How did you make your way into the world of startups, and what was that founding moment with Reddit, Steve?

A I don't know if there was a founding moment. It's more of a founding really long story, but I'll give you the, give you the high points, which is, Alexis and I were college roommates, and we were going to start a company together called My Mobile Menu, which was, it would allow people to order food from their cell phones, which seemed novel at the time, but it was really, I think, an idea ahead of its time because the smartphones didn't exist, and most restaurants were not online, and so there's a lot of reasons why it wouldn't have worked then, but we applied the Y Combinator with that idea because we had met Paul Graham, and we were optimistic because Paul had given us initial feedback that was really positive, but ultimately, We were not accepted into Y Combinator. And the way Reddit came about was basically Paul told us, hey, we really like you guys. We just don't like that idea. So if you're willing to work on something else, you can be in YC. And so through the ensuing conversations about what else we could work on, the idea of Reddit was born. That is, we'd build a website where people could find new and interesting content.

AI assessment note: “through the ensuing conversations about what else we could work on, the idea of Reddit was born”

Q What was that like? I mean, we see this kind of historic monument in YC State. What was that first batch like?

A It was like the early days of any startup. YC barely knew what they were doing. You know, they had maybe a vision. I think Paul even described it then as an experiment to see whether something like this could even work. And there were, I think, eight startups in that batch. About half that batch, about half the people on that batch, I'm still fairly close with. I and Alexis with Reddit. There was Chris Slo and Zach Stone doing Memamp. Chris Slo is our CTO. There was Justin and Emmett, who were the founders of the time Kiko, but then Justin TV and ultimately Twitch. Justin's little brother, Dan Kahn is, is the guy who connected us. Sam Altman is on Reddit's board and led one of our investment rounds, was in that batch. Aaron Swartz was in that batch and was with us at Reddit for about a year. And so, it was a pretty tight-knit group, and looking back on it, a lot of those characters are really significant people in my life, even today.

AI assessment note: “It was like the early days of any startup. YC barely knew what they were doing.”

Q It's incredible to see that group of people in your life and the ecosystem, but you mentioned Daniel Kahn there, and I actually chatted to him before our show, Steve, and he said, not many people go back to the company they sold. I'd love to hear Steve's perspective on why did you return to Reddit post-sale?

A Sure. So, the timeline there, of course, is we started in Alexis and I left in 2009, and then I returned in 2015. And between 2009 and 2015, a couple of things happened. One, Reddit spun out from Condé Nast, our acquirer, and so they had grew a lot. Three, I started another company with a friend, Adam Goldstein, Hitmonk, and we were selling plane tickets. We were a travel search company. And four, I'm losing count now, four, Reddit was going through some really difficult times by 2015, and I had accumulated a lot of A perspective while I was away and felt I can help. And so when Reddit came to me asking if I would help, it seemed like a pretty good opportunity all around to try to help Reddit live up to its potential. But the way I felt in 2015 was Reddit has this, it's this incredible opportunity represents, has more potential than any company I can think of. And it feels like they're missing it. It feels like that opportunity is slipping through their fingers. And so that was the thinking. It's like, can we help Reddit grow into what I think Its destiny is.

AI assessment note: “when Reddit came to me asking if I would help, it seemed like a pretty good opportunity”

Q loss years it has been seeing you at the helm and seeing the rise of Reddit once again. I do want to ask, were there any takeaways from the Hitmonk days? The travel industry is, you know, bluntly fucking hard. Were there any takeaways for you from that Hitmonk journey and just what a beast of a market it is being travel that you've taken back with you to Reddit?

A Well, I mean, you've just said a lot of it. Hitmonk was a really challenging company to run because travel is a really difficult industry to survive in, let alone grow in. It's a duopoly between Expedia and Priceline. And our customers, our partners, were both people like Expedia and Priceline, who would just prefer we don't exist, and airlines, who don't have any margin to share. But they're perpetually going out of business themselves. So even right there, we're off to a rough start. And one of the things about Hitmonk, one of the most important lessons I learned at Hitmonk, is that I am not our customer. So what I mean by that is in the early days of Reddit, Our product development philosophy was we would build stuff that I wanted, basically. And my reasoning was I'm not unique. There are millions of people just like me. And if we build stuff that I like, then millions of other people will also like it. And that worked, to be fair. The reality is there are millions of people just like me with similar tastes and whatnot, but there aren't hundreds of millions or even, of course, not billions of them. And so at Hitmonk, we did the same strategy. Adam and I built stuff, product that we loved and that we wanted to use. And in fact, I still use it, and I presume Adam does as well. Adam was the CEO and my co-founder. But what we learned, and we learned this lesson the hard way, is …

AI assessment note: “one of the most important lessons I learned at Hitmonk, is that I am not our customer.”

Q loss years it has been seeing you at the helm and seeing the rise of Reddit once again. I do want to ask, were there any takeaways from the Hitmonk days? The travel industry is, you know, bluntly fucking hard. Were there any takeaways for you from that Hitmonk journey and just what a beast of a market it is being travel that you've taken back with you to Reddit?

A Well, I mean, you've just said a lot of it. Hitmonk was a really challenging company to run because travel is a really difficult industry to survive in, let alone grow in. It's a duopoly between Expedia and Priceline. And our customers, our partners, were both people like Expedia and Priceline, who would just prefer we don't exist, and airlines, who don't have any margin to share. But they're perpetually going out of business themselves. So even right there, we're off to a rough start. And one of the things about Hitmonk, one of the most important lessons I learned at Hitmonk, is that I am not our customer. So what I mean by that is in the early days of Reddit, Our product development philosophy was we would build stuff that I wanted, basically. And my reasoning was I'm not unique. There are millions of people just like me. And if we build stuff that I like, then millions of other people will also like it. And that worked, to be fair. The reality is there are millions of people just like me with similar tastes and whatnot, but there aren't hundreds of millions or even, of course, not billions of them. And so at Hitmonk, we did the same strategy. Adam and I built stuff, product that we loved and that we wanted to use. And in fact, I still use it, and I presume Adam does as well. Adam was the CEO and my co-founder. But what we learned, and we learned this lesson the hard way, is …

AI assessment note: “one of the most important lessons I learned at Hitmonk, is that I am not”

Q of the thought process of a more mature kind of serial entrepreneur, serial CEO. But if we think about, like, you specifically in terms of your leadership, how have you seen how you think about what being a great CEO yourself is and how you act when you compare it from Reddit Gen One to Hitmonk to Reddit Round Two now? How do you think you've changed as a CEO?

A I've changed in, gosh, so many ways. And part of it, I think, has less to do with my job or the CEO role and more just Maturing, I think as a person and as an adult, I mean, I could give you a dozen examples. Talked about the stress management part, just simply not doing things or trying to kill the behaviors that cause me constant stress, big or small. Another is not holding on to emotions, not holding on to things like anger, for example. When I was younger in my twenties, I used to hold grudges and it was almost something like a point of pride for me. Like I'm never going to forget about this slight And when I look back on that, I just think, wow, what a waste of time and energy to do that. To actually, like, set aside a little piece of my brain to remember that this person screwed me once upon a time. And by the way, that probably wasn't even true. It was probably some perspective I invented. So just letting all of that go made me more effective. The other stuff we've also talked about, which is making decisions quickly. I think as a leader of any team, two really important duties that we have are Are set the strategy and set the guardrails, and then I guess a sub point is ending out of the way. Make it really clear where we're going, what is acceptable, what is not acceptable, and then give the rest of the autonomy to the team to figure it out. And the mistakes that I've m…

AI assessment note: “I've changed in, gosh, so many ways.”

Q I'm super interested. You said that about know your customer, and it really brings me to actually one of the questions that I was Literally just about to ask, and it was around art versus science when it comes to product feedback, and so how do you think about constructing your product feedback loop structure today with Reddit, given that experience with Hitmonk, and what does that look like?

A It's a great question. So I think about the data that influence our design decisions and strategy is coming in three parts. The first is the data, like what are people actually doing? What are the results of our experiments? How are they using the site and the apps and all of that stuff? But that's the raw data. And I think that stuff is really, really important. So it's probably some of the most difficult at Reddit because there's so much of it. You can probably tell just about any story that we want to tell. The second is what our customers are actually saying. So the, the qualitative, I like this feature. I don't like this feature. I wish you had that feature. That information is really important. And I think it requires some finesse to interpret because users don't always, they're not product managers. They're, they're customers, they're users. And so Some, I think, amount of skill is required to interpret their data into what might be actually relevant for, for our teams. And the final dimension is intuition. What do we think would be good for the users or strategically important to the business, and looking further into the future of what are the sorts of things that we should be building. And so this is raw data, customer feedback, and intuition. And I think it's important to have a healthy balance of all three of those things. And the mistake I see people make often, an…

AI assessment note: “raw data, customer feedback, and intuition. And I think it's important to have a healthy balance”

Q Can I ask, when it comes to decision making, you have some amazing people around the table. How do you think about idea sharing, thought sharing, and kind of decision making decentralization almost versus actually too many voices can spoil the decision and you need to be concentrated in the decision making effort? How do you get the balance of external perspectives, but also clarity and decision making?

A Sure. I think we just kind of touched on two things there. One is decision making, and then second is the kind of idea generation and debate around those decisions. I think it's important that there be a decider, a single decision maker, and that decision be made in time and very clear. I don't believe in decision through consensus at all. I think that's a great way to actually not make decisions. But leading up to that moment where a decision is made, collecting feedback from lots of people Bouncing ideas off of people, having conversations, formal and informal. Whenever, like, for example, I have an idea about a product direction we should be going, I just don't stop talking about it. I find an excuse, practically, sometimes consciously, usually subconsciously, I find an excuse to bring it up in just about any conversation I'm in, and have a conversation about it, and collect feedback, and continue thinking about it, until ultimately, I'm ready to do a little bit more of a formal product. Hey, here's what I'm thinking. I want feedback from you 10 people by next week. And then at that point, I'm going to make a decision on this. But I think making a decision has to be ideally done by a single person and very, very clear.

AI assessment note: “I think it's important that there be a decider, a single decision maker”

Q reasons. I'm sure that's a lot to take in. I'm going to give you a minute just to absorb what I've said. To me, that sounded really cold and quite hard in its kind of explanation. Would you agree with the mentor? And what advice would you give to me and to other founders listening in terms of the right way to communicate both effectively, but also kind of compassionately?

A I tend to agree with the mentor. And I think to use your words, I think that's both effective and compassionate. The most compassionate thing that we can do for somebody is tell them the truth in a difficult situation. It feels good in the moment, but the objective here isn't to feel good. The objective here is to have an honest conversation and not have the conversation feels really bad in the long run, and I think it's borderline cruel, um, because now that person is going to go into their next job not having that context, not knowing how they can be better, and that usually comes with, let's say if I'm in that situation, that also comes with me lying to myself, right, and so that means I'm going to go into my next hire having not acknowledged where I messed up, where things I could have done to be better in that situation, so it just It's a missed opportunity all around. Now, I have let people go before where the conversation is more like, I don't think this is working, and they agree, and I say, like, we don't need to get into details. If you want to, we can, but I think we just agree that, like, neither of us is enjoying this relationship. I don't think it's a requirement to, like, name everybody's faults and failures every time. Sometimes it's just, it's pretty obvious. But at least in the latter, it's being honest with ourselves about the situation, like, hey, you know w…

AI assessment note: “I tend to agree with the mentor. And I think to use your words”

Q No, that's really interesting. And my word, I think you've done an incredible job putting up with me and my questions then, if that was the case, but an ultimate one here, what do you believe that most around you disbelieve?

A Working at Reddit has gave me this, this gift of my outlook towards people in that I believe that not only are people fundamentally good, that people are overwhelmingly good. Reddit has given me a very optimistic view of the world because what I see when I look at Reddit is despite the ugly stuff that we have here or there, people misbehaving here or there, that people are more interesting and funny and collaborative, supportive than I think we give them credit for. It's so easy to have a cynical view of humanity, right? Just get behind the wheel of a car. And you know, the thoughts that go through my head are like, all of you people are like, you're going to kill us all. How did you get your license? Like, why are you doing this asshole thing? You're so rude, whatever. But then if I were face to face with those people over a meal, perhaps, or in another context, I'm sure we'd find something to talk about. Reddit's really good at people putting, putting people in that latter context where they can have a more human connection. And when people have that human connection, I think that can do really incredible things, and so I've learned that everybody has a lot more to offer than I think we might assume at first glance.

AI assessment note: “I believe that not only are people fundamentally good, that people are overwhelmingly good.”

Q Love it. No, it's a cracker. Tell me, you've said before, self-driving cars in cities are a waste of time. Why?

A Okay, so Dan's going to be really pissed at me for this one. So part of this, of course, consider the context, which is I live in San Francisco, which is a city without weather. I think cities, there's a couple of trends. That I think are important here. One is cities are legislating cars away. For example, in San Francisco, it's really difficult to drive downtown as a civilian. They just don't allow people to go through there. Two, the AI problem of actually driving a car through a city like San Francisco is so difficult. And I think by the time that these companies figure it out and make it safe and reliable and this and that, that there'll be other options that just work better. For example, electric bikes and scooters are really, really great. Public transit works really well. Personal drones and that sort of thing, I think, are on the horizon. So I just think cars in general in cities are probably the least efficient way of getting around in terms of the space and speed and time trade-off. And so there'll be other options, certainly in dense cities. Now, if somebody lives out in the country or even in suburbia where the computer science problem is a lot easier and there's more space required, they might make sense there. But in dense cities, I just don't see self-driving cars Being that useful.

AI assessment note: “by the time that these companies figure it out... there'll be other options”

Q I'm super interested. You said that about know your customer, and it really brings me to actually one of the questions that I was Literally just about to ask, and it was around art versus science when it comes to product feedback, and so how do you think about constructing your product feedback loop structure today with Reddit, given that experience with Hitmonk, and what does that look like?

A It's a great question. So I think about the data that influence our design decisions and strategy is coming in three parts. The first is the data, like what are people actually doing? What are the results of our experiments? How are they using the site and the apps and all of that stuff? But that's the raw data. And I think that stuff is really, really important. So it's probably some of the most difficult at Reddit because there's so much of it. You can probably tell just about any story that we want to tell. The second is what our customers are actually saying. So the, the qualitative, I like this feature. I don't like this feature. I wish you had that feature. That information is really important. And I think it requires some finesse to interpret because users don't always, they're not product managers. They're, they're customers, they're users. And so Some, I think, amount of skill is required to interpret their data into what might be actually relevant for, for our teams. And the final dimension is intuition. What do we think would be good for the users or strategically important to the business, and looking further into the future of what are the sorts of things that we should be building. And so this is raw data, customer feedback, and intuition. And I think it's important to have a healthy balance of all three of those things. And the mistake I see people make often, an…

AI assessment note: “I think about the data that influence our design decisions and strategy is coming in three parts.”

Q if you go really hard into data, at least you have objective proof to back up your reasoning behind why you did that. So if, you know, 10 out of 12 said they liked red, and you went for red, even though you thought it was wrong, you objectively have kind of scientific proof that you were right in your decision making. Do you think that's why people do it?

A I think that might be one reason. I actually think the reason that people go hard in data is because they're scared to trust their intuition, or they're scared of They're justifying their decision without being able to back it up. But going only in data is a great way to optimize to a local maximum. It's a great way to not see around a corner. I think this is true of investing as well. It's a great way to ignore the future. It's a great way to ignore what's actually happening. And I realize those are actually funny words to say, but if we just look at the data and try to make time on site go up or make the login rate go up, but our users are saying, hey, we don't really like this. Then in a year, we're not going to have any users. So I actually think going hard on data is the weakest of the arguments, even though it at first glance feels the most scientific or objective.

AI assessment note: “I actually think the reason that people go hard in data is because they're scared”

Q that single point of decision making, but some people do find decision making very stressful. And we chatted before about anxiety and stress in particular. Emotion all CEOs feel constantly, I'm sure, but the very best CEOs are able to manage and control their stress. And so talk to me about stress management. What were you like in the past? And how would you define your relationship to stress today?

A Sure. I mean, it's certainly evolved over time, and I think even using the decision-making process as an example is helpful. As you say, decisions can be stressful, and I think the stressful part is the, the part where it's the procrastinating part. Oh, I have to make this decision. I don't want to do it. Okay, maybe I'll get other people to do it, but I don't really like the way they're going, and oh, I'm just gonna, like, spin and spin. It actually just saves a lot of time and stress to make the decision and move on, and so there's this form, I think, of, like, procrastination causes stress. Actually, there's, that's a lesson right there. I don't think I've said those words before, but I do have this kind of thought process that goes through my head constantly. I used to be a procrastinator. Now I'm not. And the thing that's changed in my head is, well, I'm going to have to do this eventually. So I might as well do it now because if I don't do it now, I'm going to have to have this thought over and over and over again. And I don't want to have this thought. And so one of the big improvements I've made in my life over the last couple of years is I don't want to be thinking about this. What can I do to not be thinking about this? Because it feels like a waste of time. And so that usually means Whether it's a big important thing, make this decision at work, or put my laundry awa…

AI assessment note: “I used to be a procrastinator. Now I'm not.”

Q of the thought process of a more mature kind of serial entrepreneur, serial CEO. But if we think about, like, you specifically in terms of your leadership, how have you seen how you think about what being a great CEO yourself is and how you act when you compare it from Reddit Gen One to Hitmonk to Reddit Round Two now? How do you think you've changed as a CEO?

A I've changed in, gosh, so many ways. And part of it, I think, has less to do with my job or the CEO role and more just Maturing, I think as a person and as an adult, I mean, I could give you a dozen examples. Talked about the stress management part, just simply not doing things or trying to kill the behaviors that cause me constant stress, big or small. Another is not holding on to emotions, not holding on to things like anger, for example. When I was younger in my twenties, I used to hold grudges and it was almost something like a point of pride for me. Like I'm never going to forget about this slight And when I look back on that, I just think, wow, what a waste of time and energy to do that. To actually, like, set aside a little piece of my brain to remember that this person screwed me once upon a time. And by the way, that probably wasn't even true. It was probably some perspective I invented. So just letting all of that go made me more effective. The other stuff we've also talked about, which is making decisions quickly. I think as a leader of any team, two really important duties that we have are Are set the strategy and set the guardrails, and then I guess a sub point is ending out of the way. Make it really clear where we're going, what is acceptable, what is not acceptable, and then give the rest of the autonomy to the team to figure it out. And the mistakes that I've m…

AI assessment note: “I've changed in, gosh, so many ways. And part of it, I think”

Q Can I ask, what did you do to gain confidence, or were there moments where you noticed step changes in your own confidence in backing your own intuition, be it it happened once and it was right, be it the data that once you went with was wrong, were there moments where you really gained confidence in your own intuition, and you'd advise founders on the same?

A Well, I mean, I probably did this backwards. When we started with Red, we didn't have any data. And we didn't know how to look at our data. We were fortunate in that I think our intuitions were largely correct, but that actually probably held us back to the data side. So I started on the extreme version of the other kind, which has its own pitfalls, right? Which is, you can be blind to your own arrogance. Because if you're right, if you assume that I'll be right again in the future. And that was largely the story of Hitmonk. It took us years to realize that our intuition wasn't correct. So my advice to founders would be really look at both. Like really honestly look at both. Recognize kind of which flavor you are and try to compensate for that weakness or blind spot.

AI assessment note: “So my advice to founders would be really look at both.”

Q reasons. I'm sure that's a lot to take in. I'm going to give you a minute just to absorb what I've said. To me, that sounded really cold and quite hard in its kind of explanation. Would you agree with the mentor? And what advice would you give to me and to other founders listening in terms of the right way to communicate both effectively, but also kind of compassionately?

A I tend to agree with the mentor. And I think to use your words, I think that's both effective and compassionate. The most compassionate thing that we can do for somebody is tell them the truth in a difficult situation. It feels good in the moment, but the objective here isn't to feel good. The objective here is to have an honest conversation and not have the conversation feels really bad in the long run, and I think it's borderline cruel, um, because now that person is going to go into their next job not having that context, not knowing how they can be better, and that usually comes with, let's say if I'm in that situation, that also comes with me lying to myself, right, and so that means I'm going to go into my next hire having not acknowledged where I messed up, where things I could have done to be better in that situation, so it just It's a missed opportunity all around. Now, I have let people go before where the conversation is more like, I don't think this is working, and they agree, and I say, like, we don't need to get into details. If you want to, we can, but I think we just agree that, like, neither of us is enjoying this relationship. I don't think it's a requirement to, like, name everybody's faults and failures every time. Sometimes it's just, it's pretty obvious. But at least in the latter, it's being honest with ourselves about the situation, like, hey, you know w…

AI assessment note: “I tend to agree with the mentor. And I think to use your words, I think that's both effective and compassionate.”

Q Love it. No, it's a cracker. Tell me, you've said before, self-driving cars in cities are a waste of time. Why?

A Okay, so Dan's going to be really pissed at me for this one. So part of this, of course, consider the context, which is I live in San Francisco, which is a city without weather. I think cities, there's a couple of trends. That I think are important here. One is cities are legislating cars away. For example, in San Francisco, it's really difficult to drive downtown as a civilian. They just don't allow people to go through there. Two, the AI problem of actually driving a car through a city like San Francisco is so difficult. And I think by the time that these companies figure it out and make it safe and reliable and this and that, that there'll be other options that just work better. For example, electric bikes and scooters are really, really great. Public transit works really well. Personal drones and that sort of thing, I think, are on the horizon. So I just think cars in general in cities are probably the least efficient way of getting around in terms of the space and speed and time trade-off. And so there'll be other options, certainly in dense cities. Now, if somebody lives out in the country or even in suburbia where the computer science problem is a lot easier and there's more space required, they might make sense there. But in dense cities, I just don't see self-driving cars Being that useful.

AI assessment note: “cars in general in cities are probably the least efficient way of getting around”

Q No, that's really interesting. And my word, I think you've done an incredible job putting up with me and my questions then, if that was the case, but an ultimate one here, what do you believe that most around you disbelieve?

A Working at Reddit has gave me this, this gift of my outlook towards people in that I believe that not only are people fundamentally good, that people are overwhelmingly good. Reddit has given me a very optimistic view of the world because what I see when I look at Reddit is despite the ugly stuff that we have here or there, people misbehaving here or there, that people are more interesting and funny and collaborative, supportive than I think we give them credit for. It's so easy to have a cynical view of humanity, right? Just get behind the wheel of a car. And you know, the thoughts that go through my head are like, all of you people are like, you're going to kill us all. How did you get your license? Like, why are you doing this asshole thing? You're so rude, whatever. But then if I were face to face with those people over a meal, perhaps, or in another context, I'm sure we'd find something to talk about. Reddit's really good at people putting, putting people in that latter context where they can have a more human connection. And when people have that human connection, I think that can do really incredible things, and so I've learned that everybody has a lot more to offer than I think we might assume at first glance.

AI assessment note: “I believe that not only are people fundamentally good, that people are overwhelmingly good.”

Q Can I ask, what did you do to gain confidence, or were there moments where you noticed step changes in your own confidence in backing your own intuition, be it it happened once and it was right, be it the data that once you went with was wrong, were there moments where you really gained confidence in your own intuition, and you'd advise founders on the same?

A Well, I mean, I probably did this backwards. When we started with Red, we didn't have any data. And we didn't know how to look at our data. We were fortunate in that I think our intuitions were largely correct, but that actually probably held us back to the data side. So I started on the extreme version of the other kind, which has its own pitfalls, right? Which is, you can be blind to your own arrogance. Because if you're right, if you assume that I'll be right again in the future. And that was largely the story of Hitmonk. It took us years to realize that our intuition wasn't correct. So my advice to founders would be really look at both. Like really honestly look at both. Recognize kind of which flavor you are and try to compensate for that weakness or blind spot.

AI assessment note: “So my advice to founders would be really look at both.”

Q Basically, if you live in San Mateo, you are in luck. But I do want to ask, what would you most like to change about Silicon Valley and tech state? You're at the epicenter. What would you most like to change?

A I think one of our greatest strengths is also our greatest weakness, which is the arrogance that the world can be better. I think to be successful in business, I think requires some amount of like naive optimism, a belief that the idea in my head is better than the status quo. And then therefore the determination and the arrogance or whatever it takes to execute on that in the face of other people telling you that you're wrong. And I think that's our greatest weakness for obvious reasons, which is the blind spots that might come along with that. Because I think to be successful, you have to willingly sometimes ignore the counter arguments. And sometimes those counter arguments are true. Sometimes they're really important. The balance there, I think, is extraordinarily difficult to get right.

AI assessment note: “one of our greatest strengths is also our greatest weakness, which is the arrogance”

Q Basically, if you live in San Mateo, you are in luck. But I do want to ask, what would you most like to change about Silicon Valley and tech state? You're at the epicenter. What would you most like to change?

A I think one of our greatest strengths is also our greatest weakness, which is the arrogance that the world can be better. I think to be successful in business, I think requires some amount of like naive optimism, a belief that the idea in my head is better than the status quo. And then therefore the determination and the arrogance or whatever it takes to execute on that in the face of other people telling you that you're wrong. And I think that's our greatest weakness for obvious reasons, which is the blind spots that might come along with that. Because I think to be successful, you have to willingly sometimes ignore the counter arguments. And sometimes those counter arguments are true. Sometimes they're really important. The balance there, I think, is extraordinarily difficult to get right.

AI assessment note: “one of our greatest strengths is also our greatest weakness, which is the arrogance”

Q They absolutely don't. I do just have one more question just thinking through what I have to ask you, and it's over the next five years. Do you think we'll see the unbundling of human interest groups into specific verticals, or do you think that the core platforms which actually contain most of human interest in different groups in themselves will remain as kind of the bundled platform offering?

A Okay, so I think there are kind of two competing ideas in what you're asking there. One is The bundling or unbundling of the kind of the groups and then the platforms. And on the platform side, I would like to see an unbundling, right? I just generally think more is better and competition is good and it creates more surface area for creativity in the marketplace and allows us to solve a broader set of problems, a more diverse set of problems or challenges. But on Reddit specifically, I mean, that's how we think about Reddit specifically. I want Reddit to grow, but the way we grow is by having more communities. Having more diversity of communities and letting people self-organize into communities and around topics or interests that we almost certainly cannot predict ourselves. So I think in nature, diversity equals resilience, and I think that's true in business as well. And so while I want my business to be big and successful, I also think that for governments or for people, diversity in the ecosystem and in the platforms is really important.

AI assessment note: “On the platform side, I would like to see an unbundling, right?”