CEOInterviews.AI
Start App
Steve Huffman
CEO & Co-Founder, Reddit

Reddit's CEO on why 'we're meant to die' and other things he's right about (Steve Huffman)

📅 May 05, 2026 Fifty Years 33 MIN 189 SEGMENTS · 2 SPEAKERS
Steve Huffman is the Co-founder and CEO of Reddit, a network of online communities that is now one of the internet's primary ...

What Steve Huffman said

Written from the verified transcript and checked against it. Every figure links to the moment it was said.

Steve Huffman, CEO and co-founder of Reddit, discussed his belief that the future will resemble the past, arguing that humanity's mortality is essential for societal turnover and that efforts to extend life indefinitely could have negative repercussions. He predicted science will eventually validate astrology, and that technology will increase human connectivity, potentially through engineered experiences like those of the Hogan twins. Huffman argued that institutions, including the IRS, are static and need to be more dynamic, but that abrupt changes could be destabilizing. He criticized treating supply problems as pricing problems, citing rent control and medical licensing as examples. He expressed support for universal basic income, citing Milton Friedman, and believes people will continue to work even with UBI. Huffman also discussed AI consciousness, admitting he treats AI with respect after a jailbreak experience, and shared his view that AI will not fundamentally change parenting concerns.

Key takeaways

  1. Huffman believes humanity should accept mortality, as eternal life would prevent generational turnover in politics and ideas.
  2. He supports universal basic income, citing Milton Friedman, and thinks people will still choose to work.
  3. He criticized the AMA for limiting doctor supply, leading to higher prices and restricted care.

Numbers and commitments

FigureWhat it refers toTypeAt
$30 billion Reddit's market value metric 0:00
400 years Average lifespan if all unnatural causes of death are eliminated other 26:38

Chapters

  1. 0:00Mortality and societal turnover
  2. 4:37Science validating astrology
  3. 6:21Human connectivity and technology
  4. 10:33Cities and scaling efficiency
  5. 12:34Institutional stasis and the IRS
  6. 17:09Supply vs pricing problems
  7. 20:41Parenting and future uncertainty
  8. 22:46UBI and the future of work
  9. 27:21AI consciousness and moral duty

Questions asked in this interview

12
  1. 0:45What do we have to gain by doing that?
  2. 2:35So whatever ideas we have today that maybe are aberrant, shouldn't they die as well?
  3. 4:13Like you think we'll figure it out?
  4. 4:33So what is the thing that changes?
  5. 10:22Is that the purpose of technology?
  6. 12:28So sorry about how... so how are institutions doing?
  7. 15:15Sam Bankman-Fried's quote about this is great, which is like he doesn't think AI is going to change much about the economy because you can't automate fake jobs, right?
  8. 17:02Or what's one thing that if you could wave a magic wand, you'd change about the way our institutions work?
  9. 18:09Prices go up. You okay, bro?
  10. 21:41Do you feel like that's materially different from the way other parents have felt sending their kids into the world?
  11. 22:15Do you think your kids are going to have jobs?
  12. 28:13Right. And are the AIs happy?
Scott Phoenix 0:00 ↗
I live for the day where the scientists have to apologize to the hippies. Steve Huffman is the mayor of the internet, also known as SPEZ. He founded Reddit, the most popular online community in the world and a $30 billion company. The people who behave like [__] in one area are perfectly normal people in other areas. Steve thinks a lot of modern work is theater. Deleting the IRS would be very destabilizing. It was almost like a UBI program at this point. And Steve told me a fact about America that I had to look up to believe.
Steve Huffman 0:30 ↗
Did you know in the United States we limit the number of doctors?
Scott Phoenix 0:33 ↗
Wild.
Steve broke with most of Silicon Valley on what people should do with their time.
Steve Huffman 0:38 ↗
Just give people money and let them do what they will.
Scott Phoenix 0:41 ↗
Some people want to surf all day.
Steve Huffman 0:42 ↗
Do it. Let them surf.
Scott Phoenix 0:44 ↗
Let them surf.
Steve Huffman 0:44 ↗
Yeah.
Scott Phoenix 0:45 ↗
And unlike most Silicon Valley billionaires, Steve doesn't even want to live forever. What do we have to gain by doing that? We're supposed to die. I'm Scott Phoenix and this is Progress.
Steve Huffman knows what it's like to be hated. In 2015, Reddit was overflowing with people who were furious at Steve, also known as Spz, for how he was running the site. Every Reddit community was shouting the same rallying cry, 'Fuck Spez.' Instead of giving up, Steve kept his head level and stayed focused on making the site better. Today, Reddit is a $30 billion company and one of the most popular sites on the internet. This is progress in the mind of Steve Huffman, co-founder and CEO of Reddit.
Steve Huffman 1:29 ↗
My weirdest idea used to be that we are either the last generation to die or the first generation to live forever.
Scott Phoenix 1:38 ↗
But now everybody I know believes that. And it's so funny because now my weirdest idea is that we will all die and be forgotten.
Steve Huffman 1:46 ↗
And that's a good thing.
Scott Phoenix 1:47 ↗
Okay, say more.
Steve Huffman 1:48 ↗
It's how humanity and life has literally always worked. My view of the future is that it will look a lot like the past.
Scott Phoenix 1:57 ↗
Things don't change as much as we think they do.
Steve Huffman 1:59 ↗
Wow. You're born, you die, you're forgotten. There's so much energy into breaking that cycle. And I wonder like for what? What do we have to gain by doing that? I can think of some problems, which is if our generation never dies, think about politics.
Scott Phoenix 2:20 ↗
Oh yeah. If the boomers never die, then the millennials will never be in charge.
Steve Huffman 2:24 ↗
The millennials never die, then the zoomers will never get to be in charge.
Scott Phoenix 2:28 ↗
Yeah.
Steve Huffman 2:28 ↗
And so just that turnover in the political class, I think, is actually really important for society.
Scott Phoenix 2:33 ↗
Have you seen Altered Carbon?
Steve Huffman 2:34 ↗
No.
Scott Phoenix 2:35 ↗
The idea is that we figure out how to move consciousness from one body to the next. And the way this manifests for super rich people is that they have an infinite number of clones and they basically have to live forever. And they accumulate ungodly amounts of wealth. And it just really exacerbates the rich-poor divide. That's like a consequence of this. So what you're articulating basically is the Altered Carbon world where power is never released. You know they talk about science proceeding one funeral at a time. Power is never released. Ideas are never released. So whatever ideas we have today that maybe are aberrant, shouldn't they die as well? And the good ones will hopefully survive. But if I think we're artificially just drawing a line and saying, 'Okay, freeze it now.'
Steve Huffman 3:20 ↗
Yeah. These people and these ideas will always be the eldest. I think that will have a lot of negative repercussions. And I'm not sure what positive comes out of that.
Scott Phoenix 3:30 ↗
It's interesting. 15 years ago or some, you know, maybe 12 years ago when Vicarious was just a couple of us, we started having meetings with Alphabet about an acquisition around the time they bought DeepMind. We met with Larry Page and one of the things that frustrated him most about Google was he would sometimes overhear people say, 'Oh, that's not how we do things at Google. We need to make sure we keep it Google is what they would say.' And he just got so animated and he was like, 'This is not a [__] museum.' You know what I mean? Like we have to evolve and we have to let that evolution happen. We need aptosis like you're saying. We need sort of programmed cell death so that new, you know, life can be created.
Steve Huffman 4:12 ↗
It'll probably work out that way.
Scott Phoenix 4:13 ↗
Yeah. Yeah. There's no action needed. Well, at some point, so do you think in some way it won't work out that way? Like you think we'll figure it out?
Steve Huffman 4:19 ↗
I will propose that the laws of nature are so strong our puny computers, you know, gene editing and little ideas or whatever, not going to cut it. Life will find a way. It always has so far.
Scott Phoenix 4:33 ↗
So I'll take the field on this. So what is the thing that changes?
Steve Huffman 4:37 ↗
Okay. Okay. Okay. Here's my favorite one. Okay. Again, this is going to be nothing changes. I think through some better understanding of physics and quantum such and such and a better understanding of these cycles that we're all living through, science is going to explain why astrology works and is real.
Scott Phoenix 5:01 ↗
We'll finally know.
Steve Huffman 5:02 ↗
Yeah. We'll finally know. So the thing, so my thing that changes is that we will actually find out that the hippies were right all along.
Scott Phoenix 5:09 ↗
Yeah.
Steve Huffman 5:09 ↗
And I hope I live for the day where the scientists have to apologize to the hippies.
Scott Phoenix 5:15 ↗
I mean I think that like for a lot of ancient wisdom it is pointing to some deeper truth because that's why it persists. The fact that an idea is still around probably means it has some merit.
Steve Huffman 5:25 ↗
Yeah.
Scott Phoenix 5:26 ↗
You know one of my favorite books that I've only read the first 20% of. It's called The Tao of Physics. I love your... The Tao of Physics.
Steve Huffman 5:36 ↗
Okay.
Scott Phoenix 5:37 ↗
It was written in the 70s and the thesis of the book is the famous physicists whose names haven't really changed because physics hasn't really changed in the last whatever 50 years, they sound a lot like Eastern philosophers.
Steve Huffman 5:51 ↗
Yeah.
Scott Phoenix 5:52 ↗
You know, Eastern philosophy is often expressed in these kind of paradoxes, these paradoxical parables, which by the way, you can only express in Eastern languages, these symbolic languages. And by the way, the language of math and physics are symbolic languages. We've invented all of these symbols for these concepts that are hard to describe. But if you look at the words of these physicists, right, the Einsteins and Bohrs and Heisenbergs and whatever, they talk about the interconnectedness of all things.
Steve Huffman 6:21 ↗
Oh, yeah. The cyclical nature. It's like you almost can't even tell the difference. Are you reading like somebody from 5,000 years ago or 50 years ago studying physics? One thing I'm very excited about the future is that we might reveal that we might actually understand with the other half of our brains what's been true all along. There's a way to take that as the nothing changes and the way to take it as the something changes is the completion of what has always been prophesized. My favorite version of this that I think would be really remarkable if it did come true is if you take what Max is building at Science or what Neuralink's working on or this, you know, the new Sam Altman company, Merge, all of these people are circling around this idea that maybe you can take what nature already figured out in the Hogan twins. You remember the Hogan twins? Hogan twins are two twins that are still alive. They're conjoined twins, conjoined at the thalamus. They're two people, but they can share experiences and see through each other's eyes. They're separate, but they're also one. And if nature stumbled into it, it also means it could conceivably be engineered.
Scott Phoenix 7:29 ↗
The tree in Avatar where you like plug in and you can experience that oneness.
Steve Huffman 7:34 ↗
We become more connected.
Scott Phoenix 7:35 ↗
More connectivity. Exactly. Yeah.
Steve Huffman 7:37 ↗
And there's an ability that could be engineered. Right now, you have that level of closeness with your children, with your spouse, maybe. And then it kind of starts to degrade the further out you get.
Scott Phoenix 7:51 ↗
And if it didn't have to degrade quite so fast and you could feel a greater sense of connectedness and empathy with, you know, wider and wider circles, then perhaps it would be easier for us to come to consensus on ideas. I can see that for sure because I see that through Reddit. One of the things I love about Reddit is that it gives me a glimpse of the world through somebody else's eyes. And so I love following subreddits for hobbies that I don't have or for careers that I don't have. And so I'm in the one for Uber drivers and Kitchen Confidential for chefs. That one blew up, but that one was really great for a while. I mean, it's still great for pilots, for air traffic controllers, for like policemen. It's like really interesting just to see other people's lives and it's made me I think much more empathetic.
Steve Huffman 8:39 ↗
Yeah. It's like the 15 cups of tea or you know, some number of cups. Are you familiar with this book? There's a woman who or a man, I'm not sure which, who wrote a book about like if you've had a dozen cups of tea with someone, it's actually very hard to hate them.
Scott Phoenix 8:52 ↗
What's so interesting is we've seen this on Reddit as well, the people who behave like [__] in one area are perfectly normal people in other areas. Honestly, you sound like if anybody's talking about politics, they sound like an [__] to somebody, right?
Steve Huffman 9:06 ↗
But then if they're talking about sports or TV shows or video games, they're perfectly normal.
Scott Phoenix 9:11 ↗
Yeah. What we should actually study, I bet we can find this. This would be really cute to do. I bet we can find two people arguing over here and getting along over here. Oh, that would be interesting. I remember there was a time, I don't know if you saw this but Facebook had a page where it just was analytics and you'd see analytics on friendships between people in India and Pakistan or friendships between groups that disagreed with each other, Christians and Muslims or something. It would be showing all of these different pairings and it was just such a powerful like visualization of unity and human connection and that's not really what Facebook ended up becoming but at the time I was like wow this is like so cool what a symbol of what could be. I do think that is the purpose of the internet if we could realize it is to make those connections right regardless of geography or language or culture like bring people closer together and when it works it's so amazing thing.
Steve Huffman 10:08 ↗
I mean, isn't that in some sense the purpose of all technology since agriculture? You know, like we are creating larger and larger circles of those who we can care about and identify with and connect with over longer and longer time scales and physical distances.
Scott Phoenix 10:22 ↗
Is that the purpose of technology?
Steve Huffman 10:24 ↗
Maybe it's the effect of it.
Scott Phoenix 10:26 ↗
Steve thinks for institutions to get stronger, they have to die first. The idea comes from how cities scale.
Steve Huffman 10:33 ↗
When I think of humanity and communities, I immediately go to cities. And so does technology allow cities to be bigger or could cities be big without technology? I think one of the interesting things that I've learned is like the scale of a city is limited by its transportation network. People will commute about an hour a day. And if you can make that... So yeah, if they're walking, that's it. Going to be a couple miles wide. If it's a train or whatever, a plane that's easy or whatever, you can actually get much bigger. But people whatever. And our programming is about an hour of transit a day. And so if technology can make either things more dense so you have to travel less or travel faster, you can have bigger cities.
Scott Phoenix 11:15 ↗
That's super interesting. I never knew that.
Steve Huffman 11:16 ↗
Cities are one of the only organisms, by the way, that scale super linearly. So animals scale sublinearly. The bigger we get, the less efficient we get. Right? A mammal can only be so big before it crumbles under its own weight.
Scott Phoenix 11:32 ↗
Mhm.
Steve Huffman 11:32 ↗
Companies scale sublinearly, right? Communication overhead, right? They just get less and less efficient. So, by extension, governments too, but cities come back to that. Cities scale super linearly. They actually get more efficient as they get bigger. You need fewer lawyers per capita as a city gets bigger, fewer restaurants, fewer, you know, hairdressers. And so, they actually get more efficient as they get bigger. And so a lot of what I've thought over the years is how do we make Reddit scale super linearly. The conclusion I've come to so far is companies are companies. So you just got to stay small.
Scott Phoenix 12:13 ↗
Okay.
Steve Huffman 12:14 ↗
But the platform is more like a city. And so the platform can scale super linearly if we do our job right. And the company I think we just have to resist the law of nature which is stay small because otherwise it will just become less efficient.
Scott Phoenix 12:28 ↗
Yeah. Okay. Let's... you've set it up. So sorry about how... so how are institutions doing?
Steve Huffman 12:34 ↗
Our institutions are institutional. That's kind of... the answer is in the name. What is the etymology of the word institution? Did we name our institutions that because they never change or do we use the word institutionalized to describe things that never change? There's the African parable of every name should be like both a challenge and a promise.
Scott Phoenix 13:00 ↗
And you know with... Exactly.
Steve Huffman 13:03 ↗
Yeah. I mean I think our institutions, by and large, you know, they're static. As somebody who works in technology, I think I want them to be more dynamic. However, sometimes I also think the stasis is a feature. So how are they doing? Sometimes it's frustrating, but maybe that's the point.
Scott Phoenix 13:26 ↗
Where do you feel like rebirth is needed most needed? Because we talked about like how apoptosis and death is like a necessary... Shouldn't that also be true of our institutions?
Steve Huffman 13:38 ↗
I think it should. Companies get replaced. They should at least... people get replaced. The culture gets replaced or evolves. And so the institutions, they don't... that's one of the challenges with government is they kind of sit outside sometimes of the free market and so there's not really institutional turnover and that is probably a challenge.
Scott Phoenix 14:01 ↗
What you say really, like we end up with sort of zombie institutions but what is the force that helps those institutions transform or die or be reborn or...
Steve Huffman 14:12 ↗
I think it's a really hard problem. I can give you an example of a solution that would just never happen. Okay, let's pick an institution. The IRS.
Scott Phoenix 14:19 ↗
I was going to say the FDA.
Steve Huffman 14:20 ↗
Okay. So, safe, right? The only people mad at me right now actually are the people who are about to audit me, I guess. So, maybe that was not so safe.
Scott Phoenix 14:27 ↗
Bad choice.
Steve Huffman 14:28 ↗
Yeah. Why do we need the IRS? Like, that's such a computer job. Such a computer job these days. But it employs a lot of people, not directly and indirectly. And so, just deleting the IRS would be very destabilizing.
Scott Phoenix 14:45 ↗
I feel like this comes up a lot in these conversations like this institution needs to go, this thing needs to change, but it's going to screw a whole generation of people. The easy to say, impossible to do would be, well, we should just pay them.
Steve Huffman 14:58 ↗
It's like here, okay, you have jobs for the rest of your life. Or you have a salary for the rest of your life, but your job is now deleted.
Scott Phoenix 15:06 ↗
Then in a generation, even though it's very expensive, in a generation, you just... you created... you've healed.
Steve Huffman 15:12 ↗
Yeah. So, I think the IRS is almost like a UBI program at this point.
Scott Phoenix 15:15 ↗
Sam Bankman-Fried's quote about this is great, which is like he doesn't think AI is going to change much about the economy because you can't automate fake jobs, right? And most jobs are fake in his worldview.
Steve Huffman 15:26 ↗
Well, I know there are fake jobs in the government.
Scott Phoenix 15:28 ↗
Is your job fake?
Steve Huffman 15:29 ↗
I don't think so. Okay. I've got a decent AB test on me cuz I was gone for a while.
Scott Phoenix 15:33 ↗
You were gone for a while. That's right.
Steve Huffman 15:34 ↗
If there was a job I knew off my head at Reddit that was completely fake, obviously I would eliminate it before this podcast was over. But I think a lot of what we do during the day is fake and that adds up to lots of hours of fakeness. You know, this meeting could have been an email or this email didn't need to happen or we already made this decision. Why are we making it again? Wow, this is a really nicely formatted doc. This is my pet peeve. Like when I see a doc, it's like, well, this is a very nicely formatted doc. I wish you had put so much thought into the words. And so if that's happening in a cutthroat industry where our survival is anything but guaranteed, it has to be happening in an institution where survival is literally law.
Scott Phoenix 16:21 ↗
But then why was the DOGE so unsuccessful or seemingly so unsuccessful?
Steve Huffman 16:26 ↗
Well, that's why the immune system is too strong. Basically, the immune system is too strong. Like politics exists, like you have to bring people along.
Scott Phoenix 16:33 ↗
Yeah.
Steve Huffman 16:34 ↗
Right. That's a part of leadership. But as soon as you're coded as right or left, instantly 50% of people disagree with you.
Scott Phoenix 16:41 ↗
Yeah. It's impossible to do anything.
Steve Huffman 16:43 ↗
And so I do think we need... I hope... here's my hope for the future. This has happened. Part of the cycle is we do occasionally get these transcendent leaders who can rise above the right-left arguments of the day and actually get something done.
Scott Phoenix 17:02 ↗
So, what's a policy that you feel like could be improved? Or what's one thing that if you could wave a magic wand, you'd change about the way our institutions work?
Steve Huffman 17:09 ↗
We treat supply problems as pricing problems.
Scott Phoenix 17:13 ↗
Say what you mean.

94 more exchanges in this transcript

Sign in free to read the rest of this interview. No card required.

Sign in to read the full transcript

Cite this transcript

APA, MLA, BibTeX
APA

Huffman, S. (2026, May 5). Reddit's CEO on why 'we're meant to die' and other things he's right about (Steve Huffman) [Interview transcript]. Fifty Years. CEOInterviews.AI. https://ceointerviews.ai/interview/887930/

MLA

Steve Huffman. "Reddit's CEO on why 'we're meant to die' and other things he's right about (Steve Huffman)." Fifty Years, 5 May. 2026. Transcript, CEOInterviews.AI, https://ceointerviews.ai/interview/887930/.

BibTeX
@misc{huffman2026_887930,
  author       = {Steve Huffman},
  title        = {Reddit's CEO on why 'we're meant to die' and other things he's right about (Steve Huffman)},
  howpublished = {Interview transcript, Fifty Years. CEOInterviews.AI},
  year         = {2026},
  month        = {may},
  url          = {https://ceointerviews.ai/interview/887930/},
  note         = {Speaker-attributed transcript with timestamps}
}