The Problem With Jon Stewart - How Algorithms, Money, & Bureaucracy Distance us from Democracy

Episode Date: October 3, 2024

With the election just over a month away, Americans are caught between a flood of political promises and the reality that we live in a time of political dysfunction. Joining us this week to explore the root causes are Ezra Klein, opinion columnist at The New York Times, host of "The Ezra Klein Show" podcast, and author of "Why We're Polarized," alongside Tristan Harris, co-founder of the Center for Humane Technology and co-host of "Your Undivided Attention" podcast. We examine how engagement-driven metrics and algorithms shape public discourse, fueling demagoguery and widening the gap between political rhetoric and public needs. Follow The Weekly Show with Jon Stewart on social media for more:  > YouTube: https://www.youtube.com/@weeklyshowpodcast > Instagram: https://www.instagram.com/weeklyshowpodcast > TikTok: https://tiktok.com/@weeklyshowpodcast  > X: https://x.com/weeklyshowpod Host/Executive Producer – Jon Stewart Executive Producer – James Dixon Executive Producer – Chris McShane Executive Producer – Caity Gray Lead Producer – Lauren Walker Producer – Brittany Mehmedovic Video Editor & Engineer – Rob Vitolo Audio Editor & Engineer – Nicole Boyce Researcher/Associate Producer – Gillian Spear Music by Hansdle Hsu — This podcast is brought to you by: ZipRecruiter Try it for free at this exclusive web address: ziprecruiter.com/ZipWeekly Learn more about your ad choices. Visit podcastchoices.com/adchoices

Hey, everybody. Welcome once again to, in addition of the weekly show podcast with me there, John Stewart. We are coming to you. This is probably coming to you Thursday, but I'm talking to myself right now on a Wednesday, following a Tuesday night of a debate between your, what do you call it there, J.D. Vance, who did a great job, I think last night of not being J.D. Vance. And Tim Walsh, who was very Walsish. and it was, I thought, certainly a bit more substantive, even though it was quite riddled with what I would consider abject falsehoods. But CBS decided we're not going to interrupt to say that's complete nonsense. We're just going to let this whole thing play out. And as long as it's nonsense stated reasonably, well, who's to say? but the one thing I will say is I am really tired of articles this morning saying here's what's wrong with the debate.

Yes, it was substantive but boring. And you're like, what the f like we're in a no win? Like nobody hit anybody. Nobody blanked out. I expect now for these things to be catastrophic for one or the other candidates. And even if it's a vice president to debate, I expect action. bloodshed. I expect

we shouldn't even do debates. We should just do those slap contests that they do. Let's just reduce them to where you stand on it and you slap it's hard and whoever goes down goes down and that's the end of it.

Maybe that's how we should choose our presidents for God's sake. But we do have, man, do we have a good pod for you today? Lauren Walker, of course, and Brittany Mehmedevich are fabulous producers. They'll be joining us at the end. But first, we are going to talk about

the nature of the threats to our fine government. And I hope everybody enjoyed the debate last night. And I hope you've made your choice about who you're going to vote for for vice president. And then later on, maybe you can make your choice about who you'll vote for president. But let's move on.

Let's get to our guests today. Folks, powerhouse program today. Our guests, we're delighted to have them. Ezra Klein, you know him. He's the opinion columnist in New York Times, host of the Ezra Klein Show podcast. You've seen it. It charts, baby.

And he's the author of Why We're Polarized. Tristan Harris, co-founder of Center for Humane Technology, which, boy, wouldn't that be nice? He is also the co-host of your undivided attention podcast. Guys, thank you both. So much for joining us. I wanted to talk to you guys today about this idea. You know, so I don't know if you guys are aware. There is an election in, it's got to be like a month in America, our country.

And one of the candidates is viewed as an enormous threat to democracy. His name is Donald Aloysius Trump. He, so we can view him as, you know, he's got autocratic tendencies. Maybe he's an authoritarian. Maybe he's just used to running the country like an episode of the, the apprentice where he just says that one dude, George and his daughter Ivanka, and they just go like, great job, boss. But I want to look at it from a different perspective today and where your guy's expertise, I think, would be incredibly valuable. Rather than looking at it as an individual who is a

threat to a democratic system, what if we look at, what are the shortcomings, discomforts of that democratic system that seed the ground for populist movements, demagogues, authoritarians? whether from the left or from the right. And can we view those fragilities within the democratic system as a way to protect ourselves, not from one person, but from these movements that tend to really polarize the country and swing the pendulum so far back and forth? Ezra Klein, we're going to start with you. Nice, modest, easy question here.

We got an hour. So the question is, what are? the fragilities in the system? Well, it's my experience with democracy is an analog system, and we live in a digital world. So it appears even slower in comparison to the way that the world is moving right now. Let's shorten the question. Citizens United, the amount of money that flows into the system. At this point, if you want any say in the system, and studies have shown this, if you've got money, legislation often reflects your desires. If you do not have money, if you are outside the system, legislation mostly does not reflect your need. Isn't that a fragility in our democratic system

that opens the door for the kind of authoritarian, anti-democratic type leaders? Yeah, it sure as hell doesn't help. So, solved. Done. So let me do one more twist on the, Which is what used to keep people like this out? There's a book called How Democracy's Die, written by two professors Zublott and Levitsky, I think. They make a point that has always stuck with me, which is that we've always had Donald Trump-like figures in American politics, right? You can think of Father John Coglin during the sort of New Deal era. You can think of Henry Ford, who was, you know, like spitting anti-Semitic filth in the Dearborn Independent. Lindbergh, right, Huey Long, right?

You know, Pat Buchanan in 88 and then 92, who's, I think, a really important forerunner to Donald Trump. What kept them out? Because they often were able to get what Donald Trump got in 2016, which is, you know, 30% of one of the parties. Two things. One was that parties had gatekeeping power. We used to run this at conventions. Primaries only became a thing in American politics that actually decided who won the nomination after 68. We have to become more undemocratic.

save democracy. Well, gatekeepers have a role. I'm not saying we're going to go back to it. But if you ask in another era, Donald Trump never wins a Republican convention. He's going nowhere. But the other piece is media. There was a lot more gatekeeping control in other eras of media. And in both of these cases, you can say this also keeps good things out, right?

The party elders, the media bosses, right? The funders. Donald Trump was not the best funded candidate in 2016, but he did have a lot of small donor donations, right? So he wasn't able to get the institutional. money, the big businesses as much, right? That was going to people like Jeb Bush and Marco Rubio, but he's able to raise money in ways you couldn't have done that effectively 60, 70 years ago. So we've made it much easier for new entrants to come in. That gets you somebody like Barack Obama on one level, makes it possible to have Bernie Sanders run a competitive candidacy in 2016,

and it also opens a door to figures like Donald Trump. And so you do have this constant tension in democracy, right, which is, you know, you can read about it in the Federalist papers, between how open you want it to be and what kind of checks you want it to have if some kind of populist demagogue rises as they have endlessly, repetitively through all of human history. It's an interesting, I tell you,

I hadn't thought about it that way. It's an uncomfortable conversation because at the heart of it is this idea that democratizing the process has opened us up to demagogues. And Tristan, let's talk to you about that. You know, he's referring to Henry Ford and Lindbergh. And this is also, Father Coglin, the advent of radio. And Tristan, is the introduction

of new forms of communication and media. Does that disruption also lend itself to demagogues kind of rising up the sort of now it's social media and AI and those kinds of things? I think you were saying earlier, John, that I was clinging into is the idea of an analog democracy, and now our democracy seems to be running on the digital world. And Mark Andreessen, there's many places I disagree with him, but he, you know, the co-founder of Netscape said software is eating the world. And I think that actually is very accurate, that software is eating media.

It is eating elections. It is eating children's development. What we're seeing with social media is it is eating the sort of life support systems. Oh my God. And I've got it on my phone. Tristan, what am I? I've got software on my phone. It's going to eat my children.

In what way? Go back and talk about what that means because that's a very interesting description. Well, I think, so this is kind of a, how much is tech running society? So people know culture is upstream from politics, but now tech is constituting culture. So how much of people's, you know, news consumption now is coming from social media. It's the vast majority. And on the one hand, you have President Biden saying we need to ban TikTok because, let's say it's a threat from the Chinese Communist Party. On the other hand, hand, he just joined TikTok a few weeks after that because you can't win an election today without actually being on TikTok. The same is true for the Republicans who also want to ban TikTok. They have to be on TikTok to win the next election. I think what that speaks to is the entrenchment and lock-in of if you can't win your next election except by joining the platform, it shows how important and significant that platform is at constituting again our cultural environment. We obviously wake up and spend most of our time in our lives looking at these devices.

And what people need to know about that, as we said in the Netflix film The Social Dilemma, is it's not about technology being neutral. It's that this entire complex that we are immersed in for hours per day is about these design choices that were not aligned with what makes democracy stronger. They were only aligned with an incentive that what maximizes engagement and attention. And we've sent society through 15 years of this sort of washing machine of, you know, spinning us out into a more addicted, distracted, polarized, sexualized, you know, society where those features are actually rewarded by that business model. We always refer to Charlie Munger who said, you know, Warren Buffett's business partner, if you show me the incentive, I will show you the outcome. And that's how we saw in 2013 that that's where this sort of social media wave would take us.

not to say that it's all new, as Ezra would say, we've had, you know, polarization and distrust in politics and all this for much more preceding social media, but social media is sort of like a jet fuel on that process. That's interesting. So I, this is one of my points of optimism, and I'm so rarely optimistic anywhere in the media. But I think there's a good argument that what you have when new waves of media emerge, new platforms emerge, is periods of destabilization before societies just build up a little bit of immunity. Right? So radio had this.

It was very immediate. It was very intimate in the way, you know, people always talk about podcasting being intimate. And it led to both great things, right? FDR's fireside chats and Hitler, right? The Nazis were geniuses at radio. And you can kind of go through this, that when new mediums and ways of communicating arise, for a while, the system, the society doesn't really know what to do with it. It doesn't sort of know what the tricks are.

It doesn't know how to control it. You have first adopters who end up in a weird place. Donald Trump in 2016 to me is a sort of golem grown in some mixture of Twitter and cable news. That's what he is. He's like Twitter like created a golem of itself. Yeah. Right. Trained on cable news.

That's right. So it's a Dr. Frankenstein kind of a situation here. And I mean, I'd be curious for your perspective on on this, John. But compared to 2016, even now, I feel like the media has not a sane, but a saner relationship to what is happening across things like, I guess now we call it X, Facebook, Instagram, even TikTok that it did before. Things rise. But the sense that everybody's there, like Donald Trump can just decide what the conversation is going to be about at any given minute. Things fractured. He's over on true social a lot of the time.

it's not a perfect system by any means. But by 2020, you sort of had Joe Biden, right? That was a move back towards people wanting normalcy. You got sort of the opposite of Donald Trump. If we had somebody who in 2016 dominated media by outrage and being in a negative way, extremely interesting, Joe Biden sort of ran a consciously boring campaign in 2020. That was strategic. It was smart. In a pandemic as well.

In a pandemic, too. And then here, it's a bit, you know, more mixed, right? And you had the sort of like the vibes rise immediately with Kamala Harris. and Tim Molls, but things have settled. Again, Donald Trump is himself nuts. You have a lot of nutty things happening in the election. But as somebody who is in the media during all these periods, we don't, I just think society has built up a little more immunity than it had then.

It doesn't think every viral tweet needs to be reported on as news. It doesn't. Things feel to me like they have not settled because the actual changes he has brought to the Republican Party are real and you have to cover them as real. but the derangement and even just how the thing like feels viscerally actually is a quite different texture. Ezra, you talked a little bit earlier about the thing that prevented these demagogues, you know, in the past, the Lindbergs and Ford and these types of people were certain guardrails. And Tristan, you're talking a little bit again about the guardrails and gatekeeping that occurred within the media. And so are we in some measure focusing too closely on the media environment and the communication environment and not enough on the legislative environment for which this occurs?

So I could make the argument, for instance, in like the 1930s, right? So you see disruption around the world after World War I. You see the rise of Bolshevism, socialism, these revolutions that occur. Workers of the World Unite. It takes hold in the United States, anarchists, all kinds of other, the wobblies. You know, there's all kinds of violence and things that are occurring within the United States and destabilizing the government. You add radio into that and Father Coglin and let's get the Irish and Italians and all that, you know, and there's all this mix. But at its heart, it was the Great Depression hit, and Roosevelt killed.

came up with a program that directly addressed the needs of the people. And without that, I don't think any of those guardrails would have meant anything. I think this country would have been in a much less stable position based on the government's ability to show the people that we see. Now, admittedly, it's a catastrophe. Depression is a catastrophe, and it shouldn't take that kind of catacly. to spur direct action. Okay, we'll be right back. Look, you know, the holidays are over.

Let's face facts. Christmas gifts, you got crushed. Why? Why do you have to buy a favorite? You just bought them last year. And it all adds up. Luckily, MidMobil is here to help you cut back on overspending on wireless. This January, I didn't even want to say it.

I don't even want to tell you. It's too good. 50% off unlimited premium wireless. MinMobile's end of the year sale, still going on. only until the end of the month, which is January. So get on it. Cut out big wireless as bloated plans and unnecessary monthly charges with 50% off. Three, six, 12 months of unlimited.

Nation's largest 5G network, high speed data, unlimited talk and text. You could use your own phone. God, how much more do you people want? How demanding? What is wrong with you? In mobile plan, bring your phone number. This January, quit overspending on wireless with 50% off unlimited premium wireless.

Plan started $15 a month at MintMobile.com slash TWS. That's mintmobile.com slash TWS. Limited time offer. Upfront payments of $45 for three months, $90 for six months, or $180 for 12 months. Plan required, $15 per month equivalent. Taxes and fees extra, initial plan term only. Over 35 gigabytes may slow when network is busy. Capable device required.

Availability speed and coverage varies. see mintmobile.com. We're back. If you were to look at what people needed in their lives, and I'm talking about the majority of people, and you look at how responsive the government is to those needs after 50 years of supply-side economics and a whole lot of citizens united and big money,

couldn't you make the case that the government is almost at odds with its population? I want this argument to be true. It is the way I want policy to work. It is a way I want to be able to respond to the threat of populism. Yes. We have a lot of evidence from Europe on right-wing populism. Because we've seen it rise a lot. And I see the difference between our systems to go back to guardrails for a second

is because Europe has a lot of multi-party systems. It's easier for intense groups to rise pretty quickly. In America, if you can't capture one of the two major political parties, you're nowhere. Right? So on the one hand, it's pretty hard to capture one of those parties or traditionally it has been. But if you do, then you have like half the system to yourself, basically. But in Europe, it wasn't that way. So we actually have like a lot to run the regression analyses on. And the state of the economy and how people are doing and what they tell you about their finances and how they're feeling about the world, it doesn't track the way you would think it would. It doesn't mean it's meaningless. And it definitely doesn't mean that immiseration or catastrophe or failure doesn't create a, breeding ground for these kinds of strong men. Like, I believe that Donald Trump is functionally a creation of George W. Bush's wars. I mean, I already said he's a Twitter golem, so I have lots of theories for Donald Trump. But I believe he's functioning of George W. Bush destroy the Republican Party's credibility

between the wars and the financial crisis. Barack Obama represented a changing and rapidly changing demographic country, right? In terms of race, immigration was changing in that period. religion to some degree were secularizing pretty quickly. And he sort of response to that. But in terms of the sort of like this is a kind of, I would call this like the policy feedback argument, right? You can pass good policy, you pass something like Medicare or Medicaid. What do we get after we pass Medicare and Medicaid? We get Nixon, who is a more Trumpian figure in many ways than a lot of other figures in American history. Although it would be considered

probably a liberal Democrat at this point in terms of policy. Yes, although not, I would say not in the way he wrote. But yes, in terms of policy. Yes. The thing that seems to correlate is rapid social change, right? When people feel power is shifting and particularly when immigration is going up a lot. And since 1970, the percentage of foreign-born residents of the country has gone from about four-ish percent to around 15 percent now. It's a very, very rapid rise historically.

We're secularizing and becoming majority minority. It's not the only things going on. The economic pieces of this are very real. But it doesn't seem that sort of running an economy. me well or sort of passing bigger social programs will end the threat of right-wing populism, or you wouldn't have it rising in the way it is in some of these European countries where they've passed a bunch of the policies we all want, or not we all want, but I think you and I and probably Tristan want to see past in America.

But then how do you explain Britain and France? As these social media programs are incentivized to conflict and outrage and fear and anger, doesn't the distance between what the population is experiencing as reality and the analog nature of government begin to grow to the point where these are giant pendulum swings, sometimes to the right, as it was with Maloney and a lot of the other things. And in Germany, you know, obviously they're fending off, you know, these really right-wing parties. But in Britain, we just saw it. swing in a huge way back to labor. And in France, I mean, they were in many ways saved from the party of Le Pen by the far left of France. So is that explained more by the way that voters experience

the world in social media? Obviously, there's so many different factors that are driving the political environment, including climate change, migration, economic factors. what I can speak to, the thing I have expertise in is how is the social media machine driving certain kind of weird distortions in our psychological environment. That's what I'm talking about. Yeah. And what we know from people who use Twitter a lot, for example, is the more you use Twitter, you would think that the more you use social media, the more informed you should be about

what other people believe. It turns out that the opposite is true, that the longer you are on social media, the worse you are at predicting what other people in your society, other tribes of the political tribes, if you're a Democrat, how good are you at predicting what Republicans would say into the statement that racism is still a problem in the U.S. today. The longer you use social media, the worst Democrats are at understanding or predicting what Republicans believe about that, and vice versa. Tristan, can I ask you a question about that? Is there any correlation to the longer you're on social media, the less you know the reality, like, and I take like the transgender

sports conflict as this part. So if you're on social media, do you suddenly have the feeling that like, oh my God, there are no girls sports left in high school. They are being utterly dominated. And as somebody who has kids in high school, or like, what does it also warp, not just what you think other people would say, but your own reality of what's happening. What's something that's so obvious but also so subtle is the way that these news feeds are personalized. So whatever is the boogeyman that gets you up at night or freaks you out or makes you angry, it will keep showing you a personalized feed and infinite evidence of that bogeyman taking over society. So if that boogeyman is the transgender sports movement and that's something that you click on a couple times,

how does the algorithm work? It says, oh, there's these keywords in these tweets that this person keeps clicking on, and it includes the word transgender and sports. And so it just gives you way more things like that. And so you end up thinking that this is this massive issue. It's taking over the world. It's the most important thing going on in the world. And for every issue, it's doing that for everybody, but into a different Truman Show,

into a different bespoke reality. And it's so obvious we all know that. But it's so subtle, I think, in the way that fragments are shared ability to have conversations because when I talk to someone who's been living inside of that reality, they'll give me millions of facts or data points or news articles about things that have been happening in that world. And I might say, I'd never even heard of that. I'm not even, you know, I don't know what's going on in that world. Wow.

And it really does shape it. But then getting back to you, Ezra, if you're living in that reality, then no matter what governments do, that's not going to change unless it's directly addressing that, as Tristan said, your personal boogeyman. But Ezra, the thing that, you know, when you were talking about, it's the way you wish it would work. boy, that puts us in a really difficult position because, you know, for someone who still believes government has a role to play in the improvement of people's lives or as just a kind of a check against corporate interests or other things that are too large for the individual or the locality to deal with, then it's all just kind of spitting in the ocean because it won't have any real effect in in the world on on opinion if that makes sense. And I'm not sure I can go there.

I don't know if I can go there. When you stare into the abyss, it stares back into you here. I don't want it staring back at me. I want the abyss to go. I'm sorry. I wouldn't, I wouldn't frame it as quite as politically pessimistic or nihilistic as maybe that or I made it sound. I don't think here's one of the ones that breaks my heart a little bit. Joe Biden, my favorite, maybe not my favorite, but one of my three favorite policies as a person who would have a list of that in the Biden administration was the expanded child tax credit that they did in the American Rescue Plan following. Yes. There was a sort of post-pandemic bill. They did it for a year.

They should have done it for longer. But the reason they did for a year is that this was the clearest, the best tax policy. I mean, you just, if you got a kid, you get a check. Right. And if you turned on TikTok, a whole new genre arose of people doing dances set to music when they got their child tax credit. Like, people saw it. They felt it. And the theory was that this would be such a potent policy, right?

Republicans are always setting popular tax credits to expire. And Democrats always, like, blanch at the last minute and extend most of them. Right. That no way the Republicans would let anybody get rid of it the next year. They wouldn't want to lose the midterms. but they did get rid of it. And the fact that it had been there didn't seem to help Joe Biden. Well, I think the midterms, though, you could make the case that the midterms really turned out surprisingly well for the Democrats, given the conditions. Yeah, I mean, it just doesn't seem to have come from that because we sort of watched the polling. I talk to so many people who studied this policy.

Now, look, if you had done that policy for five years, it might have been different. And so this is where I'd be more optimistic. So we watched the vice presidential debate last night. I don't know when this will air, but in my timeline, it was last night. And watching J.D. Vance, like, lying, but lying in this particular way where he's like, Donald Trump heroically created bipartisan action to stabilize and improve the AC. He did, like, very few people on earth have written as many words about the affordable character as I have. Like, Donald Trump did everything humanly possible to destroy that bill. He supported bill after bill to appeal it. He signed on to a Supreme Court case to try to get it. named unconstitutional. He cut the money for the navigators to actually, like, go out and tell

people to sign up, everything he could have possibly done to destroy, weaken, a road, sabotage the affordable character he did. And now they're out there saying, you know, I should vote for Donald Trump? Because he worked so hard to make Obama care better. Right. So over time, this stuff works. Donald Trump is out there saying, don't touch Medicare and Social Security. So the policies themselves can become popular and they can become useful, but it's over long time periods. And of course, it's hard to get the past in the first place. Right. Because the ACA was a disaster for Democrats in 2010, right? It took time for it to become something that was

politically useful for them. Right. So it's not that policy doesn't matter. It just, it needs time. And elections don't, don't always align with that. You know, when we talk about policy, again, like, if you really break it down, the ACA is kind of a gift to insurance companies. And what it's done is it's kind of allowed millions more people access into kind of a broken system. And so kind of getting back to, and Tristan, maybe this is something, but this algorithm that creates these incentives, it's customized to your life. So if in your life, your real pressures are much more direct than that, like policy is kind of diffuse. So you might look at it like, my kids are going to college while my parents

are now elderly and I've been working and playing by the rules. And emotionally, my feed is just coming in with all the money that could be used by the government to help me directly is going to Haitian cat recipes so that they can eat more pets. And that's the thing that you're believing. isn't there some way that we can tickle that reptilian brain in reverse? And I don't think government does a great job of this, being responsive to that squeeze, elder care, child care, all that, you know, other than maybe the child tax credit and things like that. But is there a way to, I don't want to say reverse engineer because it all sounds so Machiavellian,

but how do you unravel that pathway? How do you rewire the vagus nerve to not feel this thing that's not happening? Yeah. Well, there's these we call perverse asymmetries. Once someone believes a conspiracy theory, it's the best predictor of whether someone believes in a conspiracy theory is whether they already believe in a conspiracy theory. Once you break shared reality, it's much harder to put it back together again.

And so we should be trying to protect the breaking of shared reality rather than the amount of effort and work and labor it would take to sew it back together again. I think there are, though, ways in our work we think about if you were to change the algorithm and the design, what would you be sorting for? And of course, you run to this free speech versus censorship issue. And by the way, the tech platforms in order to avoid regulation have tried to frame the issue with everything we're talking about as a free speech issue. When they do that, it means that you will never get regulated because that debate never convert. on what we should actually do. But it's not about free speech. It's about free reach and amplification. Amplification is not the thing that we're all entitled to. And free speech specifically in a gladiator stadium, which is the way that social media organizes our public debate environment,

that's not functional for democracy. Also, gladiator stadiums did not have blue checks and did not have algorithms where they'd be like, if you're interested in eviscerating slaves, you'll really like this. Okay, we'll be right back. We're back. That's the other thing, social media is not a town square. Incentivizing speech for outrage or conflict or hate or any of those things is the opposite of a town square, which is just a vessel. It's just a box. Yeah. Well, and the better, the current incentives in social media is the better you are at identifying a new cultural fault line and adding inflammation to that cultural fault line, you will be paid handsomely in likes, rewards, followers, retweets, and visibility. And you will be,

your content will be routed directly to the people who will be most angered or enraged or sort of, you know, activated by that. So it's a gladiator stadium. It's sort of a sci-fi AI gladiator stadium where when one guy sort of raises up his sword and slays someone, exactly who would be most activated by that gets to see. So he gets this sort of bespoke crowd that is most activated by that. So if you want to change this, obviously you have to fundamentally change the engagement-driven business model of social media. You can't simultaneously have a fiduciary duty to maximize shareholder value in the form of maximizing attention and also end up with a healthy democratic environment.

But one thing in our work that we found is, you know, for example, you could change the algorithm to sort for minimizing perception gaps. So for example, what are the sources of content that tend to, over time, help people accurately estimate what the other tribes believe? And you can measure that because what's hard, you can't measure what's true. That's very hard to do objectively. What you can do is say, what are the sources of content that help people better estimate how big a deal the Haitian immigrant dogs thing actually is?

And you can sort of say, what are the content sources that do that over time? And what you're doing is you're sorting for more shared common ground. That's what we really need is systems that are designed to find common ground. It's a great idea, Tristan, because I would say that free speech suffers also through the hoarding of outrage and anger. because if I'll go on my feed, it doesn't matter what I tweet. The third response is always, why did you change your name, Jew? Like, that to me is actually suppression.

Like, you don't want to engage when the toxicity is like that. Is there a model? And Ezra, I'll put this to you, that can be created to do what Tristan said as a public utility, a process by which you can kind of, it's a gauge, a check, a balance, a guardrail on the type of incentivized outrage and all that that we know is coming from companies that are for profit. In the same way that food companies exploit, they've got guys in lab codes trying to create chips that will sneak past the thing in your brain that says, you should stop eating, you're not hungry. Like, that's what I mean by government.

as a check and being responsive to the needs of the people. The way I've come to think about this, which is, I know that I like where it takes me, but I'll be honest about where I've ended up. I think a good capsule history of the media is that we used to sort, I mean, long ago, we sorted by literacy, but in most of American history, we sorted by geography, what media you had access to and the way it differed was primarily geographic. Newspapers were bound by space. You could subscribe to some magazines.

Eventually there's some radio, but at the beginning there's not that much radio. So everybody's listening to more or less the same things on the radio. You know, then there's, you know, three network television stations and they play, I Love Lucy, but there's also at 6 or 8 p.m. the news, right? We have a lot of evidence on this period. It creates much more consensus reality. Now, that reality might differ, right? you were reading a newspaper during the Civil Rights Movement in Alabama and you're reading one in New York,

you were getting very different visions of reality. But that reality was reshaped by geography and local community tendencies. What sort of changes cable news is like the first very big change, but the internet supercharges it, is we now have media and I guess reality sorted by interest. So one of the fascinating to me questions that gets asked in media studies is, is we used to think people couldn't know that much because the amount of information they had was bound, right? There just wasn't that much. When I was growing up,

the sum total of political opinion I had access to as a teenager in Southern California was the LA Times opinion page. It just was not that much political opinion. Right. I could, like, maybe there was some conservative talk radio, but I didn't listen to that. And all the information in the world

was just one encyclopedia Britannica of A to Z. Now you can have anything. But on average, people didn't get more informed. Why? Well, they did a bunch of studies, and basically what they found, and in pretty cool studies, sort of about when cable news rolled out to a place and then when the internet rolled out to a place, is after it rolls out, you actually do have a big change in who is informed. It's just not the average level. On average, it's not that people are more informed, but what happens is that everybody used to be kind of informed. And now you have the obsessives, right, who, you know, are listening to the John Stewart podcast or the Ezra Klein Show podcast or the Undivided Detention podcast, you know, or Pod Save America or, you know, Ben Shapiro, whatever it might be. And you have the people who just want no fucking part of this, right? They're, you know, they want sports, they want video games, they want cooking, they want anything. But listening to us talk about dynamics of politics and

media honest to God who is here at this point in this. And so the problem now with, like, I love all the public utility for communications ideas, but the problem is people have to choose to use it. And what gets people using a lot of these systems and why most people are on them is not that they're good information. If you want good information, you can get it. The New York Times is not perfect, but it's pretty good. The LA Times is good. We all sort of know where to go if we want pretty good information. But it's still analog.

But right. But the problem with trying to replace engagement-driven social media with a more virtuous form of social media is people are not there for virtue. They choose it because it is grappy, right? Instagram and TikTok and all these things. What makes them powerful is not the person seeking out good information. It's a person who actually doesn't want that much information at all, particularly not political information. They're there for other reasons. And they get some on the side or they start clicking on things about Gaza or the election or vaccinations or whatever.

And those are the people on whom elections turn. So what you need, if you're thinking about this public utility model in a world where people are choosing and they have these other options, right? If you're not going to shut everything else down, which we're not, is how do you make that public utility grabby such that the mass numbers of people are there who want to, you know, there's a reason YouTube headlines are all very loud. That's the point. And I don't know about grabbing, but I take your point. I think that's a really good point. And maybe Tristan, that's what I mean is I'm saying everyone's on heroin. Is there a good heroin?

What's our methadone? What's our public utility that can do that? Now, for me, I think it takes tenacity. As Maria Ressa says, a lie travels eight times faster than the truth. Well, then the truth has to work fucking ten times harder than a lie. And to do that, you need the resources. And I think that's where 24-hour media, I think, drops the ball completely, where they've still adopted the circadian rhythms of social media

and whatever else is coming without kind of battling it in that tenacious way. And they're pushed to do that because as social media creates a 24-7 news cycle, that's even more hyper-reel, hyper up to the instant than before, then cable media has to follow. I think that's one of the perverse effects of social media is actually pulled the incentives of all other forms of media. So everyone is dancing for the algorithm. How much of a cable news clip gets its visibility on news channels versus

later on the, when it trickles through social media and it gets a lot of visibility there. That's another right. But, you know, one of the things you're making me think of John is the work of digital minister Audrey Tang of Taiwan. And she uses a database called Kofax. So she basically says, fact checking is hard, to your point, there's many more things running through the system than we can possibly fact check. So they have a crowdsourced fact checking sort of system called Kofax.

And people contribute. And then what happens is to your point that Maria Ressie made, that a lie spreads, you know, eight times faster than the truth. She actually wrote an AI so that when there are trending topics about something that is actually already in the Kofax database, it adds this context in real time. One of the things we used to say in our work is if you can make it trend, you can make it true. It's the liar's dividend. If you just sort of make something more salient and visible, it lands in people's brain. It's there. they've heard about it before,

they don't remember where they heard it or whether it's true. It's just that it's visible. And if you want to catch it in real time, one of the ways we could be using AI is actually by adding more context and synthesis of multiple perspectives that are grounded in fact and adding that in real time

to the information that we're seeing. An adjunct, a kind of something that runs kind of hand in hand with your social media. So it's almost like a tool, an overlay, that you can, place on there. I think that's a brilliant idea. In some ways, it's what I think Elon, for whatever else he's doing, I think he was trying to get at with community notes. Now, those things can be

weaponized and manipulated in different ways. They can be. They're imperfect. But I think that's a really interesting idea. Now, Ezra, to your point, because the funny thing is, so it starts out as a conversation of how does government become more responsive to the needs of its people with all this money and all this communications and AI and all that. And what we end up with is AI, IAI social media, social media, social media. I don't think this will make government any more responsive. I have a different answer to that question if we want to do that, that I would love to give you my actual answer to. Bring it. Bring it.

You cut the process it makes government unresponsive. Here's how I think democracy should work. People vote for a candidate. The candidate who wins the most votes wins power. The candidate and coalition that wins power. I know this is getting weird already. They do some rough approximation of the thing they said they would do. And people decide if they liked that.

And then they either kick them out or vote them back in the next time. The way we actually do it. Go on. Is we vote for a candidate. The candidate that wins most votes may or may not win power. It's exciting. We see what happens at the electoral college, including this year. Then when they get in, because we have staggered elections and the Senate filibuster,

they can either do none of or 10 to 20 percent of what they promised. And then people pissed off that government isn't working that well and everybody's just fighting all the time. Disatisfaction builds up. and they usually get kicked out or people sort of turn off. So look, like my program for making government more responsive

is to accept the bad and the good in this because it can go the other way, as can making a social media utility that Donald Trump might one day run or J.D. Vance might one day run. But is get rid of the filibuster. In a bunch of different places, you need to actually empower the people

we elect to do things. Liberals have a really big problem with this. Liberal proceduralism. We've made the government incredibly easy to sue. We've created huge, bounce of process between getting anything done. The regulatory state is unbelievably complicated to go through. Yes. So people don't get the policies they voted for. And then we wonder why they're

pissed off. Here's where we go. Now we're going to tie it all together. Come on, Ezra. Here we go. I love it. Tristan, jump in with me. Here's how we tie it all together. An AI overlay that is also like a moonshot for bureaucratic excess that we create, rather than going to the moon, like fuck all these other planets that we're going to. We've got this one right here, and it has water and air. And we create a bureaucratic moonshot using these tools. See, right now, these tools are being absolutely dominated by profit, margin, and incentive. There has to be a more robust public utility usage for these kinds of things, creating the thing that Ezra is talking about. Government is wildly inefficient with that. Here's the other thing I would do to that, Ezra, and you tell me what you think.

This permanent campaign turns a mild disagreement into an argument, into a feud, into two sides that don't have anything to do with each other. We're never going to get the money out of it, but let's shorten the electoral system as well. This 100-day campaign we've been in, I think has been pretty good from that perspective. Fine, let's just leave it at that. The sort of thing where we did like a reset and then it was a new race and there's only 100 days in it. 100 days. We could have had some more debates in my view. I think this is actually, it's how other countries do it. It was healthier. Much healthier.

That plus Tristan, what do you think of those public utility usages? Maybe they don't replace the social media, but they are tools that we can use to bring forth some of the things. Even Ezra's talking about about better government regulation that is not so. You know, you can't have a homeless problem and then decide that and everything has to be net zero. Like you can't do everything. It was like sort of like in Shapiro and Pennsylvania. The highway goes bust and he's like, oh, that's an emergency.

We're going to throw out all the regulations and actually fix it in two weeks. Yeah. Well, I think the thing that you were pointing to here and throughout this conversation is the complexity gap that there's more and more issues to respond to than our bureaucratic institutions have an ability to respond to as climate change and climate events go up. That's going to keep going up. As sort of the outrage machine keeps spinning up more things to be upset about and for governments to respond to, that goes up.

and you're talking about that people lose faith in their institutions when they're not responding at the speed and clip that the issues and crises are hitting us. In our work, we often reference this just totally fundamental quote by E.O. Wilson, the father of sociobiology, who said the fundamental problem of humanity is we have paleolithic brains, medieval institutions, and godlike technology. And so that is the alignment problem that we have to align. And that's to sort of weave it all through. you're saying earlier, how do we weave our reptile brains, match how they work, how the dopamine cycles that Ezra was talking about, we need some kind of engaging media that is keeping our attention at some baseline, but then have it be aligned with not medieval institutions, but 21st century institutions that are maximally using technology, not in some kind of naive, techno-optimist technology

is going to solve our problems way, but we should be certainly instrumenting our institutions with technology. And then how do we then take the last part, which is currently we have God-like technology that is incentivized towards the worst goals. In social media, it's incentivized to the race to the bottom of the brainstem. And in AI, it's incentivized to the race to take shortcuts and drive recklessness, concentration of power, huge risks in society. Stock options. Stock options. So we have to realign that whole equation. And that's what we think of in our work as humane technology. Humane technology is realigning our paleolithic instincts and our reptile brains to the sort of enlightened versions of ourselves, instrumenting our institutions with this kind of how do we use AI to like...

To get them to be not medieval. Bring up to the at least 16th, 17th century. At least 17th century. Imagine you go back and you say... Age of Enlightenment. What if we had AIs that are actually looking at all of the laws that are no longer actually accurate to the present times? And we're actually accelerating the process of identifying the loopholes in previous laws,

the failure modes of previous laws, and actually updating them and having accelerating the legislative process by doing the process. by doing that and then actually getting more citizen input and doing that with AI. With efficacy ratings like you get in a restaurant. I'm not going along with any of this. Okay. Great. Ezra.

Ezra, no. Ezra. I'm bringing in some disagreement here. We're not all going to resign, grown in resigned acceptance of this. Ezra, this was beautiful. No, this is like Tristan is, of course, right about our old. Utopian, too utopian? No, this is not the problem in our policymaking that we have complexity.

The complexity gap is not a function of the world. It is a function of the systems we have built and layered and layered and layered and layered complexity. But the reason we cannot govern effectively, the reason that Gavin Newsom stands, the governor of California stands up and promises houses, but California does not build more houses. It's not that at some point in human history, houses became too complex to build. It's that when you try to build a house, particularly an affordable home that uses tax credits for affordability, we have made, and I've written a lot on this, it's partially what my next book is about. We have chosen to make this impossibly complicated. I think that's what Tristan is saying, though, to be fair, I think what he's saying is, yeah, we need something to streamline that process to show people that we've made it too complex.

Exactly. Yeah, but I'm just saying that like inside the system, like I think like liberals, and I'm not, I'm not pin this on Tristan. I just, liberals need to be in ripping parts of the system they built out. But that's what I'm saying. I'm agreeing with that. I'm saying AI could be used to help rip out the right parts of the system. Because so much of the system is broken is to target it. Yeah, exactly. But I think where I am disagreeing is that I don't want to frame this as complexity because these are all decisions, right?

Okay, Kamala Harris's housing plan, right? When you look at that, it is a series of tax credits. Each one of them micro-targeted. So you get a tax credit for building affordable homes for first-time home. Every time you layer one of these things on, you have made it more complex. Or the $25,000 home buyer, tax credit, right? Which is not for buying, but it's not for building, but for buying. That tax credit is actually for people who have never owned a home, have had two years of rental payments,

and whose parents have never owned a home, and you can kind of keep going down the line like this. It is not, we are making the complexity. And yeah, we might all be on the same boat, but we can't solve that with AI, particularly liberals, so then not only, but in this case, I think it's liberals who share these goals, have to decide to do government differently. and have to decide to have different values in government. And they don't want to. They keep writing things this way because they don't want to change it. I honestly think we're all in agreement.

Tristan's right about the world. But I'm just saying the bureaucracy. We've got like that we've created. We've created. We can't tell ourselves it too complicated. We made it complicated. Yes. I don't I don't disagree with that.

Tristan, I say, here's what I say. What a wonderful conversation. And at the end of it through all the, I don't agree with all. I think we've come to a. kind of a kumbaya. We might have found some common ground. And if we can train an AI on this conversation. See?

And set it loose. You see how it's so, it's really so easy. But it's fabulous, guys. I truly appreciate you enlightening me on all these different things and having the conversation. It was wonderful. Ezra Klein, columnist in New York Times, hosted the Ezra Klein Show, podcast, the author of Why We're Polarized.

Tristan Harris, co-founder, Center for Humane Technology, co-host of your undivided attention podcast. Thank you both so much for joining us. Thank you so much, John. Wow. Fabulous. Those guys were fabulous. Can I tell you something? Here's what impressed me about. We're here with our

Airswall producers, Brittany Mametovic, Lauren Walker. Hello. Their facility in minutia, in details, in names and the father of something biology and the quote, and it's all in their brains. There's no, I like to think in broad sketches.

There's very little that's filled in. Those guys are just Renaissance painters. I'm just doing the, you know, the herring sketches on a subway. But those guys are, I thought it was fabulous. I thought it was the most interesting disagreement agreement I've ever seen in my life. I thought we were agreeing. I think we did agree. I know that you think you did.

How are you got? What else? How are the viewers holding up? What do they got for us? Anything exciting? Yeah, we got some good questions this week. The first is... About the Mets?

Is it mostly about the Mets? Yeah. John, I'm going to ask you again. Are you hopeful at all? No. I'm going to answer that the same way every time you ask it. No, I am not. I like it.

Yeah. Okay. If Elon Musk came on the podcast, Okay. What would you ask him slash want to talk to him about? I would do that. I think that'd be fabulous. So that's an interesting question because I do know him a little bit.

I don't, we're not friends, certainly, but I know him a little bit. And we have had conversations. And I do, I know that this is a liberal heresy. You know, I don't have, I'm just like, Musk! Like I think, I do think at his heart, he is, he does some incredible things for the society. Starlink, and he's the, the neuralinks that can help somebody who is paralyzed. But I do, I would probably agree in the liberal orthodoxy that the kind of middle school edge lord shit posting has probably gone a bit far. And normally that would be kind of

inane, but not when you run the site and your edge lord shit posts end up at the top of my for you page for no apparent reason whatsoever. But I think, and this is a conversation that we, in fact, have had, a very small one. But I am, I, it's hard for me to understand how if you're, if what you hold above all else is free speech, it's hard for me to understand supporting Donald Trump, who's been pretty explicit about penalizing speech that he disagrees with, whether it's removing a license from ABC or whether it's even, you know, I'm going to jail Mark Zuckerberg and people could say, well, that's for election interference. But it's election interference as defined by Donald Trump. And we all know that he views everything through Trump colored glasses. So election interference

is interference that he thinks might work against him in an election. It's like, what's a fair election? It's an election I won. Oh, that's actually not the real definition of it, but fine. So that is, I'm always. struck by and it's sort of this broader point at the paradox of kind of Trump's support which is his constituency is boy they love him but they're very draped and kind of patriotic and Americana paraphernalia they're all you know don't tread on me we the people with the tri-corner head the whole thing but i think at his heart Donald Trump's instincts are not respectful of constitutional checks and balances, whether it be judicial, congressional, and executive.

Just by the very nature, I mean, if you ask the Supreme Court to give you complete and total immunity, well, you've just negated the revolutionary war. So the paradox at the heart of that movement is the thing I have the hardest time coming to grips with. I mean, for God's sakes, we love the Constitution. Donald Trump literally went in front of a bunch of police organizations and like, hey, man, I wish I could give you guys just an hour to beat the shit out. of everybody who shop lives and you're like, doesn't the Fourth Amendment? Like, how do you reconcile that with all the we the people shit?

That's where I would go. I thought at the very beginning of that run, you were about to admit that you had a cyber truck. Have those made it to your neighborhood yet? Yes. It was the first time I saw one, I thought somebody had put wheels on the refrigerator. Apparently raccoons are thinking that they're dumpsters in Dream Breakings. The thing that's driven me, as I go by, it's all smudges. Like, there's one in our neighborhood that keeps going by.

And the whole time, you're just like, did someone who went in the fridge to get jelly, then go outside and try and get into the car? And it's just, it's a series of like moving thumbprints. There's something fascinating. It's so silly looking. It's obnoxious. And dangerous. It just looks like, it looks like.

it looks like something that was rendered in Minecraft and it's not quite like a 13 year old boy built it in his mind oh great now he's not coming on Lauren sorry yeah things a lot you actually touched on something which one of our other listeners commented in on which is what do you think of everything happening in Georgia with voting laws and election security rules will you be talking about that on the podcast yes we we'll be talking about in the podcast. As a matter of fact, and Lauren, boy, was that, that was a beautiful prompt and a beautiful lead. But I believe we're talking about it next week with, I believe it's Stacey Abrams and Ben Ginsburg. Is that correct? Yes, correct. I'll tell you, one of my biggest

concerns, and I'll lay this out with them as well, is the security of the individuals who are gracious enough to volunteer for the administration of our elections, who are doing this literally out of a sense of whatever, civic pride, civic duty, they're taking time out of their days. And they are being, I mean, threatened and doxed. And it's awful. Truly one of the most awful things that I think is rotting away the foundation of what we're trying to do here democratically. Brutal. We'll definitely be touching on that next week. All right. Well, done, Lauren. Well, fabulous show today. Brittany, do you have the social media? Yeah. What are we supposed to do? You can follow us on Twitter at weekly show pod. Instagram, threads, and TikTok. We are

weekly show podcast. And please like and subscribe our YouTube channel, The Weekly Show with John Stewart. And also we are putting it on a satellite dish and broadcasting it like the setty out into the universe, hoping to hear back from ETs.

Thanks everybody so much. Another fabulous program. Lead producer Lauren Walker, producer Brittany Mehmetavik, video editor and engineer Rob Vatola, audio editor and engineer, Nicole Boyce, researcher and associate producer Jillian Spear and our executive producers,

Chris McShane, Katie Gray. Look forward to seeing everybody next week. Thanks, guys. The weekly show with John Stewart is a Comedy Central podcast. It's produced by Paramount Audio and Bus Boy Productions.