While the world was busy firefighting misinformation after the Covid-19 outbreak, hate speech on social media went up exponentially — Facebook removed 22.5 million pieces of hateful content from its platform between April and June, its latest community guidelines enforcement report, released on Tuesday night said.
It’s much higher than the 9.6 million pieces removed in the first three months of the year, and, in fact, the highest of all time. “Since October 2019, we have used (an enforcement strategy) 14 times to remove 23 banned organisations,” said Monica Bickert, Facebook’s vice president (content policy), in an international press call TOI attended.
That includes The Right Stuff (a neo-Nazi conspiracy theory site), Patriot Front (a white supremacists hate group), Proud Boys (a neo-Fascist organisation that only admits men), Combat 18 (a neo-Nazi group that advocates “white-only” countries), Identity Dixie (that organised the rally which ended with the Charlottesville massacre) and KKK (one of the oldest, most infamous white supremacist groups). About 4 million pieces of content related to organised hate were taken down and another 8.7 million of terrorism.
Facebook’s vice president (integrity) Guy Rosen put it down to “the increase in proactive technology” in detecting such content. Facebook has been moving towards engaging more machine learning systems for content moderation. But the same standards, it seems, did not apply to content related to suicide and self-injury.
Facebook acted on 911,000 pieces of content since April, as opposed to 1.7 million in the first three months of the year. “With fewer reviewers, we took on fewer pieces of content on both Facebook and Instagram for suicide and self-injury and on Instagram for child nudity and sexual exploitation,” Rosen added.
So while he said that content moderation will continue to be a process where human and machine review will co-exist, it seems certain kinds of posts will be easier to flag than others in the AI-driven mix.
“Between April and June, we removed more than 7 million pieces of harmful Covid-19 misinformation from Facebook and Instagram and put warning labels on 98 million such pieces,” Rosen said.
Among those was US President Donald Trump’s video last week, in which he falsely said that children were “almost immune” to Covid-19. The Facebook Civil Rights Audit last month had pointed out the problems in differentiating political content from others while moderating. “Elevating free expression is a good thing, but it should apply to everyone. When it means that powerful politicians do not have to abide by the same rules that everyone else does, a hierarchy of speech is created that privileges certain voices over less powerful voices,” the report had said.
Facebook’s latest report came the same day that two Oxford University researchers published a paper on the influence of the social media giant at a time big tech is under scrutiny for its monopolistic hold on people’s information and, by extension, lives.
“The advertiser boycott in the wake of the Black Lives Matter protests highlights the commercial risks to Facebook of failing to respond adequately to the social justice concerns of its users and customers,” the paper, by Nikita Aggarwal and Carl Ohman in the ‘Internet Policy Review’ said. “Indeed, Facebook is by far the largest archive of human behaviour in history.
Leave a Reply