The Washington Post does it. So do The New York Times, CNN, USA Today, and The Daily Caller: fact-checking.

In fact (no pun intended), fact-checking is so successful/important/sought after/self-reproducing (take your pick) that in 18 years, the number of fact-checking organizations has grown from one in 2003 to 349 worldwide in 2021, according to the Duke Reporter’s Lab. And while verifying what really happened — discerning between what’s true and what is not — is inherently tied to journalism, it usually occurs before journalists file their stories. Fact-checking, however, usually happens after the fact — so after something worthy of fact-checking happened.

Fact-checking is often conducted by independent organizations, such as FactCheck.org or PolitiFact. But over time, more and more news organizations have also started doing their own formal fact-checks — thinking that this de facto journalistic genre of fact-checking can help keep misinformation at bay. To be clear, my prediction (or hope, rather) is about the news organization genre of fact-checking, not fact-checking as a whole.

The underlying assumption of fact-checking is, of course, that when presented with facts, people adjust their beliefs accordingly. And, in general, it seems to do just that: When presented with facts that correct erroneous beliefs, people sometimes will change their minds.

That said, their prior beliefs, political ideology, and knowledge play significant roles and can curb a fact-check’s corrective effects. A journalist is unlikely to change the mind of an ardent climate change denier with a fact-check on its anthropogenic roots, or the belief system of an anti-vaxxer with an article noting that vaccines don’t contain microchips, or the view of a Trump supporter that the 2020 election was stolen. This becomes particularly problematic in the context of a hyper-partisan political and media public sphere like in the U.S., where many topics are politicized and follow clear partisan lines.

While fact-checkers might assume that their work is inherently neutral — they’re only interested in facts, after all — in the U.S. environment, fact-checking both as a journalistic genre and as done by third-party organizations is anything but. People use fact-checks to dunk on the other side, to cherry-pick information, and to counter-fact-check other fact-checks. In addition, the decisions fact-checkers make on which claims to check — and which to ignore — is a form of gatekeeping that can enrage partisans.

But there’s a big difference between fact-checking done by existing news publishers and by third-party organizations. News organizations like The New York Times or CNN are already deeply embedded in the national media ecosystem, inherently understood as journalistic organizations. As such, the American right not only doesn’t trust these organizations — they actively mistrust them.

This raises an important question: Who is the audience for the journalistic genre of fact-checking?

A straightforward answer would be: a given outlet’s existing audience. A lot of fact-checking focuses on the “debunking” of conspiracy theories and misinformation. The New York Times, for example, has its own “Tracking Viral Misinformation” section where journalists cover misinformation-related topics and debunk false claims. So do many other news sites. A charitable reading would be that these sites want to protect their readers from misinformation.

But a more critical reading raises the question: Would those readers have even heard about the debunked conspiracy theories before the debunking? Put another way: Does publisher fact-checking have the potential to actively harm the readers by exposing them to misinformation? Does fact-checking the latest QAnon conspiracy or bogus voter-fraud claim introduce those ideas to people who would have never heard of them in the first place? Does telling your readers “No, 5G towers don’t give you Covid” make them more likely to associate 5G with some cloud of suspicion that wasn’t there before?

Fact-checks allow for two additional aspects. Firstly, they can contribute to people’s partisan identity by highlighting what the “other side” is doing. And secondly, fact-checking is, in its essence, a response to a stimulus. And this response can be and is being abused to bring disinformation into the limelight.

As danah boyd and Joan Donovan have highlighted in the context of journalistic (non-) coverage of extremism and suicide: Sometimes the best coverage is no coverage. I argue that fact-checking as a journalistic genre for news publishers likely accomplishes little more than what third-party organizations do, while potentially doing harm by responding to and echoing misinformation and disinformation.

This is, of course, not to say that journalists should stop checking facts. They should. But do it within your article, not in a separate fact-checking piece. Those separate pieces should be left to the third-party fact-checkers.