The groups, including the Jewish students union (UEJF), SOS Racisme and SOS homophobie accuse Twitter of consistently failing to meet its legal obligations by not removing hate speech from its platform quickly enough and for not providing adequate moderation services.
According to a survey carried out by the groups between the 17 March and 5 May, "racist content went up by 40.5 percent, anti-Semitic comments rose by 20 percent and homophobic or anti-LGBT content increased by 48 percent".
'Unacceptable'
The groups signalled 1,110 examples of offensive messages to Twitter but found that only 12 percent of them had been erased within 3-5 days of publication.
"These are unacceptable results," said Dominique Sopo, president of SOS Racisme, in a statement. "This testing shows the massive lack of action on behalf of the platform, which is clearly refusing to invest in the human resources necessary to moderate the content that their activity is generating."
🔴 @SOS_Racisme accompagné de @uejf, @Jaccuse98686014 et @SOShomophobie assigne Twitter en justice pour son inaction face aux messages haineux.https://t.co/BNm5l6Shph
— SOS Racisme (@SOS_Racisme) May 12, 2020
The groups have asked a court in Paris to designate an expert in charge of auditing Twitter's efforts to invest in logistics and personnel to combat the rise of hateful expression, including "incitement to hatred and violence with regards to race or sex, or denying crimes against humanity".
Call for court to designate auditor
The role of the expert, they said, would clarify once and for all the way Twitter is dealing with the problem.
Twitter France director Audrey Herblin-Stoop defended the company's record when it came to investing in moderation technology.
"More than 1 out of 2 abusive tweets is detected by an automated system, rather than a user," she told AFP, a method she says reduces the pressure on users to signal inappropriate content.
"This ratio was 1 out of 5 in 2018," she added.
Faced with an increasing number of violent and aggressive content, major online platformshave put in place procedures such as the use of algorithms to filter and signal content, and more moderation staff.
Governments have also begun to set up committees to review online content.
National Assembly to review law
In France, the National Assembly, the lower house of parliament, adopted a law on Wednesday designed to counter hateful content on the internet.
It obliges social media platforms and search engines to remove the offensive content within 24 hours, or risk facing a fine of up to 1.25 million euros.
The time frame for removal is reduced to one hour if the content is related to terrorism or child pornography.
Although supported by the majority (LREM), the Socialists abstained from the vote, while the far right and left wing groups (RN, France Unbowed) voted against it.
The Covid-19 crisis has also given Twitter a new headache, with the rise of fake news.
Labels for misleading content on Covid-19
The company said on Monday it had begun tagging "potentially harmful content" related to the coronavirus pandemic.
Under the new effort, Twitter said "warning labels" would be added for misleading statements or assertions that have been confirmed to be false or inaccurate by public health authorities, as well as other claims about which the accuracy or credibility is contested.
