|
 |
I’ve been writing some pessimistic things about AI recently, so I thought I should try to balance those out with some optimistic takes. One way I think AI could really help our society is by injecting reasonableness and moderation into our public discourse.
I’m known as a pretty nice and reasonable blogger nowadays. But when I got started, as an angry graduate student in 2011 trying to distract himself from his dissertation, I was genuinely snarky. Going back and rereading some of my posts from that era makes me chuckle, but also wince a little bit. The genteel éminence grises who sat atop the hierarchy of the very hierarchical economics profession just had no idea how to deal with a snarky, internet-native Millennial who was willing to talk back.
That snarky bravado, though sincere, was how I (accidentally) forced myself into the influencer elite. Paul Krugman, Brad DeLong, and other established bloggers liked how I tweaked the tails of the stuffy New Classical macroeconomists who pooh-poohed fiscal stimulus. So they boosted me on their own blogs, and pretty soon almost everyone in the economics profession knew my name — deservedly or not. Then I got Twitter, and I started tweeting way too much, and the rest is history. Notably, it was my political tweets — anti-Trump stuff in 2015-2020 — that got me my biggest bump in social media followership, rather than my economic insights.
In the media world of 1991, this career path would have been a LOT harder to pull off. I could have been a newspaper columnist or perhaps even a TV show host, but it would have been a long hard slog, gatekept by a bunch of editors who embodied the conventional wisdom of an older generation. My best bet for breaking in as an irreverent, independent voice probably would have been talk radio. In the media world of 1971, forget about it — I would have zero chance of breaking in to a discourse dominated by broadcast TV and big newspapers.
We can wonder whether the world would have been better or worse had I never become a public intellectual (hopefully, because you read this blog, your answer is “better”). But in my personal opinion, it’s pretty clear that the phenomenon of outsiders breaking in to the discourse with aggression and social media attention-seeking has gone too far. There is very clear evidence that social media — far more than the traditional media it replaced — has led to the elevation of divisive voices and bad actors.
For example, Bor and Petersen (2021) find that social media draws malignant, status-seeking people who use hostility to get attention and power:
Why are online discussions about politics more hostile than offline discussions?…Across eight studies, leveraging cross-national surveys and behavioral experiments (total N = 8,434), we [find that] hostile political discussions are the result of status-driven individuals who are drawn to politics and are equally hostile both online and offline. Finally, we offer initial evidence that online discussions feel more hostile, in part, because the behavior of such individuals is more visible online than offline. [emphasis mine]
Basically, spreading hate and divisiveness on social media is a form of entrepreneurship. As Eugene Wei has written, social media is all about getting social status. 10,000 followers on X may not sound like a media empire to rival CBS News, but for most people it’s more attention than they would otherwise get in their entire life. For malignant individuals who crave status and attention and enjoy spreading fear and hate, social media is a natural platform for their dark dreams.
This is especially effective because the psychology of viral content tends to spread negativity more than positivity. Here’s Knutson et al. (2024):
We analyzed the sentiment of ~30 million posts (on twitter.com) from 182 U.S. news sources that ranged from extreme left to right bias over the course of a decade (2011–2020). Biased news sources (on both left and right) produced more high arousal negative affective content than balanced sources. High arousal negative content also increased reposting for biased versus balanced sources…Over a decade, the virality of high arousal negative affective content also increased, particularly in…posts about politics. Together, these findings reveal that high arousal negative affective content may promote the spread of news from biased sources.
And Brady et al. (2021) find that social media outrage is a self-reinforcing process:
Moral outrage shapes fundamental aspects of social life and is now widespread in online social networks. Here, we show how social learning processes amplify online moral outrage expressions over time. In two preregistered observational studies on Twitter (7331 users and 12.7 million total tweets) and two preregistered behavioral experiments (N = 240), we find that positive social feedback for outrage expressions increases the likelihood of future outrage expressions, consistent with principles of reinforcement learning.
Together, these effects probably explain why negative content — especially about people’s political enemies — is so much more common than positive content on social media. Here’s Watson et al. (2024):
Prior research demonstrates that news-related social media posts using negative language are re-posted more, rewarding users who produce negative content…Data from four US and UK news sites (95,282 articles) and two social media platforms (579,182,075 posts on Facebook and Twitter, now X) show social media users are 1.91 times more likely to share links to negative news articles….[U]sers [show] a greater inclination to share negative articles referring to opposing political groups. Additionally, negativity amplifies news dissemination on social media to a greater extent when accounting for the re-sharing of user posts containing article links. These findings suggest a higher prevalence of negatively toned articles on Facebook and Twitter compared to online news sites.
And as if that wasn’t bad enough, social media platforms algorithmically amplify divisive content, probably as a business strategy! Here’s Milli et al. (2024):
In a pre-registered algorithmic audit, we found that, relative to a reverse-chronological baseline, Twitter's engagement-based ranking algorithm amplifies emotionally charged, out-group hostile content that users say makes them feel worse about their political out-group.
And research also finds that algorithmic feeds tend to increase political polarization.
In other words, the rise of social media created a revolution in political discourse. The old-school monopoly of big newspapers and TV stations — already under strain from the Web and from increased entry and competition — was overthrown by a giant mob of wannabe influencers, using divisiveness, partisanship, ideology, tribalism and negative emotions to get attention and status.
I call these people the Shouting Class. The most successful among them include people like Nicholas Fuentes, a literal Hitler supporter who has called for women to be sent to “gulags”; Candace Owens, a conspiracy theorist and antisemite; and Hasan Piker, who has said that America deserved the 9/11 attacks. But the real damage is probably done by the vast legions of smaller-time shouters, all dreaming of becoming the next Fuentes or Owens or Piker. If you’re on X or Bluesky, you can probably name a few of them.
Regular people know, of course, that social media is ruled by monsters great and small. Here’s a poll from 2020 showing that Americans think social media has a negative effect on their society: