Elon Musk has not just complained about Wikipedia’s perceived bias, he has built a full-scale alternative in his own image. Grokipedia, an AI-written encyclopedia tied to his xAI project, promises speed and “neutrality” while quietly encoding a very specific worldview. To understand why he created it and what is tucked inside, I need to trace his long-running feud with Wikipedia, the design choices behind Grokipedia, and the political battle over who gets to define truth online.
From Wikipedia superfan to sworn rival
For years, Musk treated Wikipedia as one of the internet’s crown jewels, publicly praising the volunteer-run project on its anniversaries and leaning on it as a symbol of open knowledge. That affection curdled as his politics hardened and his conflicts with mainstream media intensified, and by 2022 he was attacking Wikipedia as ideologically captured and hostile to figures like himself. The official history of Grokipedia traces this arc directly, noting how his criticism of Wikipedia’s community-driven model set the stage for an AI-generated alternative aligned with xAI founder Elon Musk’s views.
That shift did not happen in a vacuum. Musk’s complaints landed in the same ecosystem as right-wing commentators who accuse Wikipedia of smearing conservatives and sanitizing liberal figures. In one widely cited exchange, reporter ALLYN chronicled how Tim Pool, Alex Jones and Tucker Carlson amplified stories of supposed ideological policing on the site, feeding a narrative that Wikipedia labels you a bigot if you deviate from progressive orthodoxy. Musk’s own posts about “woke” editors and slanted biographies slotted neatly into that storyline, turning a personal grudge into a broader campaign against the platform’s legitimacy.
How Grokipedia was built to look familiar but work differently
When Musk finally moved from rhetoric to product, he did not try to reinvent the encyclopedia format. Instead, Grokipedia copies Wikipedia’s familiar layout, article structure and navigational cues so closely that many entries read like near twins at first glance. Reporting on Grokipedia notes that the site mirrors Wikipedia’s format but swaps in AI-generated text, and that it quietly went live for users on October 27 after Musk had teased an early version earlier in the month. The goal is clear: make the new site feel instantly legible to anyone who has ever clicked a Wikipedia link, while changing who and what actually writes the words.
Under the hood, the differences are stark. Wikipedia is edited by humans who argue on talk pages, cite sources and can be banned for breaking community rules, while Grokipedia is written by Grok, the large language model developed by xAI. The project’s own description emphasizes that Grokipedia’s rapid growth comes from Grok’s reasoning capabilities, which allow it to spin up articles far faster than an encyclopedia that is edited by humans. That speed is the selling point, but it also means there is no equivalent of a talk page, no transparent revision history and no clear way for outsiders to correct or contest what the AI has written.
The launch: a two-week tease and a flood of AI pages
Musk framed Grokipedia’s debut as a kind of insurgent product launch, promising his followers that a first version would arrive within days and that it would quickly rival Wikipedia in scope. On October 6, he announced that the early version of Grokipedia was scheduled for release in two weeks, and the service then rolled out to users in the US on October 28. By the time the dust settled, the AI encyclopedia already boasted 855,279 entries, a figure that its own documentation presents as proof that Grok can match decades of human labor in a matter of weeks.
That scale has only accelerated. Coverage of Grokipedia now puts the article count at 5.6M, almost 79 percent of the size of English Wikipedia, a staggering number for a project that did not exist a short time ago. You do not publish on Grokipedia the way you do on Wikipedia, as one explainer aimed at new users put it, because You simply prompt Grok and let the system generate the entry. That inversion of control, from a messy community to a single AI model, is the core structural difference that shapes everything hidden inside the site.
What Musk says he is fixing: bias, “wokeness” and control
Musk insists that Grokipedia exists to correct what he sees as a deep ideological skew in Wikipedia’s coverage, especially on culture-war topics and biographies of controversial figures. In his telling, Wikipedia is “too left-wing,” a phrase echoed in coverage of the launch that notes he is building Grokipedia precisely Because Wikipedia allegedly leans left. He has cast his project as a counterweight that will restore balance, promising that Grok will present facts without the “woke mind virus” he blames for distorting everything from climate science to coverage of his own companies.
Critics see something very different: a political project dressed up as a neutral technology product. One detailed analysis of Late attacks on Wikipedia argues that Grokipedia is part of a broader right-wing effort, involving figures like Elon Musk and Senator Ted Cruz, to discredit independent arbiters of fact and replace them with platforms they directly influence. In that reading, Musk is not just offering users more choice, he is building an infrastructure where his own ideological preferences, from skepticism of mainstream journalism to hostility toward certain social movements, are baked into the reference layer of the internet.
What is actually inside: far-right talking points and an AI-shaped reality
Once you look past the familiar layout and the rhetoric about neutrality, the content of Grokipedia reveals how tightly it tracks Musk’s online persona and the grievances of his political allies. Investigations into Grokipedia Pushes Far show that the site repeats far-right talking points on topics like pornography and gender, including false claims that mainstream institutions are promoting explicit material to children. The same reporting notes that the new AI-powered Wikipedia competitor presents lengthy entries generated by AI that downplay or reframe issues like white nationalism, echoing the grievances aired by Tim Pool, Alex Jones and Tucker Carlson about being unfairly labeled extremists.
The pattern extends beyond a few hot-button topics. A separate look at The new AI-powered encyclopedia describes how Grokipedia’s AI-generated biographies and history pages often mirror Wikipedia’s structure while subtly shifting emphasis, omitting critical context or foregrounding conspiracy-adjacent claims. Because there are no human editors and no way for outsiders to correct entries, those slants are locked in unless xAI itself decides to retrain Grok or adjust its prompts. That is what Musk is really hiding inside his Wikipedia rival: not secret files or undisclosed data, but a closed editorial pipeline where one company’s model, tuned by one billionaire’s sensibilities, quietly rewrites the reference layer of the web.
Seen in that light, Grokipedia is less a neutral encyclopedia and more a test case in an age-old struggle over who gets to define truth, now supercharged by generative AI. Commentators who have tracked the clash between Elon Musk and Wikipedia argue that this is simply the latest round in a battle that has pitted centralized authorities, partisan media and open communities against one another for decades. What is new is the scale and opacity: a system that can generate millions of articles in weeks, with no public record of how each sentence was chosen, gives unprecedented power to whoever controls the model. Whether users embrace that tradeoff, or return to the messy transparency of Wikipedia, will determine how much of the internet’s future knowledge base ends up filtered through Grok.
More from Morning Overview