The public appetite for moderating toxic speech on social media depends on the type of person being targeted, researchers have found, with billionaires being seen as in least need of protection.
Content moderation is a hot-button issue, with social media platforms taking drastically different approaches. Elon Musk bought Twitter, since renamed to X, in part because of his concern about over moderation infringing the right to free speech. Others feel platforms don’t do enough to protect users from hate and harm.
But perceptions differ wildly. “When it comes to toxic online speech, it actually matters who is targeted,” says at the Technical University of Munich, Germany.
Advertisement
Pradel and her colleagues asked around 9000 US adults to read social media posts directed at three user profiles: a billionaire, a member of the LGBTQ community and a Christian truck driver. Participants saw some posts simply disagreeing with a user, while others attacked the same user using uncivil, intolerant or threatening language, collectively known as toxic speech.
The team asked the participants how they thought a social media platform should respond to each post, with the options including doing nothing, removing the post, suspending the person’s account, placing a warning label on the post or reducing its visibility.
Generally, around 10 per cent of participants recommend action against argumentative but non-toxic posts, while that jumped to around 30 per cent for uncivil or intolerant language and over 50 per cent for threatening posts.
But the team also found that opinions on taking action varied depending on who was being attacked. In all, 58 per cent agreed that posts attacking an LGBTQ person should be removed, compared with 36 per cent for posts targeting a Christian truck driver and 21 per cent for those against a billionaire. “It shows us that it might be that when someone is not perceived as a protected group, then users may be less likely to demand moderation,” says Pradel.
One limitation of the study is that the participants skewed younger, better educated and more likely to be Democrats than the general US population, though Pradel says this also reflects the skewed demographics of social media users. The team also found that participants who were Democrats were more likely than Republicans to demand action of some sort was taken against harmful posts.
at Northumbria University, UK, says she welcomes the fact that LGBTQ hate warranted more moderation than attacks on groups, but thought it concerning that so few participants felt the platforms ought to act against hate speech, with the majority favouring no action. This is despite laws in many nations requiring platforms to actively police hate speech.
“Something the study captures inadvertently is the passivity and almost defeatism of general audiences when it comes to platform power, meaning that it will take society and the [research] field a lot of work to galvanise these people to hold platform power to account,” she says.
Pradel echoes this sentiment. “We need universal standards for content moderation,” she says. “Platforms should help hold you to a higher standard than maybe us using them might perceive.” That is the case despite users’ seeming ease with a light-touch regime. “Even if users do not demand moderation,” says Pradel, “it must be ensured that all groups are protected.”
American Political Science Review
