Taylor Lorenz’s Power User - Everything You Know About Algorithms Is A Lie: Section 230 Algorithm Problem Explained

Episode Date: March 9, 2026

Are social media algorithms actually destroying the internet? Support my independent journalism:🙏 Patreon: https://www.patreon.com/cw/taylorlorenz    🗞️ Buy a paid subscription to my Substack: https://www.usermag.co   In this episode of my Section 230 series, I sit down with Eric Goldman, Associate Dean for Research and Professor at Santa Clara Law, and one of the nation's foremost experts on internet law, to debunk the biggest myths surrounding Section 230 and algorithmic amplification.I hear all the time: platforms should lose their Section 230 protections because they use algorithms. But what actually is an algorithm? Most people don't even understand the basics. Eric and I discuss how even the beloved "reverse chronological order" feed is an algorithm, and why it actually rewards spammers, trolls, and bad actors.We dive deep into how politicians are using the "techlash" to gain power, why censorship has become a bipartisan value, and how companies like Meta actually want Section 230 gone so they can wipe out their smaller competitors. From the dangers of age verification laws to the truth about how algorithms protect us from navigating an internet that functions like a giant folder of Google Drive links, we discuss the real existential threats to the open web, and how arguments about "algorithms" miss the real powers at play. Topics Covered:Why reverse chronological feeds encourage bad contentHow algorithms function as basic editorial choicesWhy Meta/Facebook lobbies to remove Section 230 to build competitive moatsThe bipartisan push for internet censorship and government controlThe devastating impacts of age verification laws and losing youth online spaces

Censorship is a bipartisan value. It's one of the things that the parties actually agree upon. They're not arguing over the mechanics. They're only arguing over who has the power. Welcome back to my series on Section 230, the most important and most misunderstood law on the internet. Today, I want to talk to you guys about algorithms. I think we all feel the effect of algorithmic amplification on our social media experience. Over the past 10 years, social media has become increasingly dominated by recommendation algorithms, search algorithms,

Basically, all of these simple services that one day sort of just presented us content from our friends have completely revamped and redesigned themselves to promote a lot of algorithmic content. There's a lot of people online when I start to talk to them about Section 230 say, well, because these platforms, you know, use algorithms or algorithmic ranking, they should lose Section 230 protections. These people argue, well, aren't algorithms themselves editorial decisions? Doesn't that make them more like a newspaper? Platforms should be regulated like newspapers because of this. It's actually not that simple. And to sort of debunk this argument and help people get a better understanding of why this is not really an effective argument against Section 230 protections. I had to call up Eric Goldman.

Eric is an icon in the space. He's been fighting age, identity verifications, and surveillance laws online for 30 years. He's the Associate Dean for Research and a professor at Santa Clara Law. He's one of the nation's foremost experts on Section 230. And today he's joining me to break down the nuances of Section 230 as it relates to algorithms and how in, a lot of ways, the way the tech companies operate and sort content today is not actually that different from what they were doing even 30 years ago. Eric, welcome. Hi. So to start off, I think most people, if you've watched this far in my series, probably have a good understanding

of Section 230. But how do you describe Section 230 to other people, sort of in layman's terms, if it comes up in conversation? Section 230 says, to summarize, websites aren't liable for third-party content. It still leaves liability on the part of the person who's author and creating the content, but other people downstream from that person are not liable for it. I think that's a good, like, the shortest summary that you could give, probably. This description, which you just gave, which is also the description I basically give, makes a lot of people angry or uncomfortable. They're like, well, these platforms should be liable.

There's a lot of bad stuff on there. You know, why don't we want them to be liable? And I've gotten into some of that in my other videos, but today I want to talk about algorithms because I think when you start to ask people about, like, the harms of social media allegedly or, you know, what makes them upset. A lot of them talk more and more about algorithms. So what was the internet like when Section 230 passed? And was there any sort of editorial judgment

in terms of how they curated content within these portals? So we're going to take a stroll down memory lane for a moment. And let's start with the default way of presenting content circa mid-1990s. Much of that content was presented reverse chronological order. Much like many of our email accounts are presented today. But almost everything was like that. At the top was the most recent item.

And there were other efforts to change that, but those were at the margins. So, for example, there might have been the most popular content would be presented or the one that had generated the most comments or sometimes editorialists selected items would go up at the top. Some editor was deciding this conversation should be moved up front. But obviously, we're far away from thinking that reverse chronological order is the best approach. And indeed, the studies have suggested that reverse chronological order as a sorting mechanism, and rewards bad content. It actually encourages the production of bad content. I do want to point out that reverse chronological order is about the dumbest approach to

organizing content. It's as simple and thoughtless as perhaps any other approach. But it is still, and I'm going to describe it using a magic word, it's still an algorithm. It is still a formula for deciding how to rank order content. It might be an intelligent algorithm. It might be a simple algorithm that a three-year-old could come up with, but it's still an algorithm. That's why we're going to struggle discussing about Section 2A2 application algorithms, because no matter what, any form of content presentation and organization will be a form of an

algorithm. The idea is that if you get a benefit to being at the top of the content order, which is the general payoff is you get the most attention, it encourages the production of more content. People are all flooding the zone to try to occupy that top space. And without having any other filtration process, you're going to get the unfiltered slice of humanity, which means that over time, as people are all chasing the limited tension zone, the bad people are going to be competing as well. That includes the trolls, the spammers, the male factors. They're all going to be pushing themselves into that zone.

That's why we really don't want to rely upon something like reverse chronological order, because it doesn't do anything to try to curb the spammers, the trollers, the malfactors. It rewards them equally. Yeah, I think of this a lot where people will be like, I miss the early internet and everything should have a reverse Kron option. First of all, I think META actually does now on Instagram. You can go and choose that setting or you could for a while. Nobody really wanted it.

It's a pretty bad experience. And I think what a lot of people actually miss is just the internet before it was commercialized because, as you said, there is a lot of bad actors. There was a lot of spam back then. Like it is a better experience to see more engaging posts. I think of this even as a journalist. Like, I mean, now comment sections are becoming less popular. But one of my jobs back in the day when I worked at a media company was to troll through a lot of the comment sections and highlight popular, something that I'm sure is done by an algorithm now, you know, things that got enough like conversation going.

And I feel like that was always a value to see kind of like, oh, here's what's popular. Here's what's interesting. Here's what people are engaging with. But I guess this idea of like algorithms. really took hold. I mean, I guess it started in the public consciousness in terms of social media. I would say with Facebook introducing news feed in 2006, I can't remember if that was reverse cron or it sort of immediately defaulted to an algorithm. But if it didn't immediately defaulted, I think it very quickly became algorithmic because so many people were posting on Facebook.

I don't remember exactly the history of that. I do remember there was a time that the default was reverse chronological order. I don't remember when they switched over to some kind of algorithmic approach. And, you know, you. You can go and Facebook and do a reverse chronological order, but it's still curated. It's not all of the content that your friends have posted, reverse cron. It's still in the background there picking the top three or 500 items or whatever the right number is. And then they're only presenting that content reverse chronological order. So it's actually quite hard to get the true content of your friends on Facebook.

And I think most of us are okay with that at this point. We know that not everyone is equally situated in that environment. So something that you hear so constantly, or at least that I hear, when I bring up Section 230, is people saying, well, these social media companies should lose protection from Section 230 because they use algorithms. Algorithms are an editorial choice. Therefore, Facebook or Instagram is like in newspaper and should be regulated as such. I come back to the starting question, which is what is an algorithm? And I've taken the position already. that reverse chronological ordering is an algorithm, and it's automatically deployed. It's not something that is manually handled by humans.

And so if we're going to say automated algorithms are disqualifying Section 230, then Section 230 would have never existed. The whole point was that it covered even something as stupid as automated reverse chronological order. That was part of it. I think that there's a bit of technological exceptionalism in any discussion about ALA. that we see when the machines do it, it should be treated or thought of differently than when the humans do it.

I'm an old school guy. I used to subscribe to a print newspaper that would arrive at my door every day. And there were a series of algorithms that decided how to present and organize the material that is in that paper. There were standards about how many column inches a particular story might get. There were standards about what the proper placement of stories on a page would look like. There were standards about how big of a headline should be used. These are all editorial choices that were made by a human, but they followed a set of patterns and standards that look a lot like algorithms. And so when I talk about publishing content, I refer to three verbs as describing the component. the components of publishing content.

You gather the content, you organize the content, and then you disseminate the content. And the algorithm is a part of that organization. Once you have gathered all the content, how do you organize it? How do you put it all together in a way that's actually accessible and hopeful to the audience? And that approach of organizing is an integral part of the publishing function. And for that reason, to the extent that Section 230 says you're not liable for third-party content, it means you're not liable for how you gather, organize, and disseminate third-party content. In other words, embedded in the protection for publication decisions of third-party content

is the idea you're going to prioritize and sort the content accordingly for your audience. That, to me, is so inextricably bound into what Section 2 is trying to do that any effort to do that can only really work if you say, But when the machines do it, we think of it as something different. And it's when the machines do it and you think of something different, I reject that premise usually. Usually that's now just kind of techno exceptionalism that can't be supported. Well, then what would you say to people that are like, okay,

but isn't Section 230 inherently treating machines as something different? Like if you make that argument, then aren't these platforms like the newspaper that you just described? And shouldn't they be held responsible for the content within them, just as you describe these newspapers are? So now we're getting to like the core of Section 230. Section 230 is a technological exceptionalist statute. It does say that the medium matters.

And the way I give the example is you can imagine a submission from a third party, what we used to call maybe a letter to the editor, if your audience even knows that such things happen, that would be sent to a print publisher, a newspaper, or to an online publisher that is publishing the identical content. And Section 2D3 says, the print publisher, if they publish a letter to the editor, they're going to be liable for it as if it was their own content. Section 2 there says if it's third-party content, they're not liable for it. Even if it's the identical defendant, identical content, identical author of the content, and identical legal standards,

Section 20 says, because it's now codified in an electronic medium, we're going to establish different rules. So the burden is on the proponents of Section 2 to justify that exception. But there's no doubt it is an exceptional statute and is my view that we can justify that exceptional approach by looking at the things that newly exist online that have never existed in the offline world. I'm just going to give a couple quickly. Some of the things that we see online that we have no offline equivalent to are things like consumer reviews, things like how-to videos on YouTube, things like wikis, especially something like Wikipedia and things like social media where we're all in a conversation with each other that we

can all participate in. And so we could say we don't want those kinds of content, in which case we should not have exceptionalism. But what we've learned is by having exceptionalist rules, by treating the exact same pieces of content differently based on the medium in which they're published, we get a whole bunch of new content that never had an offline analogy. And to me, that's why the exceptionalist argument, Section 230 is exceptionalist. We shouldn't have that actually section two exceptionalists and there's good reasons why we have that. What about people who would say, okay, well, you're right. It is this exceptionalist argument, but we've also created this toxic ecosystem that maybe doesn't have an offline corollary.

You know, now we have all of this harm, quote unquote, happening at scale where people are saying mean, mean things on Facebook and that can reach a lot of people, et cetera, et cetera. And so, you know, we need to revoke Section 230 protections for user generated content, all Section 230, I guess, you know, to dismantle that. because we've created these harmful systems online. Yeah, I mean, there's no doubt that there's harm online. That we're just going to accept. What we don't know, and this is actually the focus of some of the next work that I'm doing,

I think in a prior conversation I mentioned my book, that's what I'm going to cover my book, is whether the harm that we're seeing online is new to the Internet, is a mirror of activity that takes place offline, we just are now seeing it better, or if actually maybe we're not seen as much online harm as we would in the offline world or in a hypothetical universe where the internet didn't exist. So the fact that there's harm online doesn't answer that question,

and that is really the existential question for the internet. Is the internet actually making us worse than we would compared to the way we engage in other facets of our lives? And I think maybe we ought to be talking about in what ways is it making us better? And in particular, what I find is if we focus solely on the heart, We don't do a proper cost-benefit accounting. There's so many communities that get enriched

by having conversations online. And then go back to things like the how-to videos. There's such a positive spillover effect, if I can use an economics term. There's positive externalities from how-to videos. And so if we are focused only on the harms are caused by social media and not accounting for those positive externalities, what we're really saying

is we're concerned about this community, but we don't care about this other community. that's getting benefits. And I think that's a conversation we need to have. We need to talk about the cost and the benefits in one uniform conversation. If we don't do that, I think we're missing the point. Yeah. And also the harm that we would be enacting if we were to repeal Section 230 and not allow user-generated content shut down. I mean, we're seeing this happen in the UK. I wrote a piece for The Guardian last year about the deep platforming that happened after the Online Safety Act where we saw addiction forums shut down, support for violence against women shut down.

forums where children could actually report and discuss, you know, instances of abuse that they were suffering. Disability communities, queer communities, like so many of those places, I think, if we were to revoke these productions, that would also cause direct harm. Like, I think it's not like this is a neutral act that is only going to eliminate some harm and not cause, you know, this other greater harm, I would argue. And I think that's really critical. And I love the way you're framing that, Taylor, because I think that what we're really seen is when we see discussions about, reform or repeal Section 230. What we're hearing is there are certain people who would be privileged by that change and we should give them the credit and not be concerned about all these other people who are currently betting from the system who are going to lose that benefit and potentially suffer life-changing consequences as a result. You know, as I've been thinking about this for 30 years,

I've been thinking, what's the better solution? I bet you have some questions about that a bit, but let me jump to a punchline, which is that there's no way to reconfigure Section 2.3 that solves some of the harms without causing other loss of benefits or other harms. All we can do here is a set of tradeoffs. And I think we really need to be candid about that. And if the politicians are saying this is just a net good, they're lying. Because of the fact that when we do tradeoffs, we ought to be hearing from the people who are going to be disadvantaged and making sure that we're going to be. we're saying to them, we hear you, and yet we're still going to screw you over. Yeah, which I, unfortunately, I think sometimes they don't like to hear from anyone, and I think they sometimes do say, okay, that's great. We're still going to, you know, pass this bad law.

I mean, and they should be held accountable for that, right? So the regulators are saying, you know what, I hear you and screw you, well, that's a choice. And when it's a majority abusing the minority, ideally we have constitutional protections that backfill that to prevent that kind of abuse of the minority. You would hope so. I don't know. I just did an episode on Sesta Fasta, and it seems like there's not, you know, there's so much harm sometimes from chipping away at Section 230 that has not been addressed. It's always heartbreaking to talk about Sesta Fasta.

You know, it's just a tragedy of what Congress did. And we have not held our politicians accountable for causing that tragedy. Yeah. I want to go back to this idea of algorithms also because I think a lot of people have maybe a very flat understanding of just how these platforms work. Because I think when you say algorithm, a lot of people just mean like ranked content. content algorithms, but there's search algorithms that I don't think we want to go away. There's algorithms that show you like trending content. There's, you know, all of this stuff in these platforms that is not just the chronological kind

of reverse feed format of social media that we're all familiar with. I guess I would frame it slightly differently, though I agree with everything you said. Sometimes you'll see in the computer science literature discussions about recommender systems or prioritization systems. And those are one form of an algorithm and one that gets a lot of attention. But algorithms are also used in things like anti-spam efforts, looking for content that should be excluded categorically, efforts to do other forms of content moderation

that could feed into the recommender algorithm, or could be screened out again entirely and taken out of the recommending system because it doesn't meet the editorial standards. We've also seen developments of things like what I call civility prompts. These are algorithms that are watching content and saying, I see that you're about to say something in civil, and do you really want to do that?

And so that's a form of an algorithm that isn't doing the recommendation, not screening out content. It's just there's like a guide on the side to help people think about what they're doing and to recognize that maybe they're transgressing a social norm. So algorithms show up all kinds of places that help shape and organize the content or the corpus of content that's in the publisher's database.

When I was researching for this episode, I kept coming across people that were basically talking about some legal people, some articles saying that basically we want to protect algorithmic ranking as well because stripping immunity from algorithms or trying to sort of separate algorithmic social media in some way would really collapse this distinction between hosting and creating speech because in a lot of ways like algorithms are speech themselves. Yeah, if the argument is that someone who's using an algorithm shouldn't be eligible for Section 230, the internet, could still exist, but it would only be a hosting internet, not a promotion-based internet. So let's think about what that looks like. Imagine that the internet was just a series of Google Drive links, or drop-bock links.

And so you get a link, and there's content that's being hosted there. And hosting is a form of publication. It's a decision, I'm going to provide greater visibility to this content than things I'm not hosting. But it's limited because it doesn't then evangelize. doesn't then evangelize that content, doesn't help it reach the right audience, it just makes it available. And I would not want to navigate the internet that was just a series of Google Drive or Dropbox links for many reasons, but I really do rely upon some curator in between me and the authors

that's helping me, A, find the things that are most relevant to me, and B, also maybe doing some safety checks. You know, if you just click on a Google Drive link or a Drop box lane, who knows what you're going to get at the end of that. And I don't want that. I want there to be some kind of previewing or some kind of judgment call that this is safe enough for me to explore and is relevant to my interest. And we just take that for granted that that promotion function is something that we get on the internet. And so anyone who's saying, let's get rid of Section 2, the Rodman, is basically saying, we don't value that function, that maybe there would still be some of that, but it would be so much more constrained because of the liability risk that people

People should just find their own way. And boy, that just sounds like a terrible internet to me. Oh, yeah. An internet of Google Drive sounds miserable, actually. I think another thing that I hear from people with this algorithm stuff is people will be like, I think they really hate like the Twitter algorithm or the Instagram algorithm or something. And so they say, you know, well, we should ban algorithms. We should remove Section 230 protections from any website that uses an algorithm.

As you mentioned, this is an entirely crazy premise to begin with because everything's an algorithm. But I'm wondering if you can talk about small sites because when I've spoken to some people that operate smaller websites or smaller platforms or like something like Blue Sky, which also uses algorithms. And these smaller sites rely on algorithms in many ways even more than some larger sites that have resources to sort of have human moderators double check things, et cetera. So I feel like this idea that like big tech is using algorithms and they're bad and we need to, you know, this would hurt big tech. Like I think it would actually hurt a lot more smaller sites than. you might think who don't have that backup human resource. Like they are relying on algorithms for a lot of things. I guess I reject the premise that we're going to want to have laws that distinguish

between bigger sites and smaller sites. And part of this is because there's not a really good methodology for determining who's big and who's small. We could use things like number of employees or we could use revenue or profits. We could use things like traffic measures. And all of these have pros and cons that really need to be discussed before we even decide that that's an appropriate way of doing it. And once you set up a system like that where big services are covered by a different

and maybe less viable rule than smaller services, then you have incentives for people to do combinations or to, you know, do game playing to try to evade the regulations. So from my perspective, I think we should have a single rule that applies across the biggest to the small list. And then to your other point, I would point out that really, I think the primary beneficiaries of Section 230, the people who benefit the most are the new entrants that are building an audience. Now, some of them may stay small the rest of their lives. Others are just on the way towards becoming a bigger platform. But if you don't have an audience, you're having to make all your investments speculatively to build the audience, build the infrastructure, get the content,

and if you don't have the kind of legal protection section that they provide, the investment capital just is at higher risk of being lost. And so if we want new entrants, Section 230 actually opens up the door for those new entrants and encourages them to come in and compete with the bigger players without having to build the industrial grade resources that the big players have. And so I see the Section 230 that's an interesting effect on competition policy that I think is underappreciated. And that's one of the reasons why big sites, especially, sorry to name names, but Meta,

has literally said, we don't really care that much about Section 230. What they're really saying is you get rid of Section 230, what you're doing is deepening my competitive mode. You're making it harder for anyone to get across that mode to come and compete with me. And META thinks that might be okay. Well, yeah, I think they would love to see their competitors wiped out. I mean, I think a lot of people forget that META at the time, Facebook, including Cheryl Sandberg, advocated for Sesta Vasta, actually. Yes, they did, and I still won't forgive them for that.

But they also have been running ads, millions of dollars of ads, especially in the DC area, saying, you know what? Section 230 is outdated. You should reform it. And to me, it's just shocking because we think that they rely heavily on Section 230. They invoke it in court all the time. And they were going to their regulators and saying, regulate us. Go ahead and take it away from us. So, you know, their attitude towards Section 230 looks very, very different than a smaller enterprise.

what knows that their existence is dependent on Section 230. Which is so crazy to me because when Joseph Gordon Levitt was recently on the hill in front of this big, you know, these senators, Lindsey Graham, Durbin, et cetera, who have this, they have this giant sign on this podium and it says something like protect kids from big tech or like end big tech. Some ridiculous, you know, saying that is not in line with the policies that they're proposing. They constantly bring up META and they bring up the harm, the very real harm that META has done. Meta's done a lot of, you know, bad things I would argue in the world. Some good things too, but, you know, me and Marr and stuff. But it is so conflated. Like, it is this idea that like the reason meta is able to do these quote unquote harmful

things is because section 230. And as you said, they've been at best ambivalent at worst, like actively lobbying against it because they don't want competitors. And you see this reaction as well with sites like blue sky, right? Where they're out here fighting these laws, these identity verification laws, et cetera. while Elon Musk and Twitter co-signed COSA. Or, you know, like, it seems like there's this idea that revoking Section 230 would cripple big tech. And that's not maybe how it would play out, as you mentioned.

A lot of Section 230 protections actually protect the small guys, the new players. There's a second order point here, I think, is worth making explicit, which is there's a presumption that Facebook and Google are big enough. They have enough money to lobby that they can paralyze. any initiative that might harm the industry, that they're going to fight the fight, they're going to carry the water for the industry, and then all the other players are going to get the benefit from that.

And what we've seen over and over again, including in Sesta Fasta, is that the industry can divide itself. And when it's divided, it becomes so easy for regulators to conquer it. And so without having a uniform standard where the industry is not fighting amongst itself, there is no one defending the industry. internet. It's not Google. It's not meta. They might be actually the source of the problem, but even if they aren't, they're just sitting on the sidelines or actively, you know,

quietly working in the backgrounds to try and advance their own interests. And nobody is actually speaking up for the internet in a way that carries the kind of weight that we think meta and Google could have. So the internet is in great jeopardy for that very reason because the industry itself isn't even coordinated. Yeah. I mean, I think it's so important, too, when you say industry. We're talking broadly about the internet. All of these web service, like not just big tech. Because I think a lot of people say, I'd love to see the tech industry crushed. And it's like, yeah, you would love to see Peter Thiel have taxes. Like, you would love to see a lot of like big players. But that is actually not the representative of the entire internet, you know, especially

reporting on the stuff that's happening in the UK. I'm just finding out like massive swaths of the internet that are being affected, these forums that you'd think in 2026 who's on forums. actually a lot of people like there's all these like niche websites and communities and actually social media networks just for people suffering with cancer like it's the internet is so much broader than people think and I think so much of our online experience is mediated by meta and Google we forget this and you see this also I know you've done so much work on identity verification you see this with meta you know funding groups like the digital childhood alliance or whatever that are actually you know pushing identity verification laws they just

want it to happen on a different level than that or you know Like, they do a lot of shady lobbying. Ultimately, their interest is in meta, not protecting civil liberties on the internet and not protecting the rights of their competitors. That means that we as internet users are just a pawn in somebody else's game. You know, it's like, you know, the metaphor about elephants and mice. You know, when the elephants are fighting, the mice get squashed. And that's exactly what we are in this circumstance. And it's heartbreaking because it didn't used to be that way.

It used to be that there was coordinated efforts. And many of your audience may not remember the battles over. SOPA, the Stop Online Piracy Act back in 2012, when the industry coordinated a way that actually changed policy outcomes, but only because there was coordination. And we've seen when we stand together, we can fight. When we don't, we get carved up. Yeah. If some of these laws and sort of chip aways at Section 230 go through, or if there is this like increased liability specifically placed on like recommendation algorithms, how could that ultimately backfire on people? Because I think, I'm sure a lot of people are still listening and saying, like, I don't care.

I want these algorithms regulated. What would you say? I think I would start with the idea of who should decide what's proper and fit for a publisher's audience. To me, the answer is obvious. The publisher should decide what's fit for their audience. Like, that's the value ad, is that they gather, organize, and present content that their audience values.

And whenever we see targeted, criticism. criticisms against algorithms, what I hear is we shouldn't let publishers decide how to cater to their audience. Somebody else should. And that somebody else is almost certainly a government regulator. And that government regulator is almost certainly at this point, a partisan culture warrior who is going to be taking advantage of their ability to dictate algorithms to make sure that their team wins and their opponents lose. And so it's really about, power? Who gets the power to decide what's for the audience? And anyone who wants to criticize publishers deciding recommendation algorithms for themselves is saying implicitly, I'm okay with handing that power to somebody else and maybe taking a little bit of a flyer on the idea that somebody else is actually going to advance their interests. Yeah, which unfortunately I think is a popular idea among a lot of people in Washington. I think they're just like opposed. I think a lot of conservatives are like, I just want the power and we'll censor the liberals. And then the liberals,

I mean, I was so horrified. I posted about it recently when I was at this content creator summit for Joe Biden, where they were just like, if we have the power, we will censor, you know, the speech that we don't like. And no one is advocating for free expression. And like, both sides end up censoring. I would think, like, actually very valuable journalistic content. Like, as a journalist, I've lost Instagram accounts under Biden and I've lost Instagram accounts under Trump. And I think it's bad no matter how it's happening.

And I think journalists and people that actually don't buy into the. those partisan binaries suffer a lot the most. I would just point out in my mind, censorship is a bipartisan value. It's one of the things that the parties actually agree upon. They want to censor content. They may not agree on what content should be censored, but they're not arguing over the mechanics. They're only arguing over who has the power. And I will note, again, that didn't used to be that way.

Most government officials end up bending towards censorship is just inherent in the job that they want to control what people are saying about them. But there was a time when both parties were trying to out jockey each other as the free speech party. And now any party that claims to be free speech is A, line. And B, they're doing that in a way that really is just part of the culture war. They're just like, you know, what we mean by free speech is we win as opposed to the kind of like First Amendment values that animated, you know, the American psychology during my childhood. What do you think changed and what led to that change? Like, I feel like there's been such a market shift in my own lifetime, but I've noticed it really accelerate since, like, the tech lash started, like Donald Trump's first presidency.

It seems like it's somewhat more of a recent thing that's happened in the past 10 years. Yeah, I think that's astute. I would agree with you about the timeline and maybe the cause. But I would say, you know, A, as I said, censorship is a bipartisan value nowadays. So, you know, it's not like they're jockeying with each other over that basic premise. And so it's not like either side sees a political win from adopting true free speech values. And so to me, the root causes that I think our population, you know, the American people have started veering more towards pro-censorship. And I think for the very reason you described that the tech lash has said, we should fear the publishers and the tech.

space such that we should hold them accountable. And in so doing, the politicians see that. And that's just like, that's just like, you know, something, they just want to pour gasoline on the fire because every moment we're fighting big tech means we're not fighting the politicians. So as long as they can just point the finger, it's like that old far side cartoon, if you remember that one, where the rifle scope is pointed at the bear and the bear is pointed at the other bear and saying, shoot that one. That's exactly what the politicians are doing.

and they're saying, shoot big tech, leave us out of it. And so it's all about allocations of power. The politicians are feasting on the American people by spinning us up to be angry at big tech so that we're not angry at the government. But we are angry at the government. Despite their efforts, you know, the government has done so many abuses against people

that that only cured you so far. I think of the reaction to the Discord news recently where Discord announced that they were doing age identity verification stuff, basically globally. And there was so much outrage and all these YouTubers were talking about it. And I thought, yes, they're finally going to talk about age verification. We're finally going to talk about these state laws and the Section 230 stuff.

And they really just focused on Discord. They couldn't understand that the government is the problem. Discord is pre-complying with a, like, frankly, global, but certainly national effort as well to enact these policies. And they're not the only tech company doing that. Google and Meda have also started to pre-compli in many states. So it's like, yeah, as you said, it's just like anything but hold the politicians accountable, unfortunately.

There's something else that you brought up to, which is I think like there is this sentiment against big tech. And I think a lot of people feel trapped by these platforms. I always ask people when they say like, well, why can't we, you know, regulate big tech out of existence? I say, well, what other platforms are you using? Like I'm on mastodon every day. Like I'm on the Fediverse or whatever. But there are alternatives. And yet people don't use them.

Even the people that are the biggest credits, they're like, I'm trapped here. And so you have to regulate this. Like, what do you think is a recipe for a better internet? Because I think so many people, they do spend all. day on these platforms and they hate the way they operate and they do want to see more government control. And I guess there are other alternatives for them that exist out there that do not operate like that, but they don't want to use those for some reason. I mean, to be fair, Taylor, you yourself mentioned that there are all these niche communities that are taking place outside of

the, you know, the big names that add extraordinary value to those participants. And so I think if people were to reflect upon their internet usage, they would start to see that there are times that they are spinning out to these other non-gions. And and maybe not for everything, but for little pieces of their lives, and recognize that that's really important, and they should recognize that, fight for that, realize that that's being put at risk. But I'm going to offer up a hypothesis,

and I haven't done the academic research to back this up, but let me work with the hypothesis, and we'll see where that takes us. It's my hypothesis that every media ecosystem devolves into an oligopoly. In other words, every media system is going to be consolidated in a small number of major players with a long tail of smaller players.

Fixing, whether it's meta or Google, doesn't fix the underlying economic dynamics. You take them out of the equation, there will be a new oligopoly form just without whoever you knocked out in the first round. And if that's the case, we shouldn't be fighting oligoplies, but we should be thinking about what do we need to do to create an ecosystem

that keeps that market as competitive possible. And I mentioned it's actually to answer to their mind as part of the solution to that. It actually keeps the door opens for the competitors to the oligopolis. And also, where we need to intervene, let's talk about the interventions that will restrict the abuses

of the oligopolistic power. And so, for example, there are some culture war battles being fought with antitrust, but antitrust plays a role in regularly not oligopolis. We should be looking for places that they are taking advantage

to their marketplace physician and abusing it and correcting that. So in the end, competition is I think the answer with the understanding that competition might always drive us towards an oligopoly. And so if we accept that, the fact that there's a couple of big giants is not a surprise. It's actually we should expect that to be the state of nature. Yeah, I think like there's always also room for these big, broad social networks and they might shift.

I mean, I think it's interesting too that like TikTok came along. The reason that TikTok had to sell to bite dance to begin with, and they could have sold to Viacom, by the way, but they sold to ByteDance because they didn't have the resources. to compete with meta because meta was engaged in so much anti-competitive behavior bite dance poured resources in they competed with you know with meta meta tried to get them banned but now they're a major social platform so i totally agree with you i think competition is key and especially competition from smaller platforms as you mentioned that like are offering different experiences for people online i do this is maybe just my pet peeve because i do think the biggest critics are the heaviest users of these big social platforms because I'm like you spend all day on Twitter like talking about this but where are you showing up

in these other spaces and I think for people that do feel very angry with big tech or very sick of big tech like there's something about these platforms that are keeping you and maybe you're resentful of that but like recognize that within yourself and try to fix that behavior and try to spend more time in other places you might enjoy those other spaces or at least just like watch Instagram reels and don't feel so like resentful about it just be like you know what Instagram's really good at giving highly interesting content and I'm going to dive in for an hour and I'm not going to like, you know, get mad at Instagram for that, just the way that I don't get mad at the television for making a really good show on Netflix.

Yeah, I mean, I want to pick up on that last example because, you know, like I generally like the movies that are the big blockbusters. They're the mass market designed for the popular audience. That's great. But those are not my only diet of movies. I watch a whole bunch of long tail content as well. And so I don't feel bad about liking the blockbusters, but I also recognize that if all I'm doing is watching the blockbusters, I'm probably missing out on something. I think the issue is people feel like these tech platforms are like really destroying society.

And there's almost nothing you can do to sort of disabuse people of that notion. But I think what you said is so important that it's like looking at a lot of these underlying economic factors that's driving the success of these companies. And also just looking at like the problems that you mentioned just in when we first started talking like, Are these problems that are internet problems that the internet has created? Or are these societal wide problems that are being exposed by the internet and being revealed by the internet? And I think so often it's the latter. Like a lot of this stuff existed anyway. Atrocities were happening.

Attrocities are happening. And the internet is the way that we're even seeing them. But deleting the internet is not going to make less atrocities happen. I mean, think about something like the Australia's social media ban for under 16s. What it basically is telling that community is, A, we don't trust you. B, we don't care about you building your skills to be prepared for your future. And C, we don't care about your voice. We're going to take away your ability to speak up and restrict your ability to express yourself.

Like, those are terrible messages to send to children. Now, do they need training about how to use Internet? Do they need potentially some guardrails about how you said? Do they need supervision? Yes, to all of that. But the message that the government is sending is so anti-child. And the children won't get that message. They're smart enough to realize what the government is saying about them.

Yeah, I mean, I see a lot of these movements here in the U.S., unfortunately, that have gotten these kids on board where they're hauling some 13-year-old or 15-year-old up in front of lawmakers saying that, and this child is saying, yes, all my problems are because of the Internet. And I hate myself now because of the Internet. And that makes me sad because I know that removing the internet is not going to fix that child's problems. Like often they have the problems and they're using the internet in an unhealthy way because they have the problems. The problems are not necessarily caused by the internet. And I think that that's what's so sad when these kids get exploited for these really reactionary campaigns that ultimately harm the vast majority of kids.

It's like these are actually hurt kids that need help and not mass censorship and surveillance for everyone. Yes, although, again, every story is different. So I get a little nervous about generalizing that statement, but... Not generalizing. There are certain stories that I just, I feel bad for the kids. I mean, I just, I think of like a story I was doing at the Atlantic years ago where I was looking into like some of the sides and stuff for social media. And like, things are very complicated.

And often the story presented by the media is very flat and it can just be like social media bad. And there's a myriad of factors that go into some of these very tragic cases. And the plaintiff's lawyers. if they're suing won't tell that full story either. They'll expect the other side to figure it out or they'll never get emerged. But I want to come back, and I can't take credit for this. I believe it's typically attributed to Dana Boyd, who is a sociologist, who's done some really excellent work.

And she says, we don't let children cross the street without some training and supervision. And I feel like there's so much of that is really people are angsting about how do we let children cross the digital street. Well, we don't just let them do it, but they're going to have. have to do it eventually in their lives. So we're going to have to teach them and guide them. And so I think a lot of the tragedies that we're hearing about with children could be a manifestation of maybe we haven't done a good job about teaching people across the digital

streets. That's on us. We can fix that though. And what would you say to people that are like, oh, well, so isn't that an argument for age verification and stratifying the internet by, you know, people that are under 18 and over 18 so they can have those training wheels, etc.? Well, okay. Where those training wills are imposed might not be at the server level.

There might be other places where it's better to impose them. This is the longstanding argument that maybe we need more client-side filtering than we need server-side corrections. But also, I'm thinking about it just as like an education process. We need the schools to teach this. We need the parents to teach this. We need the parents to role-model it. We need the parents to be engaged in their kids' lives.

We need support structures around them, whether it's other family members, whether it's other institutions like extracurricular activities, they all need to be engaged with how do we prepare children for their digital future? That is the future. There is no question about that unless we burn down the entire internet, and there are people who would prefer that. If we don't do that, we need to teach them that, and that's a whole-of-society approach.

It's not something where we can expect the server-level provider, like social media service, to fix all the problems is that hasn't been addressed by other aspects of society. Especially when, if you get meta to just censor everything and scan everyone's biometric data to find out who's a child, like there's actually enormous harm that comes with that system, whereas there's much more benefit and much, almost very little harm from just educating people.

And as you said, like having this more collaborative process between parents, teachers, et cetera, that help prepare kids for the digital world. I mean, shouldn't we put the footnote here? See our prior conversation where we went through all this. I thought in a really brilliant fashion. Yes, I know. I will be linking my age verification video below.

Eric, thank you so much for joining me today. It's always a pleasure to chat with you. Yeah, my pleasure. Thanks for having me. All right. That's it for this episode of my Section 230 mini series. Thank you so much for tuning in. If you like this series, please, please support me on Patreon or Substack via the link below.

On Substack, I send a biweekly newsletter of everything that I'm reading and following online. You can also get that newsletter on Patreon where I do bonus content. You can get ad-free episodes and more. I also do a monthly Q&A for my Patreon subscribers. And I do a lot of stuff on SubSac too. So whichever one, you basically get kind of the same content on both. But either way, your support matters so much. This series is entirely self-funded.

I couldn't even find a single advertiser to support me. And I think that should tell you a lot about the stuff that I'm covering and the message that I'm giving. So please, please support me. Thank you so much. And I'll be back soon with another episode of my Section 230 mini series.