The story was already a century old before anyone called it a crisis. A man named Jestin Coler was running a network of fake news websites from his home and pulling in $10,000 a month from advertising. Not from a government. Not from a foreign intelligence operation. From clicks. From outrage. From people sharing things they hadn't read past the headline. The websites looked real. The URLs were plausible. The stories were engineered to travel. And they did.
In 2016, a BuzzFeed News analysis dropped a finding that should have been impossible: the top fake news stories about the U.S. presidential election had received more engagement on Facebook than the top stories from every major media outlet combined. The New York Times. The Washington Post. CBS. All of them, outperformed by fabrications. PolitiFact named "fake news" its Lie of the Year. Oxford Dictionaries chose "post-truth" as its word of the year. Two different organizations, independently, arriving at the same verdict about the same twelve months.
Something had broken. The question was when — and whether anyone had seen it coming.
One person had. In 1995, the author Terry Pratchett sat down with Bill Gates for an interview and told him, plainly, that the internet was going to legitimize false information. That without peer review, without editorial gatekeeping, anything could be made to look authoritative. Gates disagreed. He argued that internet authorities would develop more sophisticated indexing and fact-checking than print had ever managed. He was wrong. Pratchett was right. It would take the world another twenty years to fully understand how wrong Gates had been.
The internet of the early 2000s had been built on an implicit faith in volume — the idea that more information meant better information, that the crowd would surface truth through sheer collective intelligence. Platforms like Facebook and Twitter weren't designed as publishers. They were designed as pipes. They didn't editorialize; they amplified. And amplification, it turned out, was not neutral. Pew Research Center found that over 60 percent of Americans were getting their news through social media. The pipes had become the press, without any of the accountability.
What made the architecture so vulnerable was also what made it so successful: engagement metrics. Platforms rewarded content that provoked reaction. Anger traveled. Fear traveled. Research would eventually confirm that false political information spread three times faster than other false news on Twitter — and that false tweets were significantly more likely to be retweeted than truthful ones. The algorithm wasn't biased toward lies exactly. It was biased toward novelty, toward emotional charge. Lies, it turned out, were frequently more novel and more charged than the truth.
The communities paying closest attention in 2016 were a fractured coalition: journalism professors, political scientists, a loose network of fact-checkers the Poynter Institute had begun organizing through its International Fact-Checking Network, launched in 2015. They watched the election cycle with increasing alarm. They documented. They published. Almost no one was listening.
The BuzzFeed analysis landed in November 2016 like a depth charge. Fake stories about Pope Francis endorsing Donald Trump. About Hillary Clinton selling weapons to ISIS. About FBI agents being found dead. Each one shared hundreds of thousands of times. Each one false. Researchers at CBS's 60 Minutes — producer Michael Radutzky and correspondent Guy Campanile among them — began pulling threads. They found Jestin Coler, who spoke openly about his operation. He wasn't a propagandist with an agenda, he claimed. He was a businessman who had found a market. The market was human credulity, and it was enormous.
By January 2017, the UK House of Commons had launched a parliamentary inquiry into the "growing phenomenon of fake news." Tim Berners-Lee — the man who had literally invented the World Wide Web — publicly identified fake news as one of three major disturbing trends threatening the internet he had built. Taiwan, watching what was happening to democracies elsewhere, introduced a media literacy curriculum for children that same school year. The problem had become institutional.
Then something strange happened to the phrase itself. Donald Trump began using "fake news" as a weapon against credible reporting he found unfavorable. The term, which had been coined to describe fabricated content, was now being deployed to describe journalism that was accurate but inconvenient. Researchers noticed the semantic collapse in real time. By late 2018, journalists at the Poynter Institute were actively discouraging use of the phrase. In October 2018, the British government officially removed "fake news" from official documents, following a recommendation by the House of Commons' Digital, Culture, Media and Sport Committee.
Claire Wardle of First Draft News had said it plainly in 2017: the phrase was "woefully inadequate." It flattened a spectrum of distinct phenomena — satire mistaken for fact, propaganda, state-sponsored disinformation, honest error — into a single, now-weaponized label. By naming the problem, the world had accidentally given its opponents a rhetorical cudgel.
What investigators confirmed was damning in its simplicity. Humans — not bots, not click farms, not automated systems — were primarily responsible for spreading false news. The machines didn't create the problem. People did, one share at a time, often with the best intentions, passing along something that confirmed what they already believed. The architecture enabled it. The engagement economy incentivized it. But the hands doing the sharing were human.
What remained contested was the question of intent and scale at the political level. Brian Stelter documented what he described as a long-term feedback loop between Donald Trump and Fox News presenters — a system that conditioned audience outrage over years — though the precise causal mechanism, who was driving whom, remained genuinely debated. Some commentators framed the entire 2016 fake news panic as moral hysteria, an overreaction by media elites to a phenomenon that had always existed. Others argued it represented a genuine and unprecedented rupture in the public's shared epistemic foundation.
The speculative edge of the research pointed somewhere uncomfortable. Inoculation theory — the idea that populations could be "prebunked," made resistant to misinformation before encountering it rather than corrected after — gained serious academic traction. Some researchers argued the rhetorical structure of content mattered as much as its accuracy; that fake news worked not because people were gullible but because it was architecturally superior at triggering the brain's pattern-recognition systems. Others theorized that declining trust in traditional media had created a vacuum, and that the vacuum was being filled by whatever was loudest.
By 2021, the Aspen Institute Commission of Inquiry had abandoned the term "fake news" entirely in favor of "Information Disorder" — a clinical reframing that acknowledged the problem was systemic, not episodic. Christine Michel Carter reported that year that one-third of Generation Alpha — children born after 2010 — could already identify false or misleading information in media. The generation raised inside the problem was developing antibodies.
The term is still everywhere. The phenomenon it was meant to describe has no agreed name. Researchers use "information disorder." Politicians use "fake news" to mean whatever serves them. Fact-checkers keep publishing. The engagement economy keeps rewarding outrage. And somewhere, someone is running a website that looks exactly like the truth, watching the clicks accumulate, waiting for the next share.
The architecture hasn't changed. Neither has the incentive.
SOURCES