Since the day Wikipedia launched on January 15, 2001—exactly 25 years ago— pundits have been predicting its inevitable demise. Eric Goldman, a Silicon Valley law professor, confidently declared in 2005 that “Wikipedia will fail within five years.” Three years later, Google rolled out Knol, its own would-be encyclopedia, which was heralded by the press as a “Wikipedia killer.” By 2012, Knol was dead, and Google was quietly piping Wikipedia content into its knowledge graph. Other challengers like Citizendium and Everipedia have had little long-term impact. Meanwhile, the “free internet encyclopedia that anyone can edit” has become the background knowledge infrastructure for everything from search engines and voice assistants to AI chatbots.

Anyone predicting Wikipedia’s death in 2026 should remember the long line of failed prognosticators and competitors. And yet, within some corners of social media, the “death to Wikipedia” chorus sounds louder than ever . . .

Today is Wikipedia’s 25th anniversary, and I was honored to write an article for WIRED for the occasion.

Read the article

More articles on Wikipedia @ 25

• When I attended a Wikipedia conference in NYC last October, I met Darren Loucaides, a journalist who was covering the event for the Financial Times magazine. His FT article is likewise out today: “Wikipedia may be the largest compendium of human knowledge ever created, but can it survive?”
  
  I appreciated how Loucaides captured a troubling incident involving a gun at the conference, using that as a springboard to discuss Wikipedia’s role in the culture. “The internet has made it feel like each of our tribes inhabit different, irreconcilable realities,” Loucaides writes. “And yet somehow, on Wikipedia, people manage to reach a consensus every day. How did that happen?”
  

• Christopher Henner is a 20-year contributor to Wikipedia and its sister projects. Over on Wikimedia, he posted an article, “Wikipedia at 25: A Wake-Up Call,” which makes the case that Wikipedia urgently needs to change course to meet the AI revolution.

  

  From Henner’s piece: “This is not an attack on what we’ve built. It’s a call to defend it by changing it. The Britannica didn’t fail because its content was bad. It failed because it couldn’t adapt to how knowledge distribution was evolving. We have an opportunity they didn’t: we can see the shift happening. We can still act.”

Yesterday’s episode of “Between the Brackets.” YouTube and Audio version.

When Elon Musk launched Grokipedia on October 28, he did so with his usual combative style. The new AI-generated encyclopedia would be a “massive improvement over Wikipedia,” he promised, delivering “several orders of magnitude” more breadth, depth, and accuracy than the website he often derides as too woke.

Since then, my inbox has been full of emails from reporters asking whether Musk’s latest project is just a marketing ploy or represents an actual paradigm shift for internet knowledge. While I don’t yet have a unified theory of Grokipedia, I do have some initial impressions shaped by years of reporting on Wikipedia as a journalist. I’m offering those nine observations here in case others find them interesting, controversial, or useful.

1. Grokipedia borrows extensively from Wikipedia. For a site presented as a dramatic improvement over Wikipedia, Grokipedia sure copies a lot from Wikipedia. A Cornell Tech study found that 56% of Grokipedia entries are wholesale copies of their Wikipedia counterparts, as shown by a disclosure at the bottom of articles indicating that the content was “adapted from Wikipedia” and referring to the Creative Commons license. (Only pages without that footer, like the article for Joseph Stalin, appear to be Grokipedia originals.)
   
   Musk fans may say that they don’t have a problem with his copying from Wikipedia because the site is openly licensed and free to use. Still, there’s something jarring about Musk railing that Wikipedia is “insane” and “trash,” only to roll out a new encyclopedia that copies directly from Wikipedia more than half the time.

I chatted about this briefly for The Washington Post (gift link):

2. When Grokipedia doesn’t copy Wikipedia, things get weird fast. I spent some time comparing entries from the two websites side-by-side, noticing instances where Grokipedia differed from the Wikipedia version. My biggest concern has been sentences with no citation at all. It seems likely that Grok’s AI is hallucinating, and without a reference, there is no way for the reader to check.

   
   PolitiFact covered one of these hallucinations in Grokipedia’s entry on the Nobel Prize for Physics, where the AI encyclopedia asserted that physics is “traditionally the first award presented” at the Nobel ceremony. There was no citation for this sentence on Grokipedia and as it turns out, it’s not true. The Nobel Prize for Medicine is traditionally awarded first. My hypothesis is that Grok’s AI is scanning the internet, noticing a bigger spike in press coverage for the Physics award, and then making an inference that Physics must be the first prize awarded. But of course, just because something seems statistically plausible to an AI system doesn’t mean that it belongs on an encyclopedia, presented to the reader as fact.

   [The original version of this article included a reference to reporting from PolitiFact that Grokipedia contained an error on the page for Nobel Prize for Physics. investigated the issue, using the fact-checking powers of AI, and found that Grokipedia was in fact correct in stating the traditional order of the prizes. PolitiFact conducted additional research and confirmed directly with the Nobel Foundation staff that physics is traditionally first. The PolitiFact article has been updated and I am likewise updating this Substack post to remove the error. I recommend reading Caulfield’s Substack post about the fact-checking powers of AI and his argument that “hallucination” is almost never the proper terminology to use.]

   
   The sourcing problem doesn’t stop there. Grokipedia also links to outlets that Wikipedia would not consider acceptable. According to the Cornell Tech study, Grokipedia includes dozens of citations to Stormfront, a neo-Nazi forum, and Infowars, the conspiracy site run by Sandy Hook denier Alex Jones. Opening the door to alternative media may sound like a free speech philosophical stance, but in practice it means Grokipedia’s algorithm is categorizing extremist propaganda as “authoritative.”

3. Grokipedia prefers first-person sources like LinkedIn. Even on X, where Musk’s most diehard fans are amplified, plenty of users have noted that Grokipedia directly mirrors a person’s LinkedIn page.
   
   Consider the example of Steven Pruitt, the most prolific editor on English Wikipedia by edit count and someone that I have reported on in the past as a notable Wikipedian. Pruitt said that his Grokipedia entry contains several errors that do not appear on his Wikipedia page. For example, his Grokipedia entry lists him as working at Science Applications International Corporation, a job he left more than a year ago. Why the outdated information? Because his LinkedIn profile still says he works at SAIC, and Pruitt freely admits he’s “notoriously lazy about updating it.”
   
   The LinkedIn issue raises the bigger question: What types of sources should an encyclopedia consider a reliable source? Grokipedia tends toward first-person, self-authored platforms like LinkedIn. Wikipedia, by contrast, favors third-party reporting such as newspapers, books, and academic sources. The theory among Wikipedia editors is that third-party reporting is more likely to be neutral; or at least, there is not the inherent conflict of interest of the first-person perspective where the individual would be tempted to self-promote.
   
   And if LinkedIn makes the cut for Grokipedia, does that mean all first-person sources are acceptable? Grokipedia already cites Facebook, Quora, Reddit, and X posts. In the case of Pruitt’s Grokipedia page, it didn’t cite any of the articles that I’ve written that report about him for The Washington Post or Slate. Instead, it cited the versions that I reposted on Medium. In my case, the Medium pieces are syndicated versions of the original stories that went through a fact-checking process. But for the most part, Medium is a blogging platform, meaning it’s the kind of first-person source that Wikipedia generally treats as unreliable.
   
   Whether Grokipedia continues to privilege first-person sources over independent reporting will say a lot about what kind of encyclopedia Musk is actually trying to build.

   ---

4. Grokipedia says very positive things about Elon Musk. Reporting for Slate, Mary Harris noted that “Grok portrays the Tesla CEO through stubbornly rose-colored lenses.” That’s not an exaggeration. Musk’s Grokipedia entry includes a glowing quote from a former SpaceX employee: “Elon was the best mentor I ever had… He’s so sharp, he just picks it up.” The line reads less like a neutral encyclopedia than the recommendations section on LinkedIn.
   
   Musk has long tried to control the text of encyclopedic summaries. Back in 2022, I wrote about how Musk was originally a big fan of Wikipedia but got angry with the site because the lead sentence included the word investor. He wanted that word gone, repeatedly tweeting at editors to remove it, despite the fact that he had invested in companies like SpaceX, Tesla, and Twitter. Musk wanted to be portrayed as an entrepreneur and visionary, not someone who purchased companies founded by others.
   
   Given that history, it’s not surprising that Grokipedia gives Musk significantly more control over how he’s described. Ryan McGrady, a researcher at UMass Amherst, argued in Tech Policy Press that Grokipedia marks a return to top-down control of knowledge. On Wikipedia, neutrality is hashed out through messy, bottom-up debate among volunteers. On Grokipedia, neutrality is whatever Musk (and his AI) decide it is.
   
   As McGrady puts it: “Wikipedia, for all its many flaws, has always aimed to ‘set knowledge free’ … In contrast, Grokipedia’s defining feature as an encyclopedic project is the use of technological power to re-exert top-down authority over information and knowledge.”

   ---

5. Despite Grokipedia’s flaws, the Wikipedia community should not be complacent about AI-generated encyclopedias. In a recent interview series in The Signpost, Wikipedia’s long-running community newspaper, several editors shrugged off Musk’s new internet encyclopedia as simply the latest challenger. Wikipedia has seen plenty of those over the years. Citizendium, Conservapedia, Everipedia . . . So far, none have had any real staying power.

While I appreciate the “keep calm and carry on” attitude, I would urge Wikipedians not to be overly chill about Grokipedia. None of those earlier rivals had the backing of the world’s richest man. More importantly, the realization of an AI-generated encyclopedia represents a potentially significant shift. Going forward, Wikipedia editors and supporters will have to make the case for human effort.

In my view, there are still strong reasons to prefer a human-curated encyclopedia. Selena Deckelmann at the Wikimedia Foundation has laid out some of the arguments previously. But this debate is likely to last decades. As AI gets better at mimicking human processes, Wikipedians will have to explain, over and over, why the human-led, messy, deliberative, consensus-building model works better.

6. My novel The Editors predicted Grokipedia in some ways, but not in others. Once Grokipedia launched in October, I began to get DMs along the lines of: “I read your novel and loved every page of it. And boy oh boy did you foresee where things were going!”

Spoiler alert for anyone who has not read the book. The Editors features a billionaire named Pierce Briggs who creates a for-profit encyclopedia designed to challenge Infopendium, the novel’s stand-in for Wikipedia. Briggs launches Infoveritas to better control his self-image and disrupt the influence of the nonprofit website. When I started writing the book in January 2020, the idea of a billionaire attacking and ultimately challenging a volunteer-run encyclopedia still sounded like an unusual plot twist.

(The Editors is on sale at Amazon and Bookshop.org if you’ve been waiting to read a fictional dramatization of internet fact disputes.)

Despite some similarities, there are a few key differences between the real-life and fictional versions. Musk’s Grokipedia is AI-written, and artificial intelligence is not a major part of the book set during the 2019 to 2020 period. Another key difference is who helps launch the rival encyclopedia. In the novel, Infoveritas arises from a partnership between Briggs and a dissident volunteer editor of Infopendium named Ed Shelton. Grokipedia, at least so far, looks less like a collaboration with Wikipedia editors and more like a Musk solo project.

Which brings me to my next point.

7. Grokipedia shows very little evidence of being influenced by Larry Sanger’s “Nine Theses” on Wikipedia. Sanger is the co-founder of Wikipedia who played an influential role during its first fourteen months in 2001 and 2002. He has spent much of the past two decades trying to create rival encyclopedias such as Citizendium and Everipedia. He has also become one of Wikipedia’s loudest critics.

In September 2025, Sanger published his “Nine Theses” on both Wikipedia and his personal site. The imagery harkens back to Martin Luther nailing his 95 theses on the Wittenberg church door. Sanger sees this as a protest against what he sees as the institutional decline of Wikipedia, together with concrete proposals for reform. The nine theses drew significant media attention for Sanger, including interviews with Tucker Carlson, the Free Press, and The Washington Post (gift link).

Despite the timing, it doesn’t appear that Sanger’s ideas had much of an effect on Musk’s Grokipedia. Sanger’s proposals focus on how to improve the human processes behind Wikipedia, whereas Grokipedia is famously curated and written by AI.

If Musk were to hire Sanger as editor in chief of Grokipedia and recruit a cadre of human editors to further develop the project, that would look more like the plot of my novel. For now, Grokipedia appears to be Musk’s own, in-house AI project. When Sanger offered early criticism of Grokipedia, Musk responded on X that he should “calm down” because the site was still at version 0.1.

8. Sanger’s Nine Theses deserve serious consideration, and many Wikipedians are already debating them. It would be understandable if some editors approach Sanger’s proposals with skepticism. After all, Sanger has launched at least three rival encyclopedia projects since leaving Wikipedia in 2002, and he has spent years criticizing the site on cable news. Volunteer editors might reasonably wonder whether his latest announcement is motivated by a sincere desire to fix Wikipedia or to raise his own profile.

Even so, plenty of Wikipedia editors are taking him at his word. Following the maxim of assume good faith, many Wikipedians are responding to the theses in detail. ranked them in an essay on his Substack, . Sanger himself has emphasized the work he put into the Nine Theses, saying on X that he spent nine months writing them.

Here’s my quick assessment of three of the theses.

Agree. In thesis five, Sanger argues that Wikipedia should retire one of its infamous rules: “Ignore all rules.” Sanger says he originally wrote IAR as a humorous idea to welcome newcomers, but that it now functions as a shield to avoid responsibility. After years of trying to explain Wikipedia’s byzantine rules to outsiders, I find it logically inconsistent that one of the rules instructs editors to ignore the rules. Wikipedia could still encourage newcomers to be bold because that’s more of an attitude than a carte blanche. But IAR is not a true policy in practice and I agree with Sanger about retiring it.

Disagree. Thesis seven proposes allowing the general public to rate Wikipedia articles. For me, this immediately calls to mind Reddit’s systems of upvotes and downvotes. While this idea may work for gauging public sentiment, it seems like a disastrous method for determining reliability. People would very likely organize campaigns designed to downvote articles with inconvenient truths. As Yaron Koren put it on the Between the Brackets podcast, this approach would lead to the Redditization of Wikipedia.

Conflicted. Thesis six calls for Wikipedia to reveal the real identities of its most powerful volunteers, including CheckUsers, Bureaucrats, and members of the Arbitration Committee, which is like Wikipedia’s version of a Supreme Court. To be clear, these users do hold significant authority. Their decisions influence Wikipedia, and through Wikipedia they influence Siri, search engines, and LLMs.

On paper, transparency sounds reasonable, but anonymity is important for practical reasons. Many of the editors in these leadership roles receive harassment, threats, and sometimes worse. I have interviewed several of them over the years, and have seen some of the death threats they receive online. Sanger himself acknowledges the safety issue and suggests that the Wikimedia Foundation should offer stronger support to targeted volunteers. Even so, I suspect that most editors in these positions would rather keep their anonymity than rely on uncertain protection from a nonprofit organization.

At the same time, total anonymity has real drawbacks. I have written about how anonymity can enable conflicts of interest and how easily it can be abused. More recently, location data from X revealed that several major political accounts were operated not by the patriotic Americans they claimed to be, but by troll farms in Nigeria and India. Anonymity creates opportunities for manipulation. The real challenge is finding the right balance between transparency and safety. My guess is that online platforms will continue moving toward a real ID framework. It is an open question whether Wikipedia will follow that trend.

One other note about the Nine Theses. Midway through thesis two, Sanger uses an acronym to describe what he sees as Wikipedia’s dominant worldview: GASP, which stands for globalist, academic, secular, and progressive. I spoke about this GASP acronym recently on the Between the Brackets podcast.

For now, I want to set aside the “P” because whether Wikipedia is progressive is contested, and there are processes on the site for flagging concerning political bias. To me, the other three letters are more interesting.

Should Wikipedia be globalist? The project is written by editors around the world and read by people in virtually every country. It is not meant to mirror a single nation’s perspective in the way that the original Encyclopedia Britannica reflected British sensibilities. The word globalist can be divisive, but it may be a natural consequence of Wikipedia’s mission: to provide everybody in the world with access to free knowledge.

Should Wikipedia be academic? The site relies on the most dependable sources available, and in many fields that means scholarly or peer-reviewed materials. In this sense, Wikipedia is conservative with a small “c.” It tends to weigh academic sources more heavily than the informal, first-person sources discussed above.

Should Wikipedia be secular? Most readers would probably expect it to be. If you look up a religious topic, you expect Wikipedia to describe beliefs rather than endorse them. A faith-based encyclopedia would present doctrinal claims as settled truths. By contrast, Wikipedia presents them as the views of those specific religious communities.

Sanger mentions GASP only briefly, but the framework shows the fault lines for discussion. Each of these dimensions raises questions about the goal for an internet encyclopedia.

9. Musk does not share the same optimistic worldview of Wikipedia founder Jimmy Wales. Unlike Sanger, Wales has remained involved with Wikipedia and the nonprofit that supports it since 2001. His new book, The Seven Rules of Trust, uses lessons from Wikipedia to offer advice on how to build institutions that last. For instance, he argues that organizations should regularly take a “trust inventory” and hold workshops that bring together people of opposing political views.

Wales calls himself a pathological optimist, and the book is suffused with that spirit. He writes about how Annie Rauwerda, the founder of Depths of Wikipedia, shares the joy of quirky things she’s found on the site. He covers the genuine community that forms among Wikipedia editors, both online and in real-life meetups.

A consistent theme of the book is bipartisanship. Wales reminds readers that, “People on the other side of the fence may be opponents sometimes. They are never enemies.” Twenty years ago, that line may have sounded earnest to a fault. Today it feels worth stating and restating.

The contrast between Wales and Musk is striking. Musk recently attacked the beloved author Joyce Carol Oates on X, calling her a “lazy liar” and an “abuser of semicolons,” and even saying that she’s “not a good human.” Her offense, apparently, was suggesting that she knew of an unnamed wealthy man who rarely posts about nature, books, or movies he admires.

Wales, by comparison, is far more likely to treat criticism as an opportunity for reflection rather than a justification for personal attacks. “Most people are still very decent and do care about other people for very good reasons,” Wales writes in his book. “The moment you think, ‘We’re great, we’re fine, anyone who criticizes us is a lunatic,’ you’ve lost it.”

Before we get started, a quick announcement: I’ve written before about Wikiracing, the game where you try to get from one Wikipedia entry to another in the fewest possible clicks. This weekend, there’s an event that’s entirely dedicated to wiki-powered games. The WikiGameJam is taking place October 3-5 at Hex House in Brooklyn. They’ll be playing WikiAsteroids, WikiTrivia, and hopefully designing the next generation of wiki-based fun.

Readers of this newsletter probably already know that the content on Wikipedia is decided by the encyclopedia’s volunteer editors, based on the site’s policies. That’s because Wikipedia is the rare large internet platform that is self-governed by its own user community. If you have a legitimate grievance with how a subject is described on Wikipedia, take it up with the editors. Or better yet, become one yourself.

What’s perhaps less clear is the role of the Wikimedia Foundation (WMF), the nonprofit that hosts Wikipedia and its sister projects like Wikidata. Like many nonprofits, the WMF is governed by a board of trustees who decide how charitable donations are spent and weigh in on the organization’s long-term strategy.

In recent months, some firebrand commentators have tried to paint nonprofits as shadowy cabals. At least in the case of the WMF, the reality is far more democratic. Still, the selection of the board is, as one Wikipedia editor told me, “the most important election on the internet.”

To understand this election better, I spoke with Jonathan Cardy (known on Wikipedia as WereSpielChequers), a longtime editor who recently put his name forward for a trustee seat. Jonathan walked me through the process: the criteria for running, the hurdles of shortlists and community votes, and why someone might devote years of unpaid time supporting the world’s free encyclopedia.

This interview has been edited and condensed for clarity.

Stephen Harrison: Hello, Jonathan. I wondered if you could tell me about the origin of your username, WereSpielChequers.

Jonathan Cardy: If you dismantle it, I’m a lycanthropic spell checker. I pick up the kind of typos that a conventional spell checker can’t pick up. My particular niche is easily confused words. When I first started editing, I would look on Wikipedia for words like “calvary” and “cavalry.” I once spent Christmas going through a list of sports team “mangers” and adding an “a” to make it team “manager.” I’m sure you could take my edits from back then and create an AI to resolve them easily. ..

In any case, that’s why I’ve got a ridiculously high edit count. I’m going through Wikipedia, dealing with a particular typo, and I can make a large number of small changes relatively quickly. As opposed to when I am writing a couple of paragraphs about a historical topic, and there may be an hour or so of reading behind one edit.

When did you first start editing Wikipedia, and what inspired you to consider running for the WMF’s Board of Trustees?

I’ve been a volunteer editor since 2007—mostly on Wikipedia and Wikimedia Commons, and at first I was fixing typos and making small corrections. It’s been a hobby of mine for 18 years now. I also worked part-time for Wikimedia UK in 2013/15 as an outreach person with galleries, libraries, archives, and museums. Beyond that, I’ve served as a trustee for other charities, including one with an endowment not far off the size of Wikimedia’s. So when I saw there was an open seat on the Foundation board, it felt like a natural next step.

[Stephen note: The WMF Board consists of up to 16 Trustees, with one founder seat for Jimmy Wales, five appointed seats, and six seats selected by Wikimedia communities and affiliates. Jonathan was running for one of the community seats.]

One Wikipedia editor told me that this is the most important election on the internet. What’s the process for someone running for a community seat on the board of trustees?

There are four stages. First, you need to meet basic criteria—like having a certain number of edits—and submit an application. From there, if more than 10 people apply, Wikimedia’s national chapters and affiliates help shortlist candidates down to six.

This year there were 12 candidates. I didn’t advance beyond the shortlist, so I never made it to the community-wide election. But for those who do, the next step is a vote among volunteers, technical contributors, and some staff.

[This year the online voting period begins on October 8 and will be open for two weeks.]

Finally, there are reference, background checks, and social media checks before the winners take their seats.

[Stephen note: Just before this article’s publication, I received word that two candidates for the community seats have been dropped by the WMF in their vetting process. One of the candidates, Lane Rasberry, posted his initial response here, where he noted that part of the reason he was removed was for speaking publicly to journalists like yours truly. That leaves four candidates for the two remaining seats: James Alexander (American), Michał Buczyński (Poland), Wojciech Pędzich (Poland), and Bobby Shabangu (South African).]

The board’s work isn’t paid. Why would someone want to join?

Trustees make important decisions about how money is spent, what policies apply across all Wikimedia projects, and issues like Creative Commons licensing and attribution. The board also drove the Universal Code of Conduct, which sets standards across the movement. So while you don’t get paid, you do have influence on how the world’s knowledge projects are governed.

Subscribe now

If you had advanced to the shortlist and been elected, what kinds of proposals would you have brought to the board?

One of the criticisms I had of the current structure is that we have this complicated arrangement where some countries, like Germany and Switzerland, have their own fundraising powers and quite robust organizations, while others—like the UK—are much smaller because the global Wikimedia Foundation handles the fundraising. And there actually isn’t a Wikimedia USA organization, only city or regional groups like NYC or DC.

One of the suggestions I made was to consider a more federal model globally, where countries that meet certain standards could have their own national organizations. Effectively, the Foundation would split into a U.S. entity handling American fundraising, and a global organization based either in the U.S. or elsewhere. That kind of decentralization could protect the movement if, say, the U.S. political climate turned hostile toward Wikipedia.

Last question. Because of your background with libraries and archives, I’m interested in your thoughts on dead links and the “Digital Dark Age” theory.

Yes, many sources that Wikipedians relied on 10 or 20 years ago are already gone. Dead links are a constant issue. That’s why partners like the Internet Archive are so important because they preserve snapshots of the web.

If we thought we were genuinely going into a cultural dark age, giving up computers for a few generations, then I think I’d be back to books. The Doomsday Book is my example, compiled 950 years ago. [Note: Pronounced “doomsday,” the Domesday Book was a survey of England by William the Conqueror completed in 1086 and preserved in parchment.] They made several copies with high quality material and deposited them with institutions they thought would last. Most were lost, but a handful survived, and we still have the Domesday book.

Wikimedia could play that same role for digital culture. The Foundation’s endowment should be large enough to guarantee that Commons and Wikisource will be around for the foreseeable future.

This article originally appeared in Slate on September 26, 2023.

The final round required navigating from the Wikipedia article on “Stroopwafel” to the page about “Jimmy Wales,” Wikipedia’s co-founder—a tough leap for even the most seasoned Wikiracers. As the challenge appeared on their screens, the contestants leaned into their laptops and tried to quickly strategize their path across the internet encyclopedia. Whoever made it to “Jimmy Wales” from “Stroopwafel” with the fewest clicks would be crowned champion.

Wikiracing has long been viewed as a quirky, low-stakes pastime for friendly nerds—a world away from the high-pressure environment of competitive Scrabble or speedcubing. But while those other “geek sports” have already established their own versions of a Super Bowl, Wikiracing has traditionally remained confined to college dorm rooms and high school computer labs.

Not anymore. Wikiracing was a featured event at this year’s Wikimania, the global conference for dedicated Wikipedia editors, which took place this past August in Singapore. “People took Wikimania 2023’s Wikiracing very seriously, and the level of competition was incredibly high,” said Zack McCune, director of brand at the Wikimedia Foundation. “We had F1 Grand Prix energy in the room.”

Although there are several variations of the game, the basic premise is that players start with one Wikipedia article and then try to reach a target article using the fewest possible links. This year’s players used the platform TheWikiGame.com to conduct game play and were allowed to submit as many solutions as possible within a two-minute period.

Kevin Payravi, a 28-year-old software engineer from Frisco, Texas, emerged as this year’s champion, and proudly posted a photo of his first-place trophy with the caption “I’ve been training my whole life for this moment.” According to Payravi, his training began sometime in middle school. He made his first Wikipedia edit in 2007 at age 12, when he registered his user account, SuperHamster (he admits he’d choose a different name now). “We’d be in study hall in the library at school, and we’d play Wikiracing on the school computers,” Payravi said. “Since usually the school would block gaming websites, Wikiracing was kind of a way around that because we could always access Wikipedia.”

Subscribe now

When he’s racing, Payravi tends to look out for geographical links—navigating from “Stroopwafel” to “South Holland,” for example—because place articles usually open up new paths. He also considers Wikipedia’s Manual of Style for linking, which lets him anticipate the probability that a word or phrase is likely to be linked versus ordinary text.

I have fond memories from my college days of Wikiracing with my friends in the computer lab, where we’d sometimes attempt to be the first to get to “Philosophy” in six clicks or fewer. On a whim, I tried again this afternoon, starting with the randomly generated article “Gyöngyösmellék,” clicking “Hungary,” then “Roman Empire,” then “Philosophy.” Three clicks—piece of cake! But, according to Wikipedia researchers, it’s not just skill: Clicking the first link of any English Wikipedia article and repeating the process leads to “Philosophy” about 97 percent of the time. Although we might need more philosophy in public discourse, the subject is literally permeating all of Wikipedia.

For the uninitiated, it might be hard to grasp why Wikiracing is so enjoyable, so I asked the top talent to explain their theories. “I think it’s just exciting to see that almost everything in the world is related,” said Annie Rauwerda, the founder of the popular Depths of Wikipedia social media accounts and this year’s third-place winner. The world is so big, human knowledge is so vast, it’s all so overwhelmingly beyond what humans are capable of understanding and yet … it’s all so related. The world of human knowledge (as summarized on Wikipedia) is more connected than we think, and Wikiracing reflects that.

There’s also the personal thrill of the chase. “Everyone likes going to Wikipedia, reading an article and clicking on interesting links and going down that rabbit hole,” said Payravi, this year’s champion. “This kind of plays off that, trying to click links and find an interesting path.”

Effective Wikiracing rewards lateral thinking—connecting subjects without an obvious relation—while concentrating on a predetermined target. But notice how the same behaviors that help people sprint across Wikipedia in record time can seem less virtuous in other contexts. After all, conspiracy theorists use “lateral thinking” to connect unrelated events and string together false narratives that range from silly (Lea Michele was a child actress, she preferred not to improvise on Glee, therefore Lea Michele can’t read) to dangerous (pharmaceutical companies made money from vaccines, therefore Big Pharma “planned” the pandemic).

Those connections may be false, but people seem to really enjoy making them. Neuroscience suggests that dopamine plays a bigger role in wanting a reward than in actually receiving it, which might explain why the human brain prefers to hunt for information instead of having it delivered conveniently without pursuit. The people chanting “Do your own research!” seem to have very fond memories of their personal search process, finding their own interesting pathways across the net.

Fortunately, Wikiracing is more wholesome than all of that. And even though it has its own Wikipedia page, I think it could use a bit more publicity. Other Wikipedians agree: “I was surprised that in 2023, some prolific Wikimedia contributors did not know the thrill of a Wikirace!” said Rauwerda, the Depths of Wikipedia creator. “Not everyone there had spent high school lunch breaks speedrunning the encyclopedia like I did.”

Last week I wrote for Slate and joined WBUR’s On Point to discuss the latest partisan attacks on Wikipedia. The Wikipedia editing community doesn’t always get it right, but journalists nevertheless have a responsibility to describe the site’s editorial process accurately.

Today, I’m expanding to a new reporting beat: investigating the importance of paper in an age dominated by screens.

Sam Altman wasn’t the kind of guest you’d expect on ’s podcast How I Write. The show usually spotlights bestselling authors sharing tips for overcoming writer’s block, developing a unique voice, and revising drafts. Altman, by contrast, is the CEO of OpenAI, the company behind ChatGPT, and one of the most recognizable faces of the AI boom. He’s the sort of Big Tech executive lawmakers invite before Congress to talk about AI risk, not consulted for his personal views on the craft of writing.

Once on the mic, Altman did what he often does in interviews (and what you’d expect from a man with billions riding on the outcome): he sold the promise of AI. He described ChatGPT as a “general purpose tool” he now uses daily, predicting that soon he’d rely on it “for most things I do.” The impression was of a man happily tethered to his screen, using digital tools to generate the next big idea.

That’s what makes the back half of the interview so striking. About 43 minutes in, Perell asks Altman if he takes notes during the week. Altman’s green eyes light up: “I’m a huge notetaker.” Perell presses him: “Ooh, tell me about that.”

Suddenly the AI pitchman turns into a fanboy gushing about his favorite paper and pens. “You definitely want a spiral notebook because one thing that’s important is you can rip pages out frequently. You want it to lie flat on the table, with a hard front and back. And you also want something that can fit in a pocket,” Altman says, using his own notebook to demonstrate. The spiral, he explains, is crucial so you can rip up pages and place them side by side. He’s equally opinionated about pens, declaring the Uniball micro 0.5 the “best pen overall” with Muji’s 0.38 dark pen coming in second.

Handwriting isn’t an occasional hobby for Altman but a daily practice. He fills a fresh spiral every two weeks, leaving his housekeeper to clear away the piles of crumpled pages. He says he’s always scribbling new observations, half-formed ideas, notes to self, and strategies to test.

Altman’s enthusiasm for spiral notebooks is so endearing that it almost obscures the irony. His job as the CEO of OpenAI is to promote the company’s AI tools to billions of potential customers, promising the transformative benefits of this digital technology. But when he does his own deep thinking, his style is decidedly analog.

Perhaps there’s a lesson here: even in the age of AI, the sharpest minds still reach for pen and paper to discover good ideas.

If you enjoyed this, you might like my post about the psychological reasons to prefer print.

Subscribe now

Take the Netflix movie KPop Demon Hunters. When it premiered on June 20, its Wikipedia article was on the shorter side, supplying the cast list and a few early reviews. Fast-forward to today: the current page runs more than 10,000 words, with a sprawling plot summary, sections on themes and public reception, and 104 citations. Last month, the article got nearly 2.3 million page views. As the movie took off in popular culture, becoming Netflix’s most-watched film, the Wikipedia page followed suit.

If that’s how it works for a movie musical in 2025, does the same dynamic apply to historical pop songs?

That’s one of the questions posed by music writer (and fellow Substack author) Chris Dalla Riva. For his forthcoming book, Uncharted Territory, he listened to every single Billboard #1 song, built a massive dataset, and turned it into a data-driven history of popular music from 1958 to 2025.

Preorder his book "Uncharted Territory"

Throughout the book, Chris explores one of his personal theories: Pop songs from the late 1980s have left a much smaller cultural footprint than those from the first half of the decade.

To test this, Chris turned to Wikipedia, comparing the size of the respective Wikipedia articles and the traffic they receive. (In other words, the KPop Demon Hunters analysis described above.)

It turns out that early ’80s songs like Blondie’s “Rapture” (1981), Michael Jackson’s “Beat It” (1983), and Madonna’s “Like a Virgin” (1984) tend to have long and well-trafficked Wikipedia pages. By contrast, later ’80s hits like Richard Marx’s “Satisfied” (1989) suffer from shorter entries and fewer page views. The pattern suggests that early ’80s songs indeed loom larger in our cultural memory.

The People Behind the Pages

I decided to push Chris’s investigation further by talking to the people who actually compile these Wikipedia articles for #1 hit songs. Some of my sources provided their real names, but a few asked me to stick to usernames, so I’ve done that throughout for consistency.

Despite the Harry Potter-influenced handle, Dobbyelf62 spends little time on wizarding pages. Instead, they devote their energy to music articles, including Fleetwood Mac’s “Dreams.” Over time, Dobbyelf62 and other editors layered in references, chart data, and additional context. But even old hit songs can have dynamic Wiki pages. “When a song experiences a resurgence in popularity, often due to an internet meme, an issue arises on how to incorporate this information into the article without giving it undue weight and succumbing to recency bias,” Dobbyelf62 told me. Case in point: “Dreams” shot back onto the charts after it was featured in a TikTok clip of a man skateboarding while drinking Ocean Spray—a fact that is now mentioned in the article’s lead section.

Other editors keep editing the same song for years. User Synthwave.94 first edited “Billie Jean” because it was their favorite song, then kept coming back as they deepened their knowledge, adding certifications, categories, and sources over the next decade.

User Hzh, who worked on Cutting Crew’s 1987 hit “(I Just) Died in Your Arms,” told me fans often try to slip in unsourced tidbits or original research. Then there are the “genre warriors” who spar endlessly about how to categorize certain tracks, and sometimes earn bans for poor behavior.

And then there’s Binksternet, a user who helped expand Steve Winwood’s “Higher Love.” Binksternet got involved with this page after improving the Wikipedia article for John Robinson, the session musician who plays the drum solo at the start of the 1986 hit song. “I actually communicated with JR regarding his wiki bio, which was a fun experience,” Binksternet told me.

Subscribe now

I find Winwood’s “Higher Love” to be interesting because it is an exception to Chris’s general rule about the second half of the ’80s. The article remains unusually popular for a late 1980s song, continuing to draw traffic. In fact, the page views spiked to several thousand per day in August 2019. The most likely explanation is that the electronic artist Kygo released a cover version in 2019 that incorporated Whitney Houston’s vocals and introduced the song to a new generation of listeners.

This exception illustrates one of Chris’s main points in the book: cultural memory is fluid. The piece of content that looks forgotten today can return tomorrow with the right remix or spotlight of attention.

***

This is a free newsletter. If you’re interested in supporting my writing, please consider buying a copy of my Wikipedia-inspired novel The Editors or giving it a quick online review. I also have a new Ko-fi page if you’d like to buy me a coffee.

For several years now, I’ve been intrigued by Wikidata, the structured knowledge base that underpins Wikipedia and many other free knowledge projects. I’d heard about it at Wikimedia events, and even enjoyed a bit of Wikidata cake to celebrate the project's birthday. Still, the practical use cases always felt a bit abstract. Wikidata is the data-driven dimension of Wikipedia—I understood that—but what does that actually mean in real-world terms?

So I was glad to have a short conversation with Lydia Pintscher, Portfolio Lead for Wikidata at Wikimedia Deutschland, to explore that very question. Lydia walked me through how Wikidata works, what makes it different from Wikipedia, and why it matters far far beyond the immediate Wikipedia community. I’m happy to share our discussion with you.

This interview has been edited for length and clarity.

Stephen Harrison: Hello, Lydia. Could I ask you to please introduce yourself?

Lydia Pintscher: Hi everyone. I’m Lydia and I’m the portfolio lead of Wikidata and I work for Wikimedia Deutschland.

Harrison: Back in 2019, I attended Wikimania, the annual user conference for Wikipedia [and the related free knowledge projects]. This was the year it was hosted in Stockholm. And I remember first of all the enthusiasm at the event, and also that there was a large cohort of people from Germany who were involved with Wikimedia Deutschland. Can you tell us more about that organization?

Pintscher: Next to the Wikimedia Foundation, it’s the biggest organization in the Wikimedia movement and the first official chapter. It’s been around for a long time now supporting the German language Wikipedia. And it’s mainly responsible for Wikidata, which is Wikimedia’s knowledge graph.

Harrison: Tell me more about Wikidata and how it’s different from Wikipedia.

Pintscher: So Wikidata is an open project where people contribute similar to Wikipedia, but instead of text, we are talking about data. For example, people are collecting data such as the number of inhabitants of a country, or the current head of government for a country. The Wikidata entry on a person might state their profession or their data of birth. And all of that is stored in Wikidata, available as a resource for anyone to use, and contribute to as they do with Wikipedia.

Harrison: My mental image for Wikidata is something like this. You have a Wikidata item for London and that includes the population of the city. And then that data can flow out of Wikidata and into the different language editions of Wikipedia, into English Wikipedia, German Wikipedia, Spanish Wikipedia, and so forth. That way there can be some sort of consistency.

Pintscher: That was the original idea that we started with in 2012. There were about 300 different language editions of Wikipedia. [Note: Each language of Wikipedia is separate and is managed by its own community of volunteer editors.] Some editions were very large with lots of editors taking care of that content, writing articles and maintaining the information. And there were also very small Wikipedias where there were just a handful of people involved.

Let’s say that someone is elected to the role of mayor of Berlin. That would be a lot of work to update it in the 300 different language editions of Wikipedia, and there may not be enough people on the smaller wikis to make that change. It’s also not very meaningful work to make that change over and over. So we thought: why not just have one place where that gets updated and then the other Wikipedia editions can pull from there? That’s basically how Wikidata started.

Harrison: Have you found that the contributors to Wikidata overlap with the editors of Wikipedia?

Pintscher: Yes, there’s a lot of overlap. But there are also people who have found their place in the Wikimedia movement through Wikidata specifically. They preferred to work with data and were not so excited about writing articles.

Harrison: When I was first looking into Wikidata, I thought that you had to be a programmer to contribute to the project. I found that, actually, the properties in the structured data were pretty easy to understand. You can just point at and edit the data fields.

Pintscher: Exactly. Sometimes it’s much easier to contribute to Wikidata because you’re making those small contributions. As opposed to writing something longer, you can just add a data point. You’re updating the mayor of a city, you’re adding a rating for your new favorite movie… Whatever floats your boat.

Harrison: I understand there are a lot of different use cases for Wikidata—even several that extend far beyond the world of wikis.

Pintscher: Yes, the data from Wikidata is spreading our knowledge even further. One example that I find really cool in the public policy sector is a project called Govdirectory, which was started by two Wikidata editors. Govdirectory provides an easy way for people to figure out how they can contact their government for a specific topic that they care about. For example, it would provide an easy way to find an official government social media account. You could find who in your government is responsible for an environmental matter. Govdirectory is based on the information in Wikidata about how the government of a country is structured and the various contact points.

Harrison: For a project like Govdirectory, does the Wikidata volunteer manually put that information into Wikidata?

Pintscher: Wikidata contributors do the work. [laughs] Some of its handcrafting. But there are a lot of tools like “Mix’nmatch” to support the editing. With Mix’n’match, you can match a Wikidata entry to an existing catalog. There are also mass editing tools that can allow the user to bring a recent governmental report into Wikidata.

Harrison: I would expect that in some countries, the information for contacting government officials may be more easily accessible, but it would be less so in others.

Pintscher: That’s right. We have a project called Every Politician that has a registry of data for all the politicians for all the countries in the world. Because this is something journalists and activists need. It was so important to have an up-to-date list. We’ve done a lot of work to ensure that this pipeline of data is open and accessible to people.

Harrison: Your Wikidata team also told me about this project called EqualStreetNames.

Pintscher: Yes, the theory behind EqualStreetNames was to bring attention to the fact that when we name streets, we very often name them after famous men. There’s a lot of reasons for that, and it’s something we wanted to bring attention to. So if you go to the EqualStreetNames project, you can get a map for your city based on the data found in Wikidata. Then you can get a colored map of how many of your city streets are named on men versus women. Then somebody did an experiment on top of that, which was to see if they could travel only following streets that were named after women.

Harrison: Oh wow. Was that even possible? I mean, did it even work?

Pintscher: I wouldn’t want to travel like that, let’s say. It would be very difficult.

Harrison: It would be challenging to lug your books around like that. And I heard that Wikidata has a project that’s specifically dedicated to physical books?

Pintscher: Yes, there’s the book sharing network Inventaire. The idea is to support the sharing economy for books by registering which books you have. Without Wikidata, building something like this would be very difficult, because you’d have to start from scratch in putting in the data points: the author, the genre, the publisher, and so forth. But with Wikidata, you can have the system pull up an author, say, Douglas Adams, and it will bring in a blurb about him and his author photo.

Harrison: Funny that you mention a sci-fi author because I was just thinking about how we haven’t yet talked about AI yet and that seems like a miss on my part. From what I’ve read, the AI apps like ChatGPT pull a lot of their information from Wikidata.

Pintscher: Since the beginning of Wikidata, we’ve wanted it to be both human and machine readable. The data should be readable to a human, but it might reach a wider audience if it is also readable and usable by a machine.

Harrison: What language is the machine using to read Wikidata? Is it different from the human-readable text?

Pintscher: All of Wikidata’s data is published following linked open data standards. So, for example, you can get an RDF [raw data file] export, which may be easier for certain tools to use. Or you can get the data in a JSON export and use that to build, whichever type you need.

Harrison: Seems like Wikidata was set up in a very smart way since the beginning! Can you speak about any challenges that Wikidata is having as a project?

Pintscher: I would say one challenge is that Wikidata is sort of a background project. It’s seen as the data backbone to Wikipedia and other projects, but it’s not always in people’s faces. People don’t necessarily know where the data is coming from. So we’ve started a campaign to encourage the organizations that are building applications with Wikidata to name Wikidata as a source for the data that they’re presenting.

Another issue is scalability. Right now Wikidata has about 116 million entries but only about 25,000 people who contribute every month. That’s a lot of data to keep up to date!

Harrison: If you had to make a pitch for why someone should volunteer their time to Wikidata, what would you say?

Pintscher: First of all, it’s an amazing project with an amazing community, of course. But more seriously, it is the source of so much of the technical infrastructure that you use everyday without you even being aware of it. Your contributions to Wikidata spread so much further than a lot of contributions you could make in other places. So if you’re looking to really have an impact, go to the source.

Thanks to Lydia Pintscher for her insights. You can read more about Wikidata at its intro page.

I’m most known in online circles as a journalist who has written lots of news stories (and even a book) inspired by Wikipedia—a website that exists in part because people decided that print encyclopedias were too bulky, too gatekept, and too slow to keep up with the pace of knowledge.

I’m also a tech lawyer by trade. My area of practice? Online platforms and IT transactions.

In short, I’m the last person you would expect to be making a case for printed works. For paper books, magazines, and newspapers. For all these old-school, analog materials.

And yet here we are.

In an age ruled by short-form video and algorithmic curation, I find myself more and more pulled toward traditional paper and ink.

Here are three psychological reasons that I’ve come to believe that print is uniquely good for us, and that we turn away from printed works at our peril:

1. Print gives the reader more control. Ray Bradbury’s dystopian novel Fahrenheit 451 is set in a dystopian world where all books are burned, and TV dominates the culture. People spend their lives at home watching television on giant “parlor walls” that bombard them with nonstop entertainment. Everyone is obsessed with their personal screens. The main character’s wife is petrified of leaving her house for fear of missing a moment with her on-screen “family.”

   Although Bradbury wrote his book in 1953, there’s a natural parallel to today’s screen addiction. Whether it’s scrolling TikTok or Instagram Reels on your phone, or spending hours watching cable news, most video content is designed to discourage reflection and keep you watching.

   
   Sometimes in the morning, or late at night, I have found myself getting sucked into scrolling shortform videos. I only meant to watch a few minutes, but before I know it, I’ve wasted an hour staring at the tiny screen. In moments like this, it can certainly feel like the phone has godlike powers, and I’m a mere mortal subject to its cruel whims.
   
   I never feel this way when I’m reading physical books. Sure, successful authors encourage readers to keep turning pages. But we know deep down that printed works tend to be less addictive than video. Print is less likely to ensnare you. Bradbury has a character address this difference in his novel:

   “You can shut [a book]. Say, ‘Hold on a moment.’ You play God to it. But who has ever torn himself from the claw that encloses you when you drop a seed in a TV parlor? It grows you any shape it wishes! It is an environment as real as the world.”

   Isn’t that a beautifully empowering thought? You can play God to a book. By the way, the latest neuroscience supports Bradbury’s statement: Digital media is more likely to turn us into dopamine addicts. By contrast, reading a book is a less stimulating but harder task. And that ultimately leaves us feeling more fulfilled and in control.
   

2. Print helps readers concentrate, leading to higher comprehension. We all know that electronic devices are more likely to distract us from an immersive reader experience, pinging us with advertisements, email notifications, or the temptation to check on social media. The WiFi connection on a Kindle shows the other books on sale at Amazon—enticing us to do anything except for focus on the text.

   That’s why the author and famous social media minimalist Cal Newport has a strong preference for physical books. Yes, he’ll sometimes deviate and read a book on Kindle (but not the iPhone), but that’s only on the rare occasion that he doesn’t have time to wait for delivery of a physical book. And that’s a very limited exception. As he wrote on his blog, “For demanding books, I’ll almost always use physical, as I have an easier time mustering concentration when I have a physical object to manipulate.”
   
   But does enhancing concentration lead to greater understanding of the text? I was under the impression that there was still significant debate among experts about whether people learned more from physical books than e-readers. While some critics say that studies with a larger sample size still need to be conducted, the research so far suggests a verdict: Reader comprehension is significantly better with physical books than e-readers.

   It seems like physical books work better for us precisely because we are embodied beings. Holding the weight of the book in your hand, turning the pages, writing notes in the margins, and highlighting your favorite passages—these are all sensations that are experienced in the body and thus leave a more lasting impression in our brains. According to researchers, turning pages as we read creates an “index” in the brain, mapping what we read visually to a particular page.

   In other words, the indelible ink characters on a page help us to create a mental map. That’s not the case with the flickering ghost characters on a screen. On-screen text appears and disappears without a fixed location, which makes it harder for our brains to index.

3. Print may help cure smartphone addiction. I have had an intuition for a while now that one way to improve our increasingly short attention spans is by reading from a very long novel. I’ll speak for myself—since I’m feeling increasingly limited in my ability to concentrate, I’ve tried building up my endurance by reading George Eliot’s 800-page tome Middlemarch.

   Notice how I was thinking about the issue of concentration as if it’s a muscle that had weakened over time because of smartphones and therefore needed more exercise. I did not frame the smartphone problem as an addiction in need of a cure.

   That’s why this latest study caught my attention: According to research from Catholic University of Indonesia, printed media is more effective than social media in reducing the smartphone addiction levels of participants. Could old-fashioned print help cure the smartphone zombie epidemic?
   
   I took a look at the research paper itself to see what I could glean beyond the headline. The study placed 54 subjects into two test groups that were each given a cognitive behavioral therapy (CBT) course that covered topics such as “negative impacts of excessive use of smartphones” and “steps to reduce smartphone usage.” Like most CBT programs, the lessons focused on paying greater attention throughout the day and adopting healthier beliefs about the technology.
   
   The only difference between the two test groups was in the type of media used to deliver the lessons to recipients. One group got the training via social media and the other through printed media—specifically leaflets.
   
   And the leaflets won the day! While both cohorts saw a statistically significant decrease in smartphone addiction rates, the group receiving leaflets (printed media) saw greater improvement than those who got the training digitally.

   The study’s authors caution that more research is needed in this area, but I suspect their preliminary research is already pointing to an important theme: We experience a lot of our current social ills because of the ethereal world, the world of websites and apps. But just because that’s where the disease originates doesn’t mean that’s where we’ll find the cure. We’re more likely to draw our medicine from the material world. The world of leaflets, the world of print.
   

   ---

   
   If there’s interest, I’d like to continue this series and meditate on the social and cultural virtues of printed works. Please let me know what you think!
   

   Subscribe now

   
   On a different note, I’ve had the chance to talk about Wikipedia and information ecosystems in a few recent places:

   • Today, Explained (Vox’s news explainer podcast) - “What did Wikipedia do?” (May 9, 2025)

   • Connecticut Public Radio (The Colin McEnroe Show) - “What Wikipedia can teach us” (May 14, 2025)

   • The Washington Post - “Wikipedia’s nonprofit status questioned by D.C. U.S. attorney” (April 25, 2025)

     ---

Wishing you and yours a happy Memorial Day weekend.

This list originally appeared on my personal website. I’m copying it here for newsletter readers.

Since 2018, I have written several journalistic pieces covering Wikipedia, and I write a recurring column on the topic for Slate.

Below I have compiled some of my favorite journalistic articles about Wikipedia, organized by topic. You can access my more recent coverage through my newsletter.

Wikipedia, General Background

• '“Wikipedia is twenty. It’s time to start covering it better.” Columbia Journalism Review. January 14, 2021. Co-authored with Omer Benjakob.

• “From Anarchy to Wikiality, Glaring Bias to Good Cop: Press Coverage of Wikipedia’s First Two Decades.” MIT Press. October 15, 2020. Co-authored with Omer Benjakob.

• “Happy 18th birthday, Wikipedia. Let’s celebrate the Internet’s good grown-up.” The Washington Post. January 14, 2019.

International Censorship of Wikipedia

• “Why China Blocked Wikipedia in All Languages.” Slate Magazine. May 21, 2019. (“…the timing of the block seems to coincide with the Tiananmen Square protests, which ended violently on June 4, 1989”).

• “Wikipedia Has Been Unblocked in Turkey, Finally.” Slate Magazine. January 29, 2020. (“But what worked there may not help in other countries that censor the encyclopedia such as Venezuela”).

Wikipedia in the Midst of the Pandemic

• “The Coronavirus is Stress-Testing Wikipedia’s Systems—and Editors.” Slate Magazine. March 19, 2020. (“The deletions, the editor fights—all of it will be important to researchers studying this period”).

• “Future Historians Will Need Access to Coronavirus Misinformation.” Slate Magazine. May 20, 2020. (“How does the internet’s encyclopedia stay accurate with news and knowledge changing by the hour?”).

Profiles of Wikipedia Editors

• “If You See Something, Write Something.” The New York Times. March 28, 2018. Ryan Ng (user EpicGenius) and Shaul Picker (Kew Gardens 613) have made extensive contributions to the Wikipedia articles about the New York Subway system.

• “The Wikipedia contributor behind 2.5 million edits.” The Washington Post Magazine. October 2, 2018. Steven Pruitt is the most prolific Wikipedia contributor to English Wikipedia.

• “Closing Wikipedia’s Gender Gap.” OneZero. March 27, 2019. An interview with Wikipedia editor Rosie Stephenson-Goodknight, co-founder of Wikipedia Women in Red.

Wikipedia’s Impact on Culture & Society

• “How Wikipedia Became a Battleground for Racial Justice.” Slate Magazine. June 9, 2020. (“Contributors are rethinking what Wikipedia’s commitment to neutrality actually means.”)

• “How Wikipedia Can Influence Judges’ Legal Writing.” Slate Magazine. August 17, 2022. (A new study suggests Irish judges are relying on Wikipedia.)

Internal Politics of Wikipedia & Its Editors

• “Why Wikipedia Banned Several Chinese Admins.” Slate Magazine. October 26, 2021. (“What type of user would be most willing to take the risk and jump through all of those hoops: a pure free-knowledge enthusiast, a Chinese propagandist, or someone in between?”).

• “Wikipedia’s ‘Constitutional Crisis’ Pits Community Against Foundation.” Slate Magazine. July 2, 2019. (“The banning of a single editor named Fram is roiling Wikipedia”).

Public Service Announcements

• “Yes, Copying from Wikipedia is Plagiarism.” Slate Magazine. January 19, 2024. (“The question of whether Neri Oxman’s Wikipedia use is plagiarism has a pretty simple answer”).

• “WikiLeaks Is Not a Wiki.” Slate Magazine. May 19, 2019. (“The false use of the prefix has hurt the Wikipedia brand”).

• “Wikipedia is facing an existential crisis. Can gen Z save it?” The Guardian. September 12, 2024. (“Fortunately, gen Z doesn’t seem inclined to immediately surrender Wikipedia to AI”).

Hopeful Takes

• “Wikipedia is Helping Keep the Welsh Language Alive Online.” Slate Magazine. August 7, 2019. (“Welsh and other smaller language movements on Wikimedia projects suggest there may be ways to train technology to allow for cultural differences.”)

• “Wikipedia Will Survive A.I.” Slate Magazine. August 24, 2023. (“Rumors of Wikipedia’s death at the hands of ChatGPT are greatly exaggerated”).

The Fun Side of Wikipedia

• “The Art of Wikiracing.” Slate Magazine. September 26, 2023. Getting from one random Wikipedia entry to another in the fewest possible clicks.

• “The Battle for Curling’s Wikipedia Page.” VICE Sports. February 16, 2018. There is an edit war raging behind the scenes of curling's Wikipedia page. A lot of it has to do with trolls claiming it's not a sport.

• “Why Wikipedia Is So Tough on Bigfoot.” Slate Magazine. February 13, 2023. (“The two sides of Wikipedia’s Never-Ending Cryptozoological War could both stand to make some concessions”).

Subscribe now

I had the pleasure of briefly appearing on German Public Radio (ARD) for the Morning Information Show on May 7, 2025. Hearing my voice overdubbed with a German translation was quite an experience, almost as if I were addressing a U.N. committee. Below, you'll find the English language transcript of that segment.

Birgit Eger (ARD journalist): More than 280,000 volunteer contributors write Wikipedia articles every month in more than 30 countries and over 20 languages. The online encyclopedia is the eighth most visited website on the World Wide Web. Its scale makes it appealing—not just to readers, but also to politicians. The idea is that when many editors pool their knowledge, the result should be as neutral and fact-based as possible.

But not everyone likes that. Edward Martin, a Trump-appointed U.S. attorney for Washington, D.C., has now threatened to revoke Wikipedia’s tax-exempt status and accused the encyclopedia’s creators of manipulating information. He’s sent a questionnaire and wants a response by May 15.

Criticism of Wikipedia has existed since the platform was founded in 2001, notes journalist, author, and lawyer Stephen Harrison. But this time, it feels different.

Stephen Harrison: I never thought this would happen in America—that the government would go after Wikipedia. In the eight years I’ve reported on Wikipedia as a journalist, these sorts of examples came only from authoritarian countries like China, Russia, or Saudi Arabia. As an American, I find it truly shocking that this is now happening here.

Birgit Eger: The current criticism from Trump’s legal appointee focuses on Wikipedia’s coverage of the Gaza conflict, which he claims is antisemitic. But other organizations are chiming in too. The conservative, Trump-aligned Heritage Foundation has accused Wikipedia of abusing its power and offers tips on how to unmask the pseudonyms of volunteer contributors.

Last year, Elon Musk also lashed out at Wikipedia, calling for a boycott of donations. That’s a stark contrast to his attitude during Trump’s first term.

Stephen Harrison: In 2017, Elon Musk was praising Wikipedia as a great project. And now, both Musk and Trump have turned against it, using it as a target in their culture war. That’s partly because Wikipedia doesn’t cater to the far-right perspective—or, as we say in the U.S., the conservative MAGA viewpoint.

Birgit Eger: But Wikipedia editors would argue they are maintaining a neutral stance.

When the online encyclopedia was founded, it carried the hope of being a democratic, egalitarian platform. The aim was to share knowledge without commercial interests. Wikipedia’s editorial debates are conducted openly—and sometimes, contributors are actively recruited to bring in new perspectives on a topic.

Now that right-wing politicians, lawyers, and business leaders are putting heavy pressure on Wikipedia, that pressure is also being felt by the editors.

Stephen Harrison: Some Wikipedia editors have expressed concern that the U.S. government is targeting editors—trying to identify and intimidate them in specific ways. That will discourage people from volunteering on Wikipedia.

Birgit Eger: The global umbrella organization for Wikipedia, the Wikimedia Foundation, has stood by its contributors and emphasized that the information on Wikipedia is as neutral and fair as possible. In addition to the content debate, there’s a formal question: Will Wikipedia be able to keep its nonprofit status?

Donations from individuals and companies are its main source of income. It’s unclear whether those donations will continue if they become subject to taxation. As a last resort, the foundation may consider relocating its headquarters—possibly to Germany, Norway, or the UK, according to journalist Stephen Harrison.

But the final decision is still pending. Edward Martin has not yet been confirmed in his post; he still needs to be approved by the Senate. And even among Republicans, there are still doubts. Birgit Eger, reporting from Washington.

The audio for the interview is available here from Deutschlandfunk.

Following President Trump’s announcement of global tariffs during his April 2 “Liberation Day” speech, stock markets around the world plunged sharply. The S&P 500 suffered its most significant one-day drop since the COVID-19 pandemic, while the Dow Jones Industrial Average fell 1,500 points on back-to-back days, marking the steepest decline in its 127-year history. Then on Wednesday, Trump partially backtracked, pausing certain tariffs for 90 days (with the notable exception of China), and the Dow posted its best one-day jump since 2020. But it's not a total rebound. At the time of reporting, both U.S. and global markets remain significantly down since “Liberation Day,” with investors grappling with the uncertainty of a U.S.-China trade war and its potential damage to the world economy.

Amidst the economic upheaval, Wikipedia’s volunteer editors have been diligently chronicling the turbulence on the open-access encyclopedia. Currently, the main Wikipedia entry is titled “2025 stock market crash,” however there is a prominent note at the top of the page pointing to an ongoing debate about whether the title should be changed to “2025 stock market decline.” The back-and-forth about calling it a crash versus a decline is taking place on the article’s talk page, a forum for editors to deliberate potential edits. So far, the discussion has mostly been a healthy disagreement about how to properly apply Wikipedia’s policies, like using reliable sources and writing from a neutral point of view. This discussion is not nearly as contentious as previous naming disputes such as “Taiwan” vs. “Republic of China,” or whether to call January 6 an “insurrection” or an “attack.” But the crash vs. decline debate reveals some cracks beneath the surface: editors have starkly different views of what Wikipedia is and how it best serves readers.

Continue reading my latest story for Slate—out today.

Upcoming Event

If you’re in New York, I would love to see you at this event on May 9.

MAY 9 | 6:30pm at McNally Jackson Seaport

The Depths of Wikipedia
feat. Annie Rauwerda, Michael Mandiberg, and Stephen Harrison

RSVP

Join Annie Rauwerda, Michael Mandiberg, and Stephen Harrison as they discuss the challenges and opportunities for Wikipedia, a.k.a. "The Last Best Place on the Internet." The conversation will explore Wikipedia's role as an open-source archive, address questions of bias and reliability, examine the relationship between editors and the Wikimedia Foundation, and discuss the politics of "notability."

Annie Rauwerda is a Wikipedia editor and enthusiast in New York City. She posts unusual Wikipedia articles on social media on accounts called "Depths of Wikipedia" and is writing a book about the behind-the-scenes drama and delight of creating the world's largest and most accessible information source. It will be published by Little, Brown in 2026.

Michael Mandiberg is an interdisciplinary artist who created Print Wikipedia, edited The Social Media Reader (NYU Press), and co-founded Art+Feminism. Their work has been exhibited at Los Angeles County Museum of Art, The Whitney Museum of American Art, and Musée d’Art Moderne de la Ville de Paris, amongst others. Mandiberg is Professor of Media Culture at the College of Staten Island, CUNY and Doctoral Faculty at The Graduate Center, CUNY.

Stephen Harrison has covered Wikipedia as a freelance journalist for the New York Times, The Guardian, WIRED, and Slate. Known as the leading reporter on the Wikipedia beat, he is also a tech lawyer and the author of The Editors, a novel inspired by Wikipedia and its volunteer contributors.

Subscribe now

This article originally appeared in Slate on June 14, 2019. I’m sharing it again on the Source Notes blog because the historic “North Face incident” came up in several conversations I had at South by Southwest (SXSW) in Austin last weekend.

The North Face is known for its popular jackets, fleeces, and vests. But warm apparel is not insulating the brand from blowback after it was revealed that the company had unethically manipulated Wikipedia for free advertising.

In a May video ad, the North Face and its advertising partner, Leo Burnett Tailor Made, explained the scam: Before traveling, people typically search the destination on Google, and Wikipedia ranks highly in the search results for both webpages and images. So the North Face uploaded photos to the Wikipedia pages of popular destinations that included the company’s clothing and equipment. When people searched destinations such as Guarita State Park in Brazil and Huayna Picchu in Peru, the top results in Google Images would serve as subtle product placement for the North Face’s brand.

In the video published on Ad Age, the North Face states: “We hacked the results to reach one of the most difficult places: The top of the world’s largest search engine. Paying absolutely nothing just by collaborating with Wikipedia.” But the word collaborating is highly misleading here. Neither the Wikipedia community nor the Wikimedia Foundation collaborated with the North Face.* The North Face’s “hack” violated the site’s terms of use, which prohibit paid contributions to the encyclopedia without disclosure. Besides that, it’s also brazenly shady and unethical to hijack a free educational resource like Wikipedia for secretive commercial use.

Once word got out, the Wikipedia community quickly removed the North Face’s images from the compromised Wikipedia pages—or at very least, cropped out the company’s logo. The Wikimedia Foundation also released a statement criticizing the North Face, saying, “They have risked your trust in our mission for a short-lived marketing stunt.” The foundation’s executive director, Katherine Maher, likewise blasted the North Face on Twitter: “Hey thenorthface, your brand was built on the passion of exploring places in the public trust. Maybe don’t litter in Wikipedia’s digital public park?”

The North Face eventually apologized on Twitter, “We believe deeply in @Wikipedia’s mission and apologize for engaging in activity inconsistent with those principles. Effective immediately, we have ended the campaign and moving forward, we’ll commit to ensuring that our teams and vendors are better trained on the site policies.”

That’s where the story ends, right? Not at all.

Revelations that the North Face had edited Wikipedia’s images for commercial reasons sparked a spirited debate among the volunteers who contribute to Wikipedia: Should the recent history of the North Face’s editing Wikipedia be included on the company’s Wikipedia page, and if so, how much coverage should it receive?

Pausing here to acknowledge that this is all highly meta: a discussion taking place on Wikipedia about whether Wikipedia should include information within that subject’s Wikipedia article about how that subject covertly and unethically edited Wikipedia. (Taking it a step further, there’s a relatively high possibility that the very article you are reading will at some point be cited on Wikipedia since this is usually the case with Slate and other outlets that have covered the site.)

On the talk page for the North Face’s Wikipedia article, multiple editors cautioned that Wikipedians should resist the urge to retaliate against the company using the encyclopedia itself. “[W]e cannot let our own (justifiable) emotions drive coverage or try to get our readers to understand our outrage,” wrote Barkeep49. “We should, instead, be endeavoring to write a neutral long-lasting article on the topic.”

Over the course of the lengthy text discussion, Wikipedia editors cited several pertinent policies and principles: Due and Undue Weight, part of the site’s Neutral Point of View policy, states that an article “should not give undue weight to minor aspects of its subject.” The North Face’s Wikipedia page stood at 16 short sentences. Adding a detailed new section about the hacking incident when the rest of the company’s 51-year-history was only a high-level summary would arguably throw the entire article out of balance.

Another principle mentioned in the debate was recentism, the notion that an article can become unduly skewed toward coverage of recent events. This matters because Wikipedia’s stated goal is to be an encyclopedia, not a newspaper, which generally means the project should focus on the information that will be historically significant for the long term. Would people still be talking about the North Face’s Wikipedia hack in five, 10, 20 years? On the other hand, as Wikipedia editor David Gerard pointed out, the North Face’s Wikipedia page was receiving a lot more traffic than normal due to the incident. Even if Wikipedia is not a newspaper, he argued that the editors should take note of the currently high level of public interest. Furthermore, the public relations fallout for the North Face had been covered in outlets like the Guardian, BBC News, and the New York Times. To some editors, this significant coverage in reputable publications suggested that Wikipedia’s notability criteria had been satisfied, and therefore information about the incident should be included in the article.

Ultimately, the consensus from the editors who participated in the discussion was to include a two-sentence paragraph describing the incident on the North Face’s Wikipedia page. As of Friday, it reads:

In May 2019, Leo Burnett Tailor Made, a marketing agency for The North Face Brazil, revealed that they had surreptitiously replaced photos of popular outdoor destinations on Wikipedia with photos featuring North Face products in an attempt to get these products to appear more prominently in search engine results. Following widespread media coverage and criticism from the Wikimedia Foundation, The North Face ended and apologized for the campaign, and the product placement was undone.

The rationale for including this summary was that some information was warranted—especially given the significant press coverage—but that an entire section or a separate article about the incident would be undue weight. Curious readers can also read more detail about the incident on the separate Wikipedia article for conflict-of-interest editing of Wikipedia, where the North Face is included along with other corporations and politicians that have attempted to edit the site for their own purposes. For example, in 2017 a Burger King employee allegedly edited the lead of the Whopper’s Wikipedia page before the company aired a commercial designed to trigger Google Home speakers to define the restaurant’s signature sandwich. (Like other smart assistants, Google Home sources information from Wikipedia.) The “OK, Google: What is the Whopper burger?” campaign also received bad press.

After the dust had settled, I contacted Kevin Li, a Wikipedia administrator and 18-year-old freshman at Stanford who had been involved in the discussions on the North Face’s page. Li was not entirely happy with the result—he would have preferred not to include the incident on the page because he did not think it would be genuinely relevant in the long term. But he was glad to see how the community had genuinely and thoughtfully engaged with issues like neutrality, undue weight, and notability. “These are tough questions,” Li told me. “We have to be very careful.”

The sense of social responsibility shown by Li and similarly-minded Wikipedia editors bears a striking contrast to the North Face’s casual disregard as it attempted to hijack a digital public resource. After the incident, Leo Burnett Tailor Made issued a statement saying: “Our team has accepted an invitation by Wikipedia to learn more about the platform and their work to share unbiased, fact-based knowledge.” Let’s hope these advertising professionals learn some important lessons from the Wikipedians, and that those lessons stick.

FYI: The eBook for The Editors, my suspense novel inspired by Wikipedia, is on sale today for $1.99 across all platforms.

Amazon

Apple

Barnes and Noble

Bookshop.org

Google

KOBO

Subscribe now

The last time that I attended South by Southwest (SXSW)—Austin’s sprawling festival for tech, film, music, and more—it was 2013, and I was in my mid-twenties. Back then, startups were throwing parties left and right, and you had free admission if you showed that you’d downloaded their new app on your phone. The highlight of the trip was randomly running into Snoop Dogg and convincing him to take a picture with us.

Flash forward to this Saturday, March 8—I’m heading back, but this time as a speaker and the author of The Editors. Our panel is on the surprising ways that fiction can help strengthen truth—one of my favorite paradoxes.

I’m a bit intimidated, to be honest, because the panel includes esteemed novelist Laila Lalami, Pulitzer Prize finalist and author of The Dream Hotel (a chilling new release where even dreams are under surveillance), alongside Rebecca MacKinnon, with Sewell Chan of the Columbia Journalism Review moderating. In other words: I’m the least qualified person in the room—but honored to be there.

Thinking Out Loud Before SXSW

It seems to me—at least on an intuitive level—that there’s a truth in fiction that isn’t captured by straight nonfiction reporting. William Faulkner once said, “The best fiction is far more true than any journalism.” I tend to agree.

On the panel, I might speak a bit from my practical experience as a journalist covering Wikipedia for several years. Many times, I was overwhelmed by the sheer amount of data available on the platform: the number of articles, the thousands of edits on very contentious pages, the hundreds of volunteer editors involved. When I was writing my journalistic articles, I found I was more successful reporting the facts if I forced myself to focus on a specific event. For example, This company tried—and failed—to manipulate Wikipedia last week, or This prolific Wikipedia editor has been banned: Here’s why.

Then again, I always worried that by limiting my scope to that week’s newsworthy event, I wasn’t appropriately communicating the fullness—or you might say, the truth—of Wikipedia.

When it came time to write a book on the subject, I gravitated toward writing it as a novel, and specifically, a suspense novel. It wasn’t an entirely conscious decision, but in retrospect I see two key benefits to writing it as fiction.

(1) Distillation - The real-life English Wikipedia has something like 1,500 power users, and a true, holistic nonfiction account would document each of their contributions. Not only would that lengthy account be impossible for me to write (as a human), but it would also be unreadable (as a human). So for the novel, I focused on four main perspectives: young editor, seasoned editor, journalist, and rulebreaker. In other words, I distilled.

Distillation is a tricky business because you don’t want the characters to be reductive or cliché. A lot of the work as a novelist comes in ensuring the characters are authentic with their own motives and not mere stereotypes or pawns of the plot. And distillation ≠ summary. It must preserve essence; not diminish it. The aim is lightning in a bottle that’s still lightning.

You may have heard the maxim: “In a world of endless information, curation is the key.” There’s a practical, can-do spirit to curating your feeds and collecting reliable sources. I like this principle, but I’ve started to think distillation is the better word. In a world drowning in data, distillation is the key—especially through fiction.

(2) Perspective - Fiction’s second power may be even greater than the first. Fiction can explore viewpoint and the inner lives of characters in a way that isn’t captured by pure data journalism, much less AI reports.

For instance, I can now ask an AI tool to summarize the back-and-forth debate on the talk page of a contentious Wikipedia article. It can give me a bulleted list of the key arguments raised on each side. However, the AI tool cannot tell me about the inner world of the person who is advocating for a certain change. It can’t communicate their thoughts and feelings in the moment. This is the domain of fiction, which provides a simulation of consciousness. As a reader, you can step into the skin of the character and experience their perspective.

Here’s a question: Which version of events is truer—the AI summary of a real Wikipedia debate, or a novel that simulates the human experience of Wikipedia editing? The answer isn’t obvious.

AI can summarize the facts pretty well, particularly by drawing on a dataset like Wikipedia. Fiction, however, can communicate something deeper by immersing us in the characters’ lived experiences. I think that’s what Faulkner was getting at.

***

As you can probably tell, I’m still kicking around these ideas in my head. What else do you think I should say?

And please let me know if you’ll be in Austin for SXSW on March 8. I would love to meet some of my readers in person.

Other things worth sharing

• The Balsillie School of International Affairs has launched Balsillie Case Studies, a “repository of evidence-based true stories of technology-related governance dilemmas across various sectors globally.” My case study for the series is titled “Wikipedia’s Governance Challenge: Policies and Guardrails for New Generative AI Technologies.” I’m excited to see how students grapple with this real-world scenario and what recommendations for Wikipedia come from the classroom.

• Last month, my book club read Fahrenheit 451. This March, we’re reading The Stranger by Albert Camus. Drop me a note to join.

Before we start, a few quick updates:

• New Article Alert: I have a new article out this morning in Slate: How Wikipedia Editors are Bracing for Targeted Attacks in 2025.

• Fact vs. Fiction: Ars Technica interviewed me about how fiction gave me room to explore the small but powerful Wikipedia community.

• Book club: Last month, my book club read The Sun Also Rises. This month, we’re reading Fahrenheit 451. Drop me a note to join.

I was so pleased to attend Wikipedia Day NYC last month, where I had the honor of being on a panel discussing “Reliable Sources” together with journalist Emily Gertz and NYU professor Clay Shirky.

After the conversation, I had the chance to discuss these ideas further with some of New York’s most dedicated Wikipedia editors. One of the highlights was meeting Annie Rauwerda, founder of Depths of Wikipedia, whose enthusiasm for the quirky history of the website shines through.

While this might seem like a niche event, you might be surprised to learn that over 1,200 people registered to attend. There is a strong interest in contributing to the ubiquitous free internet encyclopedia.

Although the video from the event is now available, I wanted to capture some of the main ideas in writing. Note: This is not a transcript of the interview itself.

What do we mean by reliable sources?

I think it’s important to get specific because most of the people in the world are not highly-experienced Wikipedia editors.

Here’s the key principle: The articles on Wikipedia should be based on reliable, independent, published sources.

Reliability of a Source: The reliability of a source depends on context. Where possible, Wikipedia editors prefer publications with a reputation for fact-checking and accuracy. If a publication has people engaged with checking facts or scrutinizing the writing, then that is considered an indicator of reliability.

Independence of a Source: Evaluating the independence of a source means considering whether the source is self-published, not a primary source, and not closely affiliated with the subject. For instance, social media posts and company press releases generally do not qualify as reliable sources because they are not considered sufficiently independent.

Can you give me an example?

Let’s say you have a New York content creator who makes a biographical TikTok video about her life. In the video, she mentions that her favorite restaurant when she was growing up was Junior’s Diner (the original location in Downtown Brooklyn).

This TikTok video is likely not to be considered a reliable source for a biographical page about that content creator for at least two reasons:

1. Not From a Reputable Source. The source is not from a publication with a reputation for fact-checking and accuracy. TikTok videos are self-published and do not undergo a rigorous editorial process.

2. Lack of Independence. The source is not independent from the subject. It’s self-published without any third-party vetting. The creator might have a conflict of interest; her content could be biased by her desire to promote herself or to tell her story in a certain way.

This doesn’t sound that difficult.

The rules may seem simple in the abstract, but they can become more challenging when applying them or interacting with others. Because we are inundated by social media, it can be a bit jarring for people to hear that most social media posts are not considered reliable in and of themselves. For example, the TikTok creator might be rather offended that her video is not considered a reliable source about a fact that she knows very well: her favorite restaurant growing up.

It’s important that we communicate the reasoning for this policy. Wikipedia focuses on verifiability, and her self-published video has not been verified by people who are independent from herself. But again, you could imagine how the content creator could get peeved that her own firsthand account is not considered sufficient without third-party vetting.

What about the reliability of a news publication that’s not a social media post?

There are also very extensive discussions on Wikipedia about whether a publication is or is not considered reliable based upon whether the publication has a track record of fact-checking and accuracy. Some publications are deprecated after the discussion indicates that they are generally unreliable. Going forward, any user who attempts to cite a deprecated source on a Wikipedia page receives a warning and a request to cite a more reliable publication.

For instance, I wrote about how Wikipedia editors deprecated the UK newspaper The Daily Mail back in 2021 after finding that the paper had mastered the art of running false clickbait stories. The American news and opinion website The Daily Caller was deprecated in 2019 after discussion about how it consistently produces false or fabricated information.

Hold up. This sounds very controversial.

It can be. Wikipedia’s goal is to provide trustworthy information as reflected by reliable sources. That means Wikipedia editors can and do delete content when the underlying source is determined not to be trustworthy.

Keep in mind, Wikipedia has never billed itself as a free speech forum, a free-for-all where anybody can cite whatever source they like. The goal is to provide a reliable encyclopedic summary, which means rejecting and deleting poor sources.

As you might imagine, these discussions about whether a publication is or is not considered reliable can be very contentious. That’s why it’s crucial for Wikipedia editors to repeatedly remind the public that the discussions about reliable sources are all quite public and transparent. Anybody can become a volunteer Wikipedian—it’s free—and engage in these conversations. There’s not some exclusive cabal of editors (based in New York) making secret decisions behind the scenes.

I’d like to shift gears and talk about the recent news that the Heritage Foundation is planning to “identify and target” Wikipedia editors.

Last month, the Jewish-American news site Forward reported a shocking scoop: the Heritage Foundation, the conservative think tank behind Project 2025, was planning to “identify and target” Wikipedia editors. Through analyzing text patterns, usernames, technical data, and employing social-engineering tactics, Heritage aimed to reveal the identities of anonymous Wikipedia editors they believed were “abusing their position” on the platform.

What effect might these targeted attacks have on the free encyclopedia?

If the think tank succeeds in its effort to identify and target editors, the consequences could be profound. Faced with the risk of harassment or real-world retaliation, many volunteer editors—especially those covering politically sensitive topics—may simply stop contributing. Those who remain are likely to be the most ideologically-driven voices, further eroding Wikipedia’s stated goal of neutrality.

Happy New Year! Even if you’re as curmudgeonly as Larry David, I still think we’ve got a few more days where saying “HNY!” is socially acceptable. Whether or not that statute of limitations has expired, I’m offering this: my post to set the tone for 2025.

This year, instead of bold resolutions, I’m grouping a few ideas that have been rattling around my brain. These aren’t goals so much as principles I want to carry with me, and I’ll share them here in case they spark something for you:

1. Handwriting

Toward the end of 2024, I noticed a flood of posts extolling the benefits of handwriting. One study especially stood out: “Handwriting but not typewriting leads to widespread brain connectivity.” The authors explained that handwriting activates more elaborate brain connectivity patterns than typing does—particularly in regions tied to memory formation and learning. Their conclusion was compelling:

“We urge that children, from an early age, must be exposed to handwriting activities in school to establish the neuronal connectivity patterns that provide the brain with optimal conditions for learning.”

Handwriting requires complex motor movements, while typing involves far simpler, repetitive actions. It makes sense that the former would have a stronger impact on our brains.

Yet, as a society, we’ve largely replaced handwriting with digital devices. Well-meaning school administrators likely believed that teaching typing skills would better prepare students for a tech-based workforce. What they didn’t anticipate was that handwriting wasn’t just a nostalgic pastime—it’s a far more effective tool for learning and encoding information.

I think back to my favorite math teachers who worked out problems step by step on the chalkboard. Watching them physically write out their process helped to make the concepts click.

Inspired by this research—and a desire to try things differently this time—I’m writing the first draft of my next novel by hand. Writing out the chapters on yellow legal pads is a much slower process than typing, but it feels easier to pick up the story where I left off each day. Perhaps the act of handwriting does indeed build more connections across the brain.

2. Metamodernism

Wikipedia defines metamodernism as:

“a cultural discourse and paradigm that has emerged after postmodernism. It refers to new forms of contemporary art and theory that respond to modernism and postmodernism and integrate aspects of both together. Metamodernism reflects an oscillation between, or synthesis of, different ‘cultural logics’ such as modern idealism and postmodern skepticism, modern sincerity and postmodern irony, and other seemingly opposed concepts.”

If you’re curious about metamodernism, you might find a well-made YouTube explainer, like this overview from Living Philosophy, to be the best starting point.

That said, let me break it down with a superhero analogy:

• Modern idealism is like the original Captain America (1941): a clear-cut symbol of nationalism and duty, fighting for freedom and democracy with moral clarity.

• Postmodernism, by contrast, deconstructs these ideals. Think Watchmen (1986), where the heroes are deeply flawed, truth is murky, and defining the “good guy” depends on your perspective. Or Deadpool, with its self-aware humor, irony, and cynical take on heroism.

Now, enter metamodernism: a “vibe shift” that moves beyond modern idealism and postmodern skepticism. Instead of choosing one mode, it oscillates between the two. Some say it pushes through postmodern irony to arrive at something more earnest. This “new sincerity” blends hope with self-awareness.

Take Everything Everywhere All at Once (2022), a film many critics consider metamodern. The narrative revels in absurdity but lands with a moment of heartfelt sincerity. Or look at the resurgence of the band Creed. Once dismissed as cringe (a classic postmodern insult), they’re now being reclaimed as post-cringe.

Even superheroes are evolving. On YouTube, Deep Talks pastor Paul Anleitner predicts the forthcoming Superman movie will introduce the first explicitly metamodern Superman. He won’t be defined by naïve idealism or ironic detachment. Instead, he’ll embody a hopeful yet self-aware sincerity—a hero for the modern age.

Look up in the sky! It’s a bird! It’s a plane! It’s metamodernism soaring into the zeitgeist!

3. Longform

I’ll admit that I often forget what I hear on podcasts—my attention tends to drift. But every now and then, something sticks. A while back, I heard Yuval Noah Harari, author of Sapiens, share advice that felt spontaneous yet profound. He was speaking about the value of longform content over shortform:

“Throughout the day, have an information diet […] Be very aware of what enters your mind.
So for instance, I read long books. I do many interviews. I prefer three hour interviews to five minute interviews. The long format, it’s not always feasible, but you can go much, much deeper. […] Give preference to big chunks. […] Books over Twitter, definitely.”

Longform—whether it’s a book, an in-depth article, or a sprawling podcast—offers depth, nuance, and the chance to engage with ideas on a more meaningful level. In contrast, shortform content, like TikTok videos, provides a quick dopamine hit but rarely leaves you any wiser. I’ve fallen into the scrolling trap more times than I’d like to admit, only to realize how little value I’ve gotten from that time.

As 2025 unfolds, I suspect the pull of shortform will only grow stronger. Better to stick with more “nutritious” options: books, thoughtful essays, and longform podcast discussions—literature or content (more on that word later) that’s more likely to be transformative.

4. Down-versus-up

’s Substack, The Honest Broker, has become one of my go-to reads. Last year, he wrote about what he sees as an emerging dimension of cultural conflict.

“Analysis of cultural conflict is still obsessed with left-versus-right strategizing, but the actual battle lines are increasingly down-versus-up.”

The targets in this conflict are the elites—those perceived as sitting at the top of the social hierarchy. Gioia lists wealthy CEOs, DC politicians, celebrity TV newscasters, law enforcement authorities, experts of all stripes, Ivy League academics, movie stars, and other figures of authority or expertise. “All of the cultural energy right now is on the bottom. And that energy has been intensifying,” he observed.

After publishing The Editors last year—a novel centered on a crowdsourced internet encyclopedia—I’ve been paying attention to how this “down-versus-up” dynamic is playing out in discussions about Wikipedia. A recent example came on Christmas Eve, when Elon Musk attacked Wikipedia, accusing it of being controlled by a woke “editing authority.”

From one angle, you could interpret Wikipedia as an elite institution—it’s the official, authoritative internet encyclopedia. But isn’t this a bizarre twist in the “down-versus-up” narrative? Musk, the world’s wealthiest man, is positioning himself as a populist punching up at Wikipedia. Meanwhile, Wikipedia itself is, in many ways, deeply egalitarian: anyone can edit it, regardless of their academic or professional credentials. (Granted, there’s no guarantee your edits will stick.)

What does this down-versus-up dynamic mean for Wikipedia—or for other battlefronts—in 2025? I don’t have the answers yet, but I suspect we’ll see more strange and paradoxical manifestations of this tension in the year ahead.

5. Definition of content

I’ve always been reluctant to call writing “content” because I worry that the term cheapens the craft. As Jon Christian put it in Slate back in 2016:

“I find the find the idea of someone aspiring to create content, in as many words, to be almost indescribably sad. It seems like an act of pre-emptive surrender, of giving up hope that you’ll ever create something with a higher calling than attracting clicks for some monolithic publisher.”

And yet, here we are—living in the age of content. Platforms thrive on user-generated content, politicians use content to dominate conversations, and authors are encouraged to release content to build their audiences.

But what exactly is content? And how does it differ from art?

On this point, I found an illustration from ’s particularly insightful. Stein highlighted the New York Times’s article, “100 Best Books of the 21st Century,” as a prime example of content. (When the list was published, I noted how it was surprising that so few of these books engaged with the internet, despite focusing on books published since 2001).

As Stein correctly observed, the Times’s list wasn’t journalism or literary criticism. It was content—a post designed to encourage reader participation and spark more user-generated content, like the torrent of passionate videos covering the books the Times had snubbed.

The books themselves? That’s literature. The list? That’s content.

This distinction has helped me rethink content not as a diminished form of writing but as a participatory game—an invitation for audiences to engage, react, and toss their own content back into the mix.

One More Thing

I’m launching a monthly virtual book club for 2025! Who says the literary man is going extinct?

For January, we’ll kick things off with Ernest Hemingway’s The Sun Also Rises. The novel takes us through the literary circles of Paris, the bullfights of Pamplona, and the decadence of the so-called Lost Generation after World War I. Although the book is nearing its 100th birthday, I have a feeling its themes will resonate. Lost Generation vibes are strong these days.

If you’d like to join, drop me a comment or an email! For now, we’ll keep the book discussion here on Substack, but I’m open to organizing a live session if there’s interest. Bonus points if you also pick up Everybody Behaves Badly, an account of the real-life events that inspired Hemingway’s breakout novel.

HNY (again),
Stephen

Note: This article originally appeared in Slate on May 25, 2022. On December 24, 2024, Elon Musk tweeted to his followers that they should “Stop editing to Wokepedia [sic] until they restore balance to their editing authority.” Since then, he has posted several disparaging comments about Wikipedia.

On April 25, after Elon Musk announced that he would be buying Twitter for $44 billion and pledged to run the social media platform as a “free speech absolutist,” Wikipedia’s volunteer editors knew that they had a lot of work ahead of them—and they were right. One contributor soon added the words “Owner of Twitter” to Musk’s Wikipedia page, language that another editor promptly deleted with the comment, “He does not yet own Twitter!” At press time, a precautionary banner on the Musk Wikipedia article warns readers about the likelihood of speedy revisions. “This article is about a person involved in a current acquisition. Information may change rapidly as the event progresses, and initial news reports may be unreliable,” the banner reads.

Since 2015, the English language version of Musk’s Wikipedia page has received more than 110 million direct page views. Snippets from Musk’s Wikipedia article also appear in the Google search results for “Elon Musk,” a term that has been searched 479 million times in the past 30 days, according to Google Trends, with a pronounced spike in searches following Musk’s announcement about purchasing Twitter on April 25.

Since the page was created in 2004 (here’s what it looked like then, when it was fewer than 100 words), some 3,240 Wikipedia editors have collaboratively written the Musk Wikipedia article, which covers the billionaire’s early life in South Africa and Canada; his leadership roles with PayPal, SpaceX, and Tesla; and his proposed acquisition of Twitter. Over the course of about 8,600 words, the Wikipedia page outlines Musk’s career highlights—like the 2020 launch making SpaceX the first private company to place a person in orbit and dock a crewed spacecraft with the International Space Station, and how the Starlink network of satellites has provided some internet access to Ukraine during the 2022 Russian invasion. The article also lists Musk’s controversies, including a 2018 lawsuit brought by the Securities and Exchange Commission alleging that Musk’s tweet about having “secured” funding to take Tesla private at $420 per share was false and damaging to investors. (Musk settled the SEC lawsuit in 2018.)

Reading the “personal life” section of a celebrity’s Wikipedia’s page has always been one of the internet’s greatest pleasures, and in Musk’s case, that portion is a total trip. It recounts Musk’s marriages, remarriage, and divorces, together with his relationships with Amber Heard and the Canadian musician Grimes. There’s also a somewhat convoluted paragraph explaining the naming process for Musk’s son—originally named “X Æ A-12,” the child is now “X AE A-XII” Musk because California regulations prohibit the use of any characters not in the modern English alphabet.

Overall, Musk’s Wikipedia page has been designated as a “good article” by the site’s volunteer reviewers, meaning that it’s noteworthy for being well-written, containing factually accurate information, providing broad coverage, and adopting a neutral point of view. According to Wikipedia, only about 0.5 percent of articles on the site are awarded with this recognition. Then again, not every Wikipedia editor thinks the Musk Wikipedia page is up to snuff. More than once, Wikipedia editors have had to remind one another that the page should provide only previously published information and should not be influenced by any opinions that the editor holds about the controversial celebrity. On the article’s talk page, a place for editors to chat with one another and raise proposed changes, many of the discussions revolve around whether the article is biased. As the user Warbayx put it, “literally 1/4 of Musk’s front page is dedicated to criticism, how can anyone think this is unbiased and fair?” To which the user PraiseVedic replied, “I would say that more than 1/4 of the coverage Musk receives in the media is critical of him, so if anything Wikipedia is under-criticizing him, if that’s a thing.” One suspects the balance of criticism might never feel right to so-called Musketeers, a group that Wikipedia sums up as “Fans of Elon Musk, usually in the pejorative sense.”

But there is in fact someone who cares more about Musk’s Wikipedia page than even his most diehard fanboy—Elon Musk himself. All signs indicate that the world’s richest man is obsessed with how he is described on the free internet encyclopedia.

Back in December 2019, Musk tweeted, “Just looked at my wiki for 1st time in years. It’s insane! Btw, can someone please delete ‘investor’. I do basically zero investing.” In a follow-up tweet, Musk suggested that his Wikipedia page should focus on his leadership of Tesla and SpaceX rather than his investing activity.

Following Musk’s tweets, editors sprang to action, first deleting the word investor from the page and then reinserting it. Molly White, a volunteer Wikipedia administrator (username GorillaWarfare), said that tweets like Musk’s put the Wikipedia community in a difficult position. “It’s challenging when someone with a large public following makes public calls for people to edit the Wikipedia article about them because typically the people who see those requests are not familiar with Wikipedia or its policies,” White said. “Although we love when new editors join us, trying to learn to edit while also jumping into a dispute where an article subject disagrees with the content of their article is not an easy task, and can result in challenges both for those new editors and for experienced editors trying to handle the dispute.”

After Musk tweeted this out, Wikipedians engaged in an extended debate about whether Musk’s request was “based more on his desired branding than the facts,” and whether it set a bad precedent for a public figure to call on their “legion of fans” to make requested changes. As a reminder, Wikipedia rules state that the encyclopedia should reflect the information published in reliable, published sources. In this case, hundreds of third-party sources ranging from the Wall Street Journal to the Hindustan Times described Musk as an “investor” given his early history with Tesla. (More on that in a second.) Musk himself had talked for years about investing in DeepMind to track the progress of artificial intelligence, which he is on record as calling the greatest threat to humanity. Against the backdrop of these secondary sources, it would not be proper for Wikipedia editors to remove the word investor simply because Musk tweeted—read: self-published—a contrarian point of view. (At press time, Musk’s Wikipedia page still describes him as an “investor” in the first sentence, which also calls him an “entrepreneur” and “business magnate.”)

Recently, Musk has upped the volume of his Wikipedia grumbling. In April, Vaibhav Sisinty, the CEO and founder of GrowthSchool, tweeted a reminder that “Elon Musk was not the founder of Tesla. He acquired it.” Musk replied to the tweet stating that when he funded Tesla in 2004, Tesla was a “shell corp with no employees, no IP, no designs, no prototype” and “Even the name ‘Tesla Motors’ was owned by others!” Afterward, Musk lashed out specifically at Wikipedia, tweeting, “They say history is written by the victors, but not on Wikipedia if the losing party is still alive & has lots of time on their hands!” Many people interpreted this as Musk attacking Martin Eberhard, a person Wikipedia (citing several sources) describes as one of the two original co-founders of Tesla in 2003 and who stayed on as the company’s CEO until 2007.

Although it’s notoriously hard to read Musk’s mind, his tweets indicate that he would prefer that Wikipedia name him as a “founder” of Tesla because of his relatively early contributions to the company and popular associations with the EV brand. But notice how, once again, Musk is asking Wikipedia editors to set aside years of third-party reporting that documents how Musk purchased shares to become Tesla’s majority shareholder and was not one of the individuals involved since Tesla’s scrappy beginnings.

Thankfully, Musk cannot decree what information about him should appear on Wikipedia, nor can he mandate that Wikipedia use his preferred nomenclature. For instance, Musk said in 2021 that he believed CEO was a “made-up title” and that he prefers to be called the “Technoking” of Tesla. A March 2021 regulatory filing showed that Tesla had officially given him his requested “Technoking” title. (According to a Tesla spokesperson at the time, Musk remained the company’s chief executive officer.) But so far, Wikipedia editors have reverted the many attempts to replace all references to “Tesla CEO” with “Tesla Technoking” on Musk’s Wikipedia page. A notice on the Musk talk page makes a simple case for using traditional language: Musk’s position at the company is still effectively the CEO, and “Technoking” would likely be confusing for readers. The current version of the article states, “In 2021, Musk nominally changed his title to ‘Technoking’ while retaining his position as CEO.”

On the talk page behind Musk’s Wiki page, there is also an editorial note discouraging users from changing the words “business magnate,” meaning captain of industry, to “business magnet,” a pun that Musk has joked about on Twitter.

One subject that has been passionately debated among Wikipedia editors is whether Musk’s page should quote people who describe him as a “sociopath.” Several former Musk employees and associates have claimed that Musk exhibits pathological sociopathy and is severely lacking in empathy. But so far, Wikipedians have been reluctant to include that verbiage on Musk’s article. “We have to be cautious with these terms so that we don’t inaccurately imply that a person has been diagnosed with a medical condition,” White told me, noting that a similar discussion often crops up when editors consider using the word narcissist.

Some Wikipedia editors have proposed a rewrite of the old Goldwater rule, arguing that someone’s Wiki page should not provide a mental health diagnosis unless it is supported by a clinical psychiatrist who has examined the subject personally. Although Wikipedia has not yet formalized a standard policy in this area, it seems that the site’s editors have generally decided to use caution. At press time, there is no language on Musk’s Wikipedia page describing him as a “sociopath” or having any sort of maladaptive personality disorder.

Since announcing his plans to purchase Twitter, Musk has promoted himself as a “free speech absolutist” who intends to restore the value of free speech to Twitter. Critics are understandably concerned that Musk’s interpretation of “free speech” will mean an uptick in misinformation and harassment, which will quickly turn Twitter into a cesspool of lies, racial attacks, and antisemitism.

At a time when all sides are batting around the term free speech, it seems worthwhile to mention that Wikipedia has never billed itself as a free speech platform. The information is free, yes; the project also firmly resists government censorship, including recent threats from Russia. But the internet encyclopedia has never been a free-for-all. Instead, the ability to post lasting content has always been restricted by the system of peer production and the encyclopedia’s policies; free expression is limited by rules and requirements, such as citing to a credible source. That matters because nonprofit Wikipedia occupies what appears to be an increasingly rare internet niche: a place where billionaires cannot purchase their preferred version of events, nor own the means of conversation.

🎧 The audiobook for my debut suspense novel, The Editors, is officially out today on Audible.

I’m happy to share that it’s narrated by two award-winning voice actors: Mia Hutchinson-Shaw (a graduate of the Royal Conservatoire of Scotland) and Tim Lounibos (known for narrating James Patterson’s books). Hearing them translate my story into this new medium felt so exciting and surreal. Tim masterfully builds suspense right from the prologue, and Mia is pitch-perfect as Morgan, the protagonist.

Tim told me, “I LOVED The Editors and hope I did it justice.” (I’d say he nailed it.) Mia added, “I absolutely thought about the internet and internet information differently after narrating this book.”

If you’re a fan of audiobooks, I’d love for you to check it out. You can grab The Editors with an Audible token here: https://amzn.to/4fShp8d

Source Notes

• The science journal Nature published a disturbing report suggesting that readers are unable to reliably differentiate AI-generated poetry from human-written poems. But it gets worse: People rated AI-generated poems more highly than human-authored poems on qualities like rhythm and beauty. So far, I think has the best explanation: AI-generated art is “almost unavoidably kitsch–comforting, straightforward, accessible, flattering. And people love kitsch!”
  

• Issues in Science and Technology, a policy journal published by the U.S. National Academies of Sciences, Engineering, and Medicine together with Arizona State University, just republished a short story of mine. “Little Assistance” is a tale about Eugene, the first judge on the moon, set in the near future. You can read both the original story and the reaction essay over at Issues.org.
  

• Wikipedia is under fire in India. The Delhi High Court has ordered the Wikimedia Foundation to remove an article about a court case and reveal information about the editors involved–a direct challenge to Wikipedia’s core principles (freedom of information, respecting user anonymity). newsletter has a good breakdown of what’s at stake, including the risk of a potential ban in India.
  

• A recent survey shows Gen Z thinks they need a significantly higher salary to feel financially successful. Are older generations simply remembering an outdated number, or is Gen Z being brainwashed by affluent social media influencers? My guess is both.

Let me know if you check out the audiobook, or if you’d like me to cover these stories in detail in a future post.

Cheers,
Stephen

I originally wrote this article for Slate in August 2023. As of November 2024, I stand by my predictions.

Wikipedia is, to date, the largest and most-read reference work in human history. But the editors who update and maintain Wikipedia are certainly not complacent about its place as the preeminent information resource, and are worried about how it might be displaced by generative A.I. At last week’s Wikimania, the site’s annual user conference, one of the sessions was “ChatGPT vs. WikiGPT,” and a panelist at the event mentioned that rather than visiting Wikipedia, people seem to being going to ChatGPT for their information needs. Veteran Wikipedians have couched ChatGPT as an existential threat, predicting that A.I. chatbots will supplant Wikipedia in the same way that Wikipedia infamously dethroned Encyclopedia Britannica back in 2005.

But it seems to me that rumors of the imminent “death of Wikipedia” at the hands of generative A.I. are greatly exaggerated. Sure, the implementation of A.I. technology will undoubtedly alter how Wikipedia is used and transform the user experience. At the same time, the features and bugs of large language models, or LLMs, like ChatGPT intersect with human interests in ways that support Wikipedia rather than threaten it.

For context, there have been elements of artificial intelligence and machine learning on Wikipedia since 2002. Automated bots on Wikipedia must be approved, as set forth in the bot policy, and generally must be supervised by a human. Content review is assisted by bots such as ClueBot NG, which identifies profanity and unencyclopedic punctuation like “!!!11.” Another use case is machine translation, which has helped provide content for the 334 different language versions of the encyclopedia, again generally with human supervision. “At the end of the day, Wikipedians are really, really practical—that’s the fundamental characteristic,” said Chris Albon, director of machine learning at the Wikimedia Foundation, the nonprofit organization that supports the project. “Wikipedians have been using A.I. and M.L. from 2002 because it just saved time in ways that were useful to them.”

In other words, bots are old news for Wikipedia—it’s the offsite LLMs that present new challenges. Earlier this year, I reported on how Wikipedians were grappling with the then-new ChatGPT and deciding whether chatbot-generated content should be used in the process of composing Wikipedia articles. At the time, the editors were understandably concerned with how LLMs hallucinate, responding to prompts with outright fabrications complete with fake citations. There is a real risk that users who copy ChatGPT text into Wikipedia would risk polluting the project with misinformation. But an outright ban on generative A.I. seemed both too harsh and too Luddite—a failure to recognize new ways of working. Some editors have reported that ChatGPT answers were useful as a starting point or a skeletal outline. While banning generative A.I. could keep low-quality ChatGPT content off of Wikipedia, it could also curtail the productivity of human editors.

These days, Wikipedians are in the process of drafting a policy for how LLMs can be used on the project. What’s being discussed is essentially a “take care and declare” framework: The human editor must disclose in an article’s public edit history that an LLM was used and must take personal responsibility for vetting the LLM content and ensuring its accuracy. It’s worth noting that the proposed policy for LLMs is very similar to how most Wikipedia bots require some human supervision. Leash your bots, your dogs, and now your LLMs.

To be clear, the Wikipedia community has jurisdiction over how their fellow editors use bots—but not how external agents are using Wikipedia. These days, generative A.I. companies are taking advantage of the internet encyclopedia’s open license. Every LLM so far has been trained on Wikipedia’s content, and the site is almost always the largest source of training data within their data sets.

Despite swallowing Wikipedia’s entire corpus, ChatGPT is not the polite sort of robot that graciously credits Wikipedia when it uses that information for one of its responses. Quite the contrary—the chatbot doesn’t typically disclose its sources at all. Critics are advocating for greater transparency, and advocating restraint until chatbots become an explainable A.I. system.

Of course, there’s a scary reason that LLMs don’t normally credit their sources: the A.I. does not always know how it has arrived at its answer. Pardon the grotesque simile, but the knowledge base of a typical LLM is like a huge hairball; the LLM may pull strands from Wikipedia, Tumblr, Reddit, and a variety of other sources without distinguishing among them. And the LLM is basically programmed solely to predict the next phrase, not to provide credit when it’s due.

Journalists in particular seem very concerned about how ChatGPT isn’t acknowledging Wikipedia in its responses. The New York Times Magazine published a feature last month on how the reuse of Wikipedia information by A.I. imperiled Wikipedia’s health and made people forget about its important role behind the scenes.

But I get the sense that most Wikipedia contributors are less concerned about credit-claiming than the average reporter. For one thing, Wikipedians are used to this: After all, before LLMs, Siri and Alexa were the ones scraping Wikipedia without credit. (As of publication time, these smart assistants have been updated to say something like “from Wikipedia.”) More fundamentally, there has always been an altruistic element in curating information for Wikipedia: People add knowledge to the site expecting that everyone else will use it how they will.

Rather than sapping away the morale of volunteer human Wikipedians, generative A.I. may add a new reason to the list of their motivations: a sincere desire to train the robots. This is also a reason that generative A.I. companies like OpenAI should care about maintaining Wikipedia’s role as ChatGPT’s primary tutor. It’s important for Wikipedia to remain a human-written knowledge source. We now know that LLM-generated content is like poison for training LLMs: If the training data is not human-created, then LLMs become measurably dumber. LLMs that eat too much of their own cooking are prone to model collapse, a symptom of the curse of recursion.

As Selena Deckelmann, the Wikimedia Foundation’s chief product and technology officer, put it, “the world’s generative AI companies need to figure out how to keep sources of original human content, the most critical element of our information system, sustainable and growing over time.” This mutual interest is perhaps why Google.org, the Musk Foundation, Facebook, and Amazon are among the benefactors who have donated more than a million dollars to the Wikimedia Endowment—A.I. companies seem to have realized that keeping Wikipedia a human-created project is in their interests. (For further context, the foundation is primarily supported by numerous small donations by ordinary Wikipedia readers and supporters, which is comforting for those of us who worry about any big tech company gaining too much influence over the direction of the nonprofit organization.)

The weaknesses of A.I. chatbots could also popularize new use cases for Wikipedia. In July, the Wikimedia Foundation released a new Wikipedia ChatGPT plug-in that allows ChatGPT to search for and summarize the most up-to-date information on Wikipedia to answer general knowledge queries. For instance, if you ask ChatGPT 3.5 in its standard form about Donald Trump’s indictment, the chatbot says it doesn’t know about it because it is only trained on the internet through September 2021. But with the new plug-in, the chatbot accurately summarizes current events. Notice how Wikipedia in this example is functioning something like a water filter: sitting on the tap of the raw LLM, rooting out inaccuracies, and bringing the content up to speed.

Whether Wikipedia is incorporated into A.I. via the training data or as a plug-in, it’s clear that it’s important to keep humans interested in curating information for the site. Albon told me about several proposals to leverage LLMs to help make the editing process more enjoyable. One idea proposed by the community is to allow LLMs to summarize the lengthy discussions on talk pages, the non-article spaces where editors delve into the site’s policies. Since Wikipedia is more than 20 years old, some of these walls of texts are now lengthier than War and Peace. Few people have the time to review all of the discussion that has taken place since 2005 about what qualifies as a reliable source for Wikipedia, much less perennial sources. Rather than expecting new contributors to review multiyear discussions about the issue, the LLM could just summarize them at the top. “The reason that’s important is to draw in new editors, to make it so it’s not so daunting,” Albon said.

John Samuel, an assistant professor of computer science at CPE Lyon, told me that prospective Wikipedia editors he’s recruited often find it difficult to get started. Finding reliable sources to use for an article can be very labor-intensive, and Gen Z has grown impatient with the chore of sifting through Google search results. An internet that has become flooded with machine-generated content will make the process of finding quality sources even more painful.

But Samuel foresees a hopeful future in which Wikipedia has integrated some A.I. technology that helps human editors find quality sources and double checks to ensure that the underlying sources in fact state what the human claims. “We cannot delay things. We have to think about integrating the newer A.I.-based tools so that we save the time of contributors,” Samuel said.

If there’s a common theme running through the A.I.-gloom discourse, it’s that A.I. is going to take people’s jobs. And what about the “job” of volunteer Wikipedia editors? The answer is nuanced. On the one hand, a lot of repetitive work (adding article categories, basic formatting, easy summaries) is likely to be automated. Then again, the work of the people editing Wikipedia has never really been about writing text, per se. The more important job has always involved discussions between members of the community, debates about whether one source or the other is more reliable, arguments about whether wording is representative or misleading, trying to collaborate with the shared goal of improving the encyclopedia. So perhaps that’s where the future is heading for Wikipedia: leave the polite busywork for the A.I., but keep the discourse and the disagreement—that messy, meaningful, consensus-building stuff—for humans.

After hearing Dopamine Nation referenced in approximately a billion podcast episodes, I finally decided I was going to download the audiobook and give it a listen. The popular nonfiction book is definitely worth the Audible credit.

Psychiatrist and author Dr. Anna Lembke explores one of the central challenges of our times: Never before in human history have we been flooded with so much high-dopamine stimuli. YouTube videos, texting, tweeting, TikTok . . . It’s never been so easy to access all these microbursts of pleasure and distraction. As Lembke puts it, “The smartphone is the modern-day hypodermic needle, delivering digital dopamine 24/7 for a wired generation.”

While the entire book is full of insights, there are two key ideas from Dopamine Nation that have stuck with me a few weeks after listening.

Self-binding

When I wake up I try not to even look at my phone until I have finished my reading and creative writing ritual. I know that if I so much as look at my phone, I’m liable to start watching Instagram reels with a cute Golden Retriever or the latest outrageous news, and inevitably lose a lot of time to scrolling. For me, it’s better to have a simple rule: No looking at the phone until I’ve done some writing.

Until I read Dopamine Nation, I hadn’t thought of this policy as a self-binding strategy. Lembke defines self-binding as “the way we intentionally and willingly create barriers between ourselves and our drug of choice in order to mitigate compulsive overconsumption.”

The book argues that self-binding has become a modern necessity. We can’t expect outside forces to limit our own intake for consumption. Even if Congress eventually passes a law against social media use by minors (or restricts other high-dopamine goods), adults will still have to devise our own barriers and policies. The rules must come from within.

One comforting insight is that the concept of self-binding has been around for centuries. Lembke observes that we even have whole myths about it. Remember how Odysseus was able to pass unharmed around the deadly Sirens? He ordered his crew to put beeswax in their ears and tie him to the mast of the sailing ship. Instead of trying to resist the Sirens’ song, Odysseus saved the day by binding himself, removing the temptation.

“What fires together wires together”

I do not have a neuroscience background, so I was unfamiliar with the old mantra from psychologist Donald Hebb that Lembke laces throughout her book: “What fires together wires together.” The idea is that if the brain has a dedicated neural process, then practicing that process strengthens those neural circuits.

We see the dark side of this every day. It’s so easy to (unconsciously) train your brain to crave something addictive and antisocial, like passively scrolling social media or watching TV. “Any behavior that leads to an increase in dopamine has the potential to be exploited,” Lembke writes.

But it’s important to remember that we can also strengthen positive neural pathways. Lembke points to research showing that practicing radical honesty can reinforce the habit and make it easier for our brains to handle the next time.

We know that scrolling is addictive. Turns out truth-telling is, too.

Audible is releasing the audiobook for my novel The Editors on November 26. If you don’t have an Audible subscription, you can grab a free 30-day trial.

Last Saturday, I appeared on Alexis Conran’s show for Times Radio London where we had a fun and far-reaching conversation about the famous internet encyclopedia. Here’s the transcript, lightly edited and annotated for clarity.

Alexis: Where do you get your information from? (Other than Times Radio, of course.) Is it Google? Is it Wikipedia?

Now, I can hear some of you scoffing, but since its founding, Wikipedia has become a surprisingly important part of the online information ecosystem. A bulwark against viral misinformation. So much so that in 2018, YouTube announced it would use Wikipedia to help fact check conspiracy videos.

But the future of the site is now under threat with the rise of artificial intelligence. So should we be doing more to protect it? Well, Stephen Harrison is a tech lawyer, a journalist, and the author of a new thriller based on Wikipedia called The Editors.

And he's been thinking hard about the site's reputation in future. I caught up with Stephen Harrison a little earlier on.

Stephen: The Editors is a suspense novel that's inspired by Wikipedia. When I first got started covering Wikipedia as a journalist, I always thought it was ripe for a suspense novel because there's so many people around the world that rely on the information. There are also foreign governments and corporate agents who might try to manipulate the information that's on the site.

Why would they interfere? Because Wikipedia influences the information that appears on Google. It changes the content that is generated by AI.

So on the one hand, we have these manipulators. And then we have the ordinary Wikipedia editors. Frankly, I think a lot of the good editors are heroic. They are vigilantly on guard against misinformation on the site.

Alexis: The book touches on this. What about the trustworthiness of Wikipedia? I mean, we've all spotted errors in there. A lot of people were worried because of its structure and the idea that people can sort of log in and correct things or add things. How do you feel about where Wikipedia is now? As a journalist, would you use it as a source?

Stephen: I wouldn't use it as a source as a journalist, but I think a lot of journalists would be lying if they didn't admit that they might look at it for initial research. You can get the summary version on Wikipedia and then go to the underlying sources to go deeper.

As far as the general trustworthiness of Wikipedia, and I admit that this sounds like a lawyer answer, but it depends. Specific Wikipedia articles can be really good. I think that articles that got a lot of visibility by editors can often present a very good summary. But of course, the pages are changing from time to time, and they can always be moving in a direction that is more factual or less. For the listeners out there, you can get involved if you want to, and start participating in these editorial debates behind the scenes.

Alexis: We’re living in an age and in a current political climate where the issue of trust in what you see online is paramount. And in your book you touch on this, where one of the editors in your Infopendium is editing on behalf of wealthy and powerful clients. I know this was a challenge for real Wikipedia, where we had politicians in this country [the U.K.] getting in and editing their own pages. How much of that has Wikipedia dealt with so far? Is this still an issue?

Stephen: I do think it’s an issue. I think what’s interesting in this era of automation and AI, the best tools for identifying these for-profit and highly biased editors are still human tools. People have an intuition if an editor is only editing about such-and-such politician or such-and-such corporate leader, right? [See this article on how AI is currently not as good as humans at detecting AI.]

So people can still kind of detect bias, and these volunteers become very vigilant when they have a suspicion that someone is editing on behalf a paying client. It gets very tricky though when a bad actor is using several sockpuppet accounts. In those cases, they’re trying to make it look like they’re multiple different people. That’s really tricky for Wikipedia editors. And I think it’s something that the volunteers have to lookout for constantly.

Alexis: Pretty much everybody listening will have at some point used Wikipedia, but probably not know how the information got onto that page. At the moment, what is that process? Are there any checks on what goes on there? How does that work?

Stephen: It’s a site that anybody can edit, but it’s not an anarchy. There are rules and there are policies. And so if someone, a new contributor, goes to the site and they have found a reliable source that states a certain statement, that statement can then be added to an article. I always tell people that every sentence in a Wikipedia page should have a citation because it’s not meant to be original writing. A lot of times new contributors are shocked by how quickly they get slapped down for failing to cite a reliable source. So in practice, the site is sort of policing itself. It’s self-policing.

Alexis: Are there editors that have more power than other editors?

Stephen: What’s interesting to me is that there are only about 1500 core editors that are very involved on a regular basis with Wikipedia. It reflects other things in society: Everyone can join, yes, but very few people choose to get really involved.

There are administrators that are selected by their [peer editors] through a process that is not called an election, but is in practice very similar to an election. Administrators have more powers. They can block a user, or add protection to a page so that it can only be edited by Wikipedia editors with a certain number of edits. So yes, there is a bit of a hierarchy among the editors, but in general most [ordinary] editors can do most things on the website. You can make changes just by signing up, or if you don’t want to sign up, the site will default to using your IP address.

Alexis: You’ve also been talking about AI using Wikipedia as a learning tool. But of course that means that all that work that those editors have done goes uncredited. How much of an issue do you think that is when it comes to AI? Because AI is being trained by all of us. Every time you're asked to click in one of those boxes, “I am not a robot, spot all the traffic lights in this image,” what you're doing is you're just training AI.

Stephen: What I am worried about with AI is that it kind of pushes Wikipedia further back so that it has less visibility to the average user. When someone goes to ChatGPT to get their answer, they’re not always realizing that it’s coming from Wikipedia. And in order for our AI systems to be good, Wikipedia needs to be constantly updated.

Wikipedia needs to maintain a fresh contributor base and people need to be excited to continue working on the site. The public perception is that AI is doing this remarkable work on its own, but no, it's actually using the free labor of these Wikipedia editors. And I think the site needs that sort of crediting and visibility.

Alexis: That’s a big problem, isn't it, Stephen? I've spoken to authors who know for a fact that their books have been fed into AI without any credit. Is there any way of going back on that and giving people that credit or has the bottle, genie bottle been open and it's just way too late?

Stephen: Hm. I worry that we're not going to be able to change the past. What I think ChatGPT and LLMs have done incorrectly is not providing any provenance or citation. The AI usually fails to provide a source.

I am hopeful that even if it doesn’t become a law—even if we can’t get agreement across the different jurisdictions of the world of what the law should be—I would still hope that it becomes a consumer demand that the information coming from AI comes with a source. We users need to push for this.

With authors in particular, it’s not only informational but also creative content. I wouldn’t want my book to be mined for data by ChatGPT, and certainly not without getting a license fee or royalty for it.

Alexis: That doesn’t exist, does it? There’s currently no law prohibiting AI companies from taking your book, The Editors, and feeding it into ChatGPT?

Stephen: We have some important legal cases that are coming down the pipe. In the U.S., the New York Times is suing OpenAI saying that their journalism has been used without consent. The comedian Sarah Silverman has sued. [Update: A judge has trimmed Silverman’s claims down to direct copyright infringement.]

The core issue is something like this: When AI is copying, is it copying the way a human does, taking the core idea or spirit of the work, or is it just copying the structure? I sort of wonder if we can divide these two things.

I’m hopeful that the courts can use the existing copyright statutes in a way that will protect authors and say that the providers of AI need to compensate authors when they’re using it to train their applications.

Alexis: Isn’t it a problem though, when it comes to trying to put some rules around AI and the development of AI, that you can regulate if you like in the U S or in the UK or in the EU. But because AI is being developed by countries and private companies worldwide, the argument will be, well, other countries that don't have to obey those rules, their AI is going to develop quicker than ours and perhaps become more advanced. There is a sort of that sort of arms race at the moment. Is that the sort of the elephant in the room, if you like?

Stephen: I think so. Maybe a good model or comparison might be privacy laws. After Europe passed GDPR [the General Data Privacy Regulation], there was some thinking here in the US that the GDPR framework might be too burdensome, that it might inhibit innovation by tech companies. So here in the US, while we don’t even have a comprehensive data privacy statute, California has the CCPA [California Consumer Privacy Act]. It’s thought to be like Europe, but a little less strict. A little more flexible.

So I could see countries doing that. They could say, we’ll see what this other country enacts, and then we’ll implement something that’s a little less burdensome for our tech companies, because we don’t want to inhibit the AI products in our own country. That’ll be interesting to watch.

Thanks for reading. As a reminder, I’d be happy to join your book club by Zoom to discuss The Editors. And you can read & listen to other media coverage of the book here.