The Fact That Wasn't: How a Made-Up Claim Becomes 'Well-Documented'

In 2016, an anonymous editor added a single sentence to a U.S. cabinet official's Wikipedia page, claiming he had served in the Gulf War. He hadn't. The claim sat there, unsourced, until major outlets including the Wall Street Journal and the New Yorker repeated it in their own reporting — at which point it stopped looking like vandalism and started looking like an established biographical fact, cited back to the very news coverage it had generated.DOCUMENTED It took until April 2018, and direct confirmation from the CIA, for the claim to be formally corrected.DOCUMENTED

Wikipedia editors have a name for this: citogenesis, a term coined by cartoonist Randall Munroe in a 2011 xkcd comic strip, describing the exact mechanism by which an invented fact acquires the appearance of being well-documented.DOCUMENTED The pattern shows up far more often than most readers assume — and understanding how it works is one of the more useful tools available for telling a genuinely documented claim from one that merely looks like it.

How the loop actually runs

The mechanism is mundane, which is part of why it works. Someone adds an unsupported claim to an easily editable page. A writer under deadline pressure checks that page for background, finds the claim, and repeats it in a published article — often without independently verifying it, since it reads as an established fact rather than a live edit. A second editor then notices the claim now appears in a "reliable source" and adds that published article back onto the original page as its citation. From that point forward, the claim has a footnote. Anyone checking the page sees a sourced fact, not an unverified assertion, and the loop closes.REVIEWED

Wikipedia's own documented list of these incidents includes cases far more consequential than trivia. In one instance, an editor fabricated a claim that comedian Sacha Baron Cohen had once worked at Goldman Sachs; news outlets picked it up, and the resulting coverage was then cited back on the page as proof.DOCUMENTED In another, a joke edit claiming a royal jester had slapped a king's buttocks was picked up by multiple books, two of which were then added as the page's own supporting citations.DOCUMENTED Even the release date of a $20 digital watch spent years listed incorrectly after a single Wikipedia edit was repeated by the BBC, with neither party checking the other's work before publishing.DOCUMENTED

The professional version of the same loop

Citogenesis describes what happens on an open editing platform, but a structurally identical pattern operates across professional biography-building generally, without requiring any single dishonest actor. A biographical detail appears in one interview. A database or "notable people" aggregator imports it. A press release references the aggregator entry as independent confirmation. A later interview cites the press release. None of these individual steps requires bad faith — each participant may be relying in good faith on what looks like an already-established, previously reported fact — but the cumulative effect is the same as citogenesis: a claim that began with zero independent verification ends up surrounded by a dense enough web of citations that checking its actual origin becomes difficult.REVIEWED

The distinguishing feature of this kind of claim is not that it has been disproven. It's that no single link in the citation chain traces back to an institution with the authority to actually confirm it — a university registrar, a former employer's HR department, a court record, a regulatory filing. Volume of repetition and depth of verification are different things, and the first is far easier to manufacture than the second.REVIEWED

When the source is an ad wearing a byline

A related but distinct problem involves content designed to look like independent journalism while actually being paid placement. The Federal Trade Commission addressed this directly in a 2015 enforcement policy statement on native advertising, stating plainly that an ad's format is deceptive "if it materially misleads consumers about the ad's commercial nature," including through any implied suggestion that it comes from an independent source rather than the party paying for it.DOCUMENTED The agency's companion guidance, "Native Advertising: A Guide for Business," directs that such content should never be indistinguishable from the news, feature articles, and editorial content surrounding it, and that disclosures should use plain terms like "Ad" or "Sponsored" rather than technical language most readers won't immediately recognize.DOCUMENTED

The FTC's core reasoning is that knowing whether something is an advertisement changes how a reader evaluates it — the same claim carries different weight depending on whether it comes from a paying client's own marketing or from a reporter with no financial stake in the outcome.DOCUMENTED When that distinction is blurred, deliberately or through inconsistent labeling across different outlets, a reader has no reliable way to calibrate how much independent scrutiny a given claim actually received before publication.

Knowing that something is an ad "likely will affect whether consumers choose to interact with it and the weight or credibility consumers give the information it conveys," the FTC's guidance states — the entire basis for requiring clear disclosure in the first place.

Wikipedia's rule, and how often it gets skipped

Wikipedia occupies a particular position in this ecosystem because search engines and casual readers alike tend to treat it as inherently neutral, even though anyone can edit it. The Wikimedia Foundation's Terms of Use legally require that anyone editing for compensation disclose their employer, their client, and any other relevant affiliation, and community guidelines strongly discourage paid editors from directly modifying affected articles at all, steering them instead toward proposing changes on talk pages for independent review.DOCUMENTED Failure to disclose paid editing is itself a violation of Wikipedia's terms of use, separate from any question about the accuracy of the edits themselves.DOCUMENTED

In practice, undisclosed paid editing remains difficult to detect and is generally only caught after the fact, through pattern analysis by volunteer editors or after a public controversy prompts a closer look at an article's editing history.REVIEWED Articles that rely heavily on promotional sourcing can also fail Wikipedia's separate notability standard, leading to formal deletion discussions — outcomes that reflect how volunteer editors assessed the available sourcing at a given moment, not a definitive ruling on any underlying facts.REVIEWED

What actually breaks the loop

The fix on the Wikipedia side is structural: citogenesis persists because journalists occasionally use an encyclopedia anyone can edit as an unverified primary source, rather than following its citations back to whatever the encyclopedia itself is citing.REVIEWED The fix on the native-advertising side is disclosure: content that originated as paid placement carries less evidentiary weight than independent reporting, and readers can only make that adjustment if the paid nature of the content is actually visible to them.REVIEWED

For readers trying to evaluate any claim that seems to be everywhere at once, the practical test is the same one investigative reporting relies on generally: trace the claim back to its earliest appearance, and check whether that earliest source is an institution with actual authority to confirm it, or simply the first link in a chain that everyone since has been citing to each other.REVIEWED Repetition is not evidence. It just looks enough like evidence, often enough, that separating the two has become its own kind of skill.

Have documents relevant to this story? Reach us through our tips channel.