The Interview: How Wikipedia Is Responding to the Culture Wars

0:28Well, Casey, it's Thanksgiving week. And they had a really interesting conversation about all of the culture wars surrounding Wikipedia right now, how it operates, the political bias accusations that it has faced, the challenges they're now facing from AI and from Grokipedia, Elon Musk's version of Wikipedia. And we should note that this interview happened just before Grokipedia was released to the public. But we thought this was a conversation well worth hearing, and we wanted to bring it to you on this Thanksgiving week. Absolutely. And the second thing we wanted to tell you is that, believe it or not, Kevin, it has been an entire year since we released our episode on the 100 most iconic technologies of all time.

1:08And in the wake of that episode, listeners said, we would like to hear about more technological icons. And so we are proud to announce today that this Friday, in this very feed, we will be releasing the long-awaited follow-up to that list, the 50 most iconic technologies of 2025. Yes. and if you're wondering why only 50 this year, Casey, why are you only 50 this year? It's because there were not that many iconic technologies this year and we have a very high bar for what counts as an icon. And you know, Kevin, just to give people a little teaser of what they can expect, we have a clip from this episode that's coming up on Friday.

1:44Rare Earth Metals! So that is coming up on Friday. If you have a long drive somewhere to see your family on Black Friday because you're going to the mall. If you're camped out in line outside of a Walmart waiting for the LaBooBoos to open up. This is a great thing you can listen to. And by the way, did LaBooBoos make the list? There's only one way to find out Hard Fork this Friday. And we hope you have a great holiday. We're thankful for you. Very thankful.

2:24Thank you.

2:54Wikipedia helps define our common understanding of just about everything. But recently, the site has gone from public utility to a favorite target of Elon Musk, congressional Republicans, and MAGA influencers, who all claim that Wikipedia is biased. In many ways, those debates over Wikipedia are a microcosm of bigger discussions we're having right now about consensus, civil disagreement, shared reality, truth, facts, all those little easy topics. A bit of history. Wikipedia was founded back in the Paleolithic era of the internet in 2001 by Larry Sanger and Jimmy Wales. It was always operated as a non-profit and it employs a decentralized system of editing by volunteers, most of whom do so anonymously.

3:43There are rules over how people should engage on the site, cordially, and how changes are made transparently. And it's led to a culture of civil disagreement that has made Wikipedia what some have called the last best place on the internet. Now with that culture under threat, Jimmy Wales has written a book called The Seven Rules of Trust, trying to take the lessons of Wikipedia's success and apply them to our increasingly partisan trust depleted world. And I have to say, I did come in skeptical of his prescriptions, but I left hoping he's right. Here's my conversation with Wikipedia co-founder Jimmy Wales.

4:28I wanted to talk to you because I think this is a very tenuous moment for trust. And your new book is all about that. In it, you sort of lay out what you call the seven rules of trust based on your work at Wikipedia. And we'll talk about all those as well as some of the threats and challenges to Wikipedia. But big picture, how would you describe our current trust deficit? I draw a distinction between what's going on maybe with politics and journalism, the culture wars and all of that, and day-to-day life. Because I think in day-to-day life, people still do trust each other. People generally think most people are basically nice.

5:12and we're all human beings bumping along on the planet trying to do our best. And obviously there are definitely people who aren't trustworthy. But the crisis we see in politics, trust in politicians, trust in journalism, trust in business, that is coming from other places and is something that we can fix. One of the reasons why you can be an authority on this is because you created something that scores very high on trust. You have built something that people sort of want to engage with. Yeah, I mean, I do think Wikipedia isn't as good as I want it to be. And so I think that's part of why people do have a certain amount of trust for us, because we try to be really transparent.

6:02You know, you see the notice at the top of the page sometimes that says, the neutrality of this page has been disputed, or the following section doesn't cite any sources. People like that. Not many places these days will tell you, hey, we're not so sure here. And it shows that the public does have a real desire for unbiased, neutral information. They want to trust. They want the sources. They want you to prove what you're saying and so forth. How does Wikipedia define a fact? Basically, we're very old-fashioned about this sort of thing. What we look for is good quality sources. So we like peer-reviewed scientific research, for example, as opposed to populist tabloid reports.

6:52You know, we look for quality magazines, newspapers, etc. So we don't typically treat a random tweet as a fact. And so we're pretty boring on that regard. Yeah, it's sort of like the publication that you cite gets cited by other reputable sources, that it issues corrections when it gets things wrong. It's all the old-fashioned sort of good stuff. And I think it's important to say, when we look at different sources, they will often come to things from a different perspective or a different political point of view. That doesn't diminish the quality of the source. So, for example, I live here in London in the UK.

7:33We have The Telegraph, which is a generally right-leaning but quality newspaper. We have The Guardian, generally left-leaning but quality newspaper. Hopefully, the facts, as you read the articles and you glean through it, the facts should be reliable and solid. But you have to be very careful as an editor to tease out, okay, what are the facts that are agreed upon here? And what are the things that are opinions on those facts? And that's an editorial job. It's never perfect and it's never easy. And Wikipedia also is famously open source. It's decentralized and essentially it's run by thousands of volunteer editors.

8:10You don't run Wikipedia, we should say. It runs me. How do those editors fix disputes when they don't agree on what facts to be included or on how something is written? How do you negotiate those differences? Well, in the best cases, what happens and what should happen always is take a controversial issue like abortion. Obviously, if you think about a kind and thoughtful Catholic priest and a kind and thoughtful Planned Parenthood activist, they're never going to agree about abortion. But probably they can come together, because I said they're kind and thoughtful, and say, okay, but we can report on the dispute.

8:51So rather than trying to say abortion is a sin or abortion is a human right, you could say Catholic Church position is this and the critics have responded thusly. You'll start to hear a little of the Wikipedia style because I believe that that's what a reader really wants. They don't want to come and get one side of the story. They want to come and say, OK, wait, hold on. I actually want to understand what people are arguing about. I want to understand both sides. What are the best arguments here? Yeah, and basically every page has what's called a talk tab where you can see the history of the discussions and the disputes, which relates to another principle of the site, which is transparency.

9:29You can look at everything and see who did what and what their reasoning was. Yeah, exactly. So, you know, oftentimes if you see something repeating, you think, huh, okay, well, why does it say that? Often you'll be able to go on the talk page and read sort of what the debate was and how it was, and you can weigh in there. And you can join and say, oh, actually, I still think you've got it wrong. Here's some more sources. Here's some more information. Maybe propose a compromise, that sort of thing. And in my experience, it turns out that a lot of pretty ideological people on either side are actually more comfortable doing that because they feel confident in their beliefs.

10:08I think it's the people who, and you'll find lots of them on Twitter, for example, they're not that confident in their own values and their own belief system. And they feel fear or panic or anger if someone's disagreeing with them rather than saying, huh, OK, look, that's different from what I think. Let me explain my position, which is where your more intellectually grounded person will come from. What you're saying is supported actually by a study about Wikipedia that came out in the science journal Nature in 2019. It's called The Wisdom of Polarized Crowds. Perhaps counterintuitively, it says that politically contentious Wikipedia pages end up being of higher quality, meaning that they're more evidence based, they have more consensus around them.

10:54But I do want to ask about the times when consensus building isn't necessarily easy as it relates to specific topics on Wikipedia. Some pages, they have actually restricted editing privileges. So the Arab-Israeli conflict, climate change, abortion, unsurprising topics there. Why are those restricted and why doesn't the wisdom of polarized crowds work for those subjects? Well, typically the subjects that are restricted are, we try to keep that as short as we can. The most common type of case is if something's really big in the news, or if some big online influencer says, ah, Wikipedia is wrong, go and do something about it.

11:36And, you know, we get a rush of people who don't understand our culture, don't understand the rules, and they're just vandalizing, or they're just sort of being rude, and so on and so forth. And we just, as a calming down, just like, okay, hold on, just slow down. We're going to protect the page. And then there are pages where, you know, the most common type of protection we call semi-protection, which just means you have to have had an account for, I forget the exact numbers, it's something like four days and you have made 10 edits without getting banned. Now, typically, and this is what's surprising to a lot of people about Wikipedia, like 99 % of the pages, maybe more than 99%, you can edit without even logging in.

12:14And it goes live instantly. That's like mind boggling. But it kind of points to the fact that most people are basically nice. Most people are trustworthy. People don't just come by and vandalize Wikipedia. And often if they do, it's because they're just experimenting or they didn't believe that would like, they're like, oh my God, it actually went live. And I didn't know I was going to do that. It's like, Yeah, please don't do that again. This brings me to some of the challenges. Wikipedia, while it has created this very trustworthy system, it is under attack from a lot of different places. And one of Wikipedia's sort of superpowers can also be seen as a vulnerability, right?

12:55The fact that it is created by human editors. And human editors can be threatened, even though they're supposed to be anonymous. You've had editors doxxed, pressured by governments to doctor information. Some have had to flee their home countries. I'm thinking of what's happened in Russia and India, where those governments have really taken aim at Wikipedia. Would you say this is an expanding problem? Yeah, I would. I think that we are seeing all around the world a rise of authoritarian impulses towards censorship, towards controlling information. And very often these come, you know, as a wolf in sheep's clothing because it's all about protecting the children or whatever it might be that, you know, you move forward in these kind of control ways.

13:45But at the same time, you know, the Wikipedians are very resilient and they're very brave. And one of the things that we believe is that in many, many cases, what's happened is a real lack of understanding by politicians and leaders of how Wikipedia works. A lot of people really have a very odd assumption that it's somehow controlled by the Wikimedia Foundation, which is the charity that I set up that owns and operates the website. Therefore, they think it's possible to pressure us in ways that it's actually not possible to pressure us. The community has real intellectual independence. But yeah, I do worry about it.

14:27I mean, it's always something that weighs very heavily on us is volunteers who are in dangerous circumstances and how do they remain safe is like critically important. I want to bring up something that just happened here in the U.S. In August, James Comer and Nancy Mace, two Republican representatives from the House Oversight Committee, wrote a letter to Wikimedia requesting records, communication, analysis on specific editors and also any reviews on bias regarding the state of Israel in particular. Their reason, and I'm going to quote here, is because they are investigating the efforts of foreign operations and individuals at academic institutions subsidized by U.S.

15:11taxpayer dollars to influence U.S. public opinion. So can you tell me your reaction to that query? Yeah. I mean, you know, we've given a response to the parts of that that were reasonable. I mean, what we feel like is there's a deep misunderstanding or lack of understanding about how Wikipedia works. You know, ultimately, the idea that something being biased is a proper and fit subject for a congressional investigation is frankly absurd. And so, you know, in terms of asking questions about cloak and dagger, whatever, we're not going to have anything useful to tell them. I'm like, I know the Wikipedians, they're like a bunch of nice geeks.

15:54Yeah, I mean, the Heritage Foundation here in the United States, which was the architect of Project 2025, have said that they want to dox your editors. I mean, how do you protect people from that? I mean, it's embarrassing for the Heritage Foundation. I remember when they were intellectually respectable. And that's a shame that if that's what they think is the right way forward, they're just badly mistaken. But it does seem that there is this movement on the right to target Wikipedia over these types of concerns. And I'm wondering why you think that's happening. I mean, it's hard to say. There's a lot of different motivations, a lot of different people.

16:38Some of it would be, you know, genuine concern if they see that maybe Wikipedia is biased or, you know, I have seen, for example, Elon Musk has said Wikipedia is biased because they have this really strong rules about only citing mainstream media and the mainstream media is biased. Okay. I mean, that's an interesting question, interesting criticism. Certainly, I think worthy of some reflection by everyone, the media and so on and so forth. But it's hardly news to anybody and not actually that interesting. Then other people in various places around the world, not speaking just of the US, but facts are threatening.

17:23And if you and your policies are at odds with the facts, then you may find it very uncomfortable for people to simply explain the facts. And I don't know, that's always going to be a difficulty. But we're not about to say, gee, you know, maybe science isn't valid after all. Maybe the COVID vaccine killed half the population. No, it didn't. Like, that's crazy. And we're not going to print that. And so they're going to have to get over it. I want to talk about a recent example of a controversy surrounding Wikipedia, and that's the assassination of Charlie Kirk. You know, Senator Mike Lee called Wikipedia wicked because of the way it had described Kirk on its page as a far-right conspiracy theorist, among other complaints that they had about the page.

18:09And I went to look at the time that we're speaking, and that description is now gone from Wikipedia. Those on the left would say that that description was accurate. Those on the right would say that that description was biased all along. How do you see that, that tension? Well, I mean, I think the correct answer is you have to address all of that. You have to say, look, this was a person. I think the least controversial thing you could say about Charlie Kirk is that he was controversial. I don't think anybody would dispute that. and to say like, okay, this was a figure who was a great hero to many people and treated as a demon by others.

18:50He had these views, many of which are out of step with say mainstream scientific thinking, many of which are very much in step with religious thinking and so on and so forth. And those are the kinds of things that if we do our job well, which I think we have in this case, we're going to describe all of that. Like maybe you don't know anything about Charlie Kirk. You just heard, Oh, my God, this man was assassinated. Who was he? What's this all about? Well, you should come and learn all that. You should learn like who his supporters were and why they supported him and what are the arguments he put forward and what are the things he said that upset people?

19:22That's just part of learning what the world is about. So those words that were there, far right and conspiracy theorists, those were, in your view, the wrong words and that the critics of Wikipedia had a point. It depends on the specific criticism. So, if the criticism is this word appeared on this page for 17 minutes, I'm like, you know what, you've got to understand how Wikipedia works. It's a process, it's a discourse, it's a dialogue. But to the extent that he was called a conspiracy theorist by prominent people, that's part of his history. That's part of what's there. And Wikipedia shouldn't necessarily call him that.

20:06But we should definitely document all of that. You mentioned Elon Musk, who's come after Wikipedia. He calls it Wokipedia. He's now trying to start his own version of Wikipedia called Grokipedia. And he says it's going to strip out ideological bias. I wonder what you think attacks like his do for people's trust in your platform writ large. because as we've seen in the journalism space, if enough outside actors are telling people not to trust something, they won't. Well, you know, it's very hard to say. I mean, I think for many people, their level of trust in Elon Musk is extremely low because he says wild things all the time.

20:52So to that extent, you know, when he attacks us, people donate more money. So, you know, that's not my favorite way of raising money, but the truth is a lot of people are responding very negatively to that behavior. One of the things I do say in the book, and I've said to Elon Musk, is that type of attack is counterproductive, even if you agree with Elon Musk. Because to the extent that he has convinced people falsely that Wikipedia has been taken over by woke activists, then two things happen. Your kind and thoughtful conservatives who we very much welcome and we want more people who are thoughtful and intellectual and maybe disagree about various aspects of the spirit of our times, come and join us and let's make Wikipedia better.

21:42But if those people think, oh, no, it's just going to be a bunch of crazy woke activists, they're going to go away. And then on the other side, the crazy woke activists are going to be like, great, I found my home. I don't have to worry about whatever. I can come and write rants against the things I hate in the world. We don't really want them either. You said you talked to Elon Musk about this. When did you talk to him about it and what was that conversation like? I mean, we've had various conversations over the years. You know, he texts me sometimes. I text him sometimes. He's much more respectful and quiet in private.

22:21But that you would expect. He's got a big public persona. When was the last time you had that exchange? That's a good question. I don't know. I think the morning after the last election, he texted me that morning. I congratulated him. Obviously, the debate that happened more recently was because of the hand gesture that he made that was interpreted in different ways. And he was upset in the way that it had been characterized on Wikipedia. Yeah, I heard from him after that. I mean, in that case, I pushed back because I went to check like, oh, what does Wikipedia say? And it was very matter of fact.

23:02It said he made this gesture. It got a lot of news coverage, many interpreted it as this, and he denied that it was a Nazi salute. That's the whole story. It's part of history. I don't see how you could be upset about it being presented in that way. If Wikipedia said, you know, Elon Musk is a Nazi, that would be really, really wrong. But to say, look, he did this gesture and it created a lot of attention and some people said it looked like a Nazi. Yeah, that's great. That's what Wikipedia is. That's what it should do. Do you think Elon Musk is acting in good faith? You're saying that in private he's nice and cordial, but his public persona is very different.

23:42You know, I think it's a fool's errand to try and figure out what's going on in Elon Musk's mind. So I'm not going to try. I don't mean to press you on this. I'm just trying to refer to something that you said, which is people human to human are nice, right? That we're good. We should assume good faith. And so you're saying that Elon one-on-one is lovely. but he is attacking your institution and potentially draining support for Wikipedia. Well, I mean, I don't think he has the power he thinks he has or that a lot of people think he has to damage Wikipedia. I mean, we'll be here in 100 years and he won't.

24:29So I think as long as we stay Wikipedia, people will still love us. They will say, you know what, Wikipedia is great and And all the noise in the world and all these people ranting, that's not really the real thing. The real thing is genuine human knowledge, genuine discourse, genuinely grappling with the difficult issues of our day. That's actually super valuable. So there's a lot of noise in the world. I hope Elon will take another look and change his mind. That would be great, but I would say that of anybody. And, you know, in the meantime, I don't think we need to obsess over it or worry that much about it.

25:07You know, we don't depend on him for funding. And yeah, there we are. I hear you saying that part of your strategy here is just to stay the course, do what Wikipedia does. Are there changes that you do think Wikipedia needs to make to stay accurate and relevant? Well, I think what we have to focus on when we think about the long run, we also have to keep up with technology. You know, we've got this rise of the large language models, which are an amazing but deeply flawed technology. And so the way I think about that is to say, okay, look, I know for a fact, like, no AI today is competent to write a Wikipedia entry.

25:53It can do a passable job on a very big famous topic, but anything slightly obscure and the hallucination problem is disastrous. But at the same time, I'm very interested in how can we use this technology to support our community. One idea is, you know, take a short entry and feed in the sources. Maybe it's only got five sources and it's short and just ask the AI, is there anything in the entry that's not supported by the sources? Or is there anything in the sources that could be in Wikipedia but isn't? And give me a couple suggestions if you can find anything. As I've played with that, it's pretty okay.

26:31It needs work and it's not perfect. But if we react with just like, oh my God, we hate AI, then we'll miss the opportunity to do that. And if we go crazy like, oh, we love AI and we start using it for everything, we're going to lose trust because we're going to include a lot of AI hallucinated errors and so on. I mean, that's interesting because Wikimedia writes this yearly global trends report on what might impact Wikipedia's work. And for 2025, it wrote, quote, we are seeing that low quality content is being churned out, not just to spread false information, but as a get rich quick scheme. And it is overwhelming the Internet.

27:07High quality information that is reliably human produced has become a dwindling and precious commodity. End quote. I read that crawlers from large language models have basically crashed your servers because they use so much of Wikipedia's content. And it did make me wonder, will people be using these large language models to answer their questions and not going to the source, which is you? Well, you know, this has been a question since they began. We haven't seen any real evidence of that. And I use AI personally quite a lot. And I use it in a different way, though, different use cases. How? See, I like to cook.

27:53It's my hobby. I fancy myself as being quite a good cook. And I will often ask ChatGPT for a recipe. I also ask it for links to websites with recipes. It sometimes makes them up, so that's a bit hilarious. And I also suggest be careful using ChatGPT for cooking unless you actually already know how to cook. Because when it's wrong, it's really wrong. But Wikipedia would be useless for that. Wikipedia doesn't have recipes. It's a completely different realm of knowledge than encyclopedic knowledge. So, yeah, I'm not that worried about it. I do worry about, you know, in this time when journalism has been under incredible financial pressure and there's a new competitor for journalism, which is sort of low-quality churned -out content produced for search engine optimization to compete with real human-written content.