Zeynep Tufecki is one of the shrewdest writers on technology around. A while back, when researching an article on why (and how) Donald Trump appealed to those who supported him, she needed some direct quotes from the man himself and so turned to YouTube, which has a useful archive of videos of his campaign rallies. She then noticed something interesting. “YouTube started to recommend and ‘autoplay’ videos for me,” she wrote, “that featured white supremacist rants, Holocaust denials and other disturbing content.”
Since Tufecki was not in the habit of watching far-right fare on YouTube, she wondered if this was an exclusively rightwing phenomenon. So she created another YouTube account and started watching Hillary Clinton’s and Bernie Sanders’s campaign videos, following the accompanying links suggested by YouTube’s “recommender” algorithm. “Before long,” she reported, “I was being directed to videos of a leftish conspiratorial cast, including arguments about the existence of secret government agencies and allegations that the United States government was behind the attacks of 11 September. As with the Trump videos, YouTube was recommending content that was more and more extreme.”
If you think that this is all about politics, think again. Tufecki tried watching videos on non-political topics such as vegetarianism (which led to videos about veganism), and jogging (which led to items about running ultramarathons). “It seems,” she reflected, “as if you are never ‘hardcore’ enough for YouTube’s recommendation algorithm. It promotes, recommends and disseminates videos in a manner that appears to constantly up the stakes. Given its billion or so users, YouTube may be one of the most powerful radicalising instruments of the 21st century.”
Tufecki’s right. We know from the research of Jonathan Albright and others that YouTube has become the key disseminator of “alt-right”, conspiracist, white-supremacist and other unsavoury propaganda. In the old days, if you wanted to stage a coup, the first thing to do was to capture the TV station. Nowadays all you have to do is to “weaponise” YouTube. After all, its first motto was “broadcast yourself”. Accordingly, if governments of the western world really wanted to cripple these disruptive forces, then shutting down YouTube would be a giant step forward. It wouldn’t prevent other such services springing up, of course, but none would have the power and reach that YouTube’s billion-strong network effect provides.
This doesn’t mean that YouTube’s owner (Google) is hell-bent on furthering extremism of all stripes. It isn’t. All it’s interested in is maximising advertising revenues. And underpinning the implicit logic of its recommender algorithms is evidence that people are drawn to content that is more extreme than what they started with – or perhaps to incendiary content in general.
So YouTube (like Facebook) is caught between a rock and a hard place. On the one hand, it’s embarrassed by the way in which it is being exploited by unsavoury actors (and also possibly worried about the longer-term threat of regulation); on the other hand, its bottom line is improved by increasing “user engagement” – ie, keeping people glued to YouTube. And since it’s a capitalist company, those revenues just have to keep growing.
Watching social media executives trying to square this circle is like watching worms squirming on the head of a pin. The latest hapless exhibit is YouTube’s chief executive, Susan Wojcicki, who went to the South by Southwest conference in Texas last week to outline measures intended to curb the spread of misinformation on her platform. This will be achieved, apparently, by showing – alongside conspiracy-theory videos, for example – “additional information cues, including a text box linking to third-party sources [about] widely accepted events, like the moon landing”. It seems that the source of these magical text boxes will be Wikipedia.
All of which makes one wonder which planet Wojcicki currently inhabits. She clearly knows nothing of conspiracy theories, for example, and has a touching faith that those who hold such beliefs are susceptible to evidence that might refute them. Nor does she understand that our current crisis of disinformation and computational propaganda will not be resolved by just finding and publishing “the facts”, whatever they are. Indeed, one of the most obvious implications of the proposed YouTube strategy is that it will turn Wikipedia into an even bigger epistemological battleground than it is at present.
In fact, this is just another distraction from the fundamental issue, which is that social media platforms cannot solve the societal problems they have created – because, ultimately, doing so will hurt their revenues and growth. This is the unpalatable truth they are all squirming to avoid. And in doing so they’re really just confirming HL Mencken’s observation about the impossibility of getting someone to understand a proposition if his income depends on not understanding it. It’s not that the companies don’t get it, just that they cannot afford to admit that they do.
What I’m reading
Out of the loop?
New York Times columnist Farhad Manjoo wrote an interesting piece about unplugging from the net for two months and how much his life had improved. But then a spoilsport on the Columbia Journalism Review spotted Manjoo tweeting during his e-detox.
Reforming Reddit
A fascinating New Yorker story by Andrew Marantz about Reddit and its efforts to purge vile content from its site. Hint: it’s not easy.
Tweet defeat
As website Cnet reports, the US library of Congress has given up archiving all public tweets, and is now only collecting them selectively.
