Social media companies Facebook, Twitter and Google’s YouTube have greatly accelerated their removals of online hate speech, reviewing over two thirds of complaints within 24 hours, new EU figures show.
The European Union has piled pressure on social media firms to increase their efforts to fight the proliferation of extremist content and hate speech on their platforms, even threatening them with legislation.
Microsoft, Twitter, Facebook and YouTube signed a code of conduct with the EU in May 2016 to review most complaints within a 24-hour timeframe.
The companies managed to meet that target in 81% of cases, EU figures seen by Reuters show, compared with 51% in May 2017 when the European Commission last monitored their compliance with the code of conduct.
EU Justice Commissioner Vera Jourova has said previously she does not want to see a removal rate of 100 percent as that could impinge on free speech.
She has also said she is not in favor of legislating as Germany has done. A law providing for hefty fines for social media companies if they do not remove hate speech quickly enough went into force in Germany this year.
“I do not hide that I am not in favor of hard regulation because the freedom of speech for me is almost absolute,” Jourova told reporters in December.
“In case of doubt it should remain online because freedom of expression is (in a) privileged position.”
Of the hate speech flagged to the companies, almost half of it was found on Facebook, the figures show, while 24% was on YouTube and 26% on Twitter.
The most common ground for hatred identified by the Commission was ethnic origins, followed by anti-Muslim hatred and xenophobia, including expressions of hatred against migrants and refugees.
Following pressure from several European governments, social media companies stepped up their efforts to tackle extremist content online, including through the use of artificial intelligence.
The Commission will likely issue a recommendation, a soft law instrument, on how companies should take down extremist content related to militant groups at the end of February, an official said, as it is less nuanced than hate speech and needs to be taken offline more quickly.
