AI social media moderators harsher than humans, MIT-led team finds

1 min read
FILED - Unwanted post sanctions are coming: Experts believe the AI moderators being deployed to take down inappropriate posts on social media are going to be far too harsh. Photo: Silas Stein/dpa

Washington (dpa) – The alarm around so-called artificial intelligence(AI) and machine learning has prompted the technology’s “godfather”Geoffrey Hinton to lament his life’s work.

AI, the warnings go, could see widespread job losses or worse, should
it “get smarter than people”, as Hinton put it when he quit his job
at Google in early May. 

Hinton’s concerns followed tech business bosses putting their names
to a letter calling for a six-month pause on AI advances, lest we
”develop nonhuman minds that might eventually outnumber, outsmart,
obsolete and replace us” and in turn ”risk loss of control of our

A less hair-raising warning came on May 10, with the publication in
Science Advances of research showing AI to be a harsher judge of
social media posts than human counterparts. 

AI moderators are used instead of people to clear the backlogs of
complaints and reports generated by the fickle billions of social
media users, but ”often do not replicate human decisions about rule
violations,” the researchers found.

Not only that, but AI is “likely to make different, often harsher
judgements than humans would,” if not trained with the right data. 

”I think most artificial intelligence/machine-learning researchers
assume that the human judgements in data and labels are biased, but
this result is saying something worse. These models are not even
reproducing already-biased human judgements because the data they’re
being trained on has a flaw,” said Marzyeh Ghassemi, an assistant
professor at Massachusetts Institute of Technology (MIT).

”We are going to end up with systems that are going to have extremely
harsh moderations, much harsher than what humans would do. Humans
would see nuance or make another distinction, whereas these models
don’t,” Ghassemi warned.

So, for anyone who ends up unemployed because of AI, there might beno point posting a fiery complaint on Twitter or Facebook, railingagainst the injustice of it all. Who knows, an AI moderator might getwind of it and sanction the account.

Join Our Mailing List

[contact-form-7 id="9206" title="Sidebar Newsletter"]