The Great Cheapfakes Caper
When debunking misinformation turns into spreading it
When we list the groups that should be embarrassed by the Great Cheapfakes News Cycle of 2024, let's not leave out the alleged experts on detecting misinformation. It's not just that some of them wound up misleading people themselves—sadly, that's no longer novel. It's that they did it in one of the ways they spent years highlighting as an online hazard.
For those of you who have forgotten the cheapfakes caper, here's a quick review. Last June, the media suddenly overflowed with stories about videos that seemed to show then-President Joe Biden detached from the events unfolding around him. These press reports dubbed the clips "cheapfakes," a term that became increasingly infamous after Biden's decline became undeniable.
This expression was not invented in the Biden era; it's a play on "deepfakes" that's been around at least as far back as 2019. It's a dumb word for a real phenomenon: photos or videos that mislead not with sophisticated new technologies but through such low-tech manipulations as cropping something important out of the frame or just mislabeling what we're watching. (Think of all the times that footage from other wars, or even from a video game, has been presented as a scene from Ukraine or Gaza.) In their simplest form, these videos were an example of another misinfo-hunting buzzword: malinformation, a term the Department of Homeland Security defines as material "based on fact, but used out of context." If a clip shows Biden with his eyes closed at a D-Day event, as though he had fallen asleep, but the full context reveals that he had merely shut them for a short period while listening to a translation, then presto! That's malinformation.
Or maybe it isn't: Even with that extra context, we don't necessarily know whether Biden was briefly nodding off. But I'll concede the point. Some of those videos were in fact misleading. Real events were presented in a deceptive way.
But that's also true of many reports about the Biden "cheapfakes," stories that bubbled over with quotes from people presented as experts on misinformation.
Two things were true at the same time: (1) Joe Biden was losing his grip, and (2) some of the videos proffered to prove this were junk. If you filed a story that focused on only the second truth while implying but never outright stating that the first truth was false, you could leave your readers with the wrong impression without ever saying a lie. Maybe because you set out to dupe people, or maybe because you yourself were so caught up in confirmation bias that you were blind to what you were doing.
And so you got passages like this lede to a piece The Washington Post ran last year:
President Biden, who at 81 is a couple of decades younger than many of the veterans he honored during Thursday's D-Day commemoration in Normandy, nonetheless found his age and fitness in the spotlight as selectively edited clips of him circulated online to paint the picture of a physically and mentally challenged commander in chief.
Every word of that sentence is true! Biden is indeed younger than those World War II vets. Some of the clips bopping around the internet were indeed selectively edited. But looking back 11 months later—hell, looking back two weeks later, in the aftermath of Biden's disastrous debate with Donald Trump—do those really seem like the two most salient facts?
People have been known to deploy the word malinformation in a bad-faith way. (As my colleague Jacob Sullum once cracked, the term sometimes seems to mean material that is "true but inconvenient.") But as with cheapfakes, there is a real phenomenon lurking behind the neologism. It is common, very common, for people to deceive by leaving out important context rather than lying outright. If you're serious about sorting truth from fiction, you need to be on alert for material that is, as the man from Futurama put it, "technically correct, the best kind of correct."
Yet the line between information and malinformation can be blurry, because there's always more context that can be added. This very article you're reading right now is missing all sorts of potentially relevant information. If you're a Republican, you might be seething that I haven't gone looking for a larger pattern of pro-Biden bias at certain outlets. If you're a Democrat, you might be tearing out your hair watching me build a whole article around Biden's mental decline without asking if President Trump's head is fully functional. I might reply that those just aren't the subjects of this story, but no doubt there are reporters who would say the same thing about the pieces on cheapfakes that they cranked out a year ago. I think there's a difference, but I also realize that we're looking at shades of gray. The moment you move beyond the limited task of judging whether an individual video is legit—the moment you're writing a bigger piece on those clips and What They Mean—is the moment you step into that gray zone.
So "malinformation" is real, but it is also capacious enough to swallow nearly everything. To quote Dan Williams, a philosopher at the University of Sussex who has established himself as one of the most capable critics of the conventional wisdom in this area, if you expand "misinformation" beyond clear-cut falsehoods to mean "anything that leads people to be misinformed," you'll find that "misleading communication isn't just widespread. It's so widespread in media, political communication, and social science that the concept of misinformation will either become useless or be applied in highly selective ways." Indeed, "cherry-picking, framing, and the omission of inconvenient context are endemic to almost all political communication and punditry. Which politicians or pundits present the 'whole truth' about an issue?"
The cheapfakes caper shows that even debunking misinformation can be misinformative, or malinformative, or whatever the proper adjective is here. We should still debunk things, of course. But one round of debunking won't always be enough.