Attempts at top down information control are counterproductive.
After eight years (perhaps better counted as two US electoral cycles) of partnering with fact checkers, Meta is making major changes to the way it handles misinformation and other harmful content on its platforms, Facebook, Instagram and Threads. Meta will, according to CEO Mark Zuckerberg:
A lot of it is pretty vague, but it’s clear there’s a major paradigm shift. Even some of those who support the move have described it as a “knee bend”.
But the media-industrial complex in miniature which has developed around the topic of misinformation, universally working on projects where ordinary people are expected to accept the rulings of unknown “experts” with humanities degrees, and fully committed to learning nothing from its failures, is having a fit about it.
‘Today Mark Zuckerberg said “Fact checkers have just been too politically biased and have destroyed more trust than they’ve created.” This is patently false. ’ – wrote Andrew Dudfield (He/Him), the Head of AI at Full Fact on Linkedin. As if such a statement, about the presence or absence of political bias, could ever be “patently” false (or true). This habit of selectively ignoring inherent ambiguity is one reason among many that people don’t trust people like him and the organisations for which they work.
He was sharing a post by something called the European Fact-Checking Standards Network, who criticised Zuckerberg saying he made “statements linking fact-checking with censorship. Fact-checking is not censorship,” but Zuckerberg never said it was. He does not use the word censorship to describe what fact checkers do (but did say that they had “destroyed more trust than they’ve created”). He only used the word censorship to describe what their automated filters do.
To be fair, they never said he did. They just said he made “statements linking fact-checking with censorship”, then, in a subsequent sentence, said “Fact checking is not censorship”, giving themselves plausible deniability when people inevitably join the dots. And down the he-said-she-said slip-and-slide we go – to hell.
Similarly, the Poynter Institute, which runs the International Fact Checking Network, republished a piece originally published on Politifact, which is a member of the IFCN, as well as being directly run by Poynter, which aims to explain the system that’s being shut down. They were keen to emphasis that they had a narrow role, determined for them by Facebook. They didn’t censor, and, according to them, they didn’t even choose which items got checked. They write “Meta surfaces potential misinformation”. But in the same article they link to Meta’s explanation which states “Fact-checkers can identify hoaxes based on their own reporting” (as well as being asked by Meta).
So which is it? And why should anyone trust such a complex and obscure arrangement?
As Zuckerberg says in his announcement, the problem with complex systems is that they make mistakes.
But as my throbbing linkedin feed (I’ve been working in this space far too long) attests, the people with jobs in that complex system resent the changes – which will mean an end of the substantial flows of cash from Meta into their absurd, useless little ecosystem.
But they’re now screaming into the void. No one is listening. They fundamentally misunderstand the moment and their place in it. Taking a big picture view, we are moving away from the institutions which defined modernity, as they are displaced by interfaces. Zuckerberg is a seed of the new elite, associated with the interfaces. Up until now, they’ve been taking their lead from the old elite, which came up out of institutions. Not any more.
This is – even if it is being done for questionable reasons – a good thing as it signals both progress in our society-wide transition, which we might as well get on with, and more specifically, the end of the top-down approach to combatting misinformation, which was always illegitimate, always counterproductive.
Follow Austin on Bluesky at @austingmackell.bsky.social and on threads at @austinmackell
