Most online hate targets women: new EU report

Women are the main targets of online hate, including abusive language, harassment and incitement to sexual violence, according to a European Union report.

This should encourage the EU and social media platforms to pay close attention to protected characteristics such as gender and ethnicity when moderating content, the EU's Agency for Fundamental Rights (FRA) said in its report.

The study was conducted on YouTube, Telegram, Reddit and X - formerly known as Twitter – in four EU countries between January and June 2022. It showed women were the main targets across all platforms and countries involved. Other affected groups included people of African descent, Roma and Jews.

The number of hateful posts targeting women was almost three times that of those targeting people of African descent across Bulgaria, Germany, Italy and Sweden, the four countries covered by the report, the EU agency said.

READ MORE:

Anti-Asian racism: What's it like to be Chinese growing up in the UK?

Asian racism: Does the Fu Manchu myth still haunt Western thinking?

Sweden threatens to deport law-breaking immigrants

Beyond specific minority groups, out of 1,500 posts assessed by content moderation tools, more than half were still considered hateful by human coders, while almost half of all such posts were direct harassment.

"The sheer volume of hate we identified on social media clearly shows that the EU, its Member States, and online platforms can step up their efforts to create a safer online space for all," FRA director Michael O'Flaherty said in a statement.

As a way to combat such online discrimination, the report recommended that platforms should pay particular attention to protected characteristics like gender and ethnicity in their content moderation, provide more guidance on what is considered illegal hate speech.

It even went so far as to suggest the European Commission and national governments should create and fund a network of trusted flaggers, while also testing technology for bias, meaning providers and users of automated content moderation tools should test their tech for built in discrimination. 

Under the EU's Digital Services Act, which came into force last year, very large tech platforms and search engines must do more to tackle harmful and illegal content or risk fines. Tech giants have been facing mounting scrutiny recently, with a surge in harmful content and disinformation following the Israel-Gaza conflict.

The FRA said it had been unable to access data from Facebook and Instagram for the study.

Subscribe to Storyboard: A weekly newsletter bringing you the best of CGTN every Friday