24 Luglio 2023

Secondo l’ADL, i giochi online faticano a contenere i nomi utente con contenuto odioso

Usernames that include racist, misogynistic, antisemitic, anti-LGBTQ+, ableist and white supremacist terms go unmoderated on some of the most popular online games, according to a report published Monday by the Anti-Defamation League.

The report from the ADL Center for Technology and Society said usernames were one of the easiest pieces of content to moderate, since they’re screened by the registration system when an account is created. The companies’ failures to moderate hateful messaging in one of the first and most identifiable places indicates that gaming companies have not been doing enough to address hate and extremism on their platforms, according to the report.

On multiplayer games, usernames are clearly displayed to other players, and offensive names have been a tool to create a hateful and hostile gaming environment, the ADL said.

“Online games are one of the most important and potent digital social spaces for building connection and finding community,” said Daniel Kelley, director of strategy and operations at the ADL Center for Technology and Society. “Unfortunately, what game companies say they are doing to fight hate in these spaces and what we are finding in our research contradict one another.”

The report adds to a growing body of research that shows that online games have been used to spread extremist ideas and images. The ADL’s annual reports on extremism found a rise in hate and harassment in online games in the last four years. ADL’s 2022 survey collected responses from more than 2,000 Americans and found that the majority of adults and teens experienced hate and harassment while playing online games.

In December, lawmakers led by Rep. Lori Trahan, D-Mass., asked the companies behind the top games to explain their efforts to address in-game hate, harassment and extremism in their online games. In their responses, only two of 14 gaming companies highlighted their mechanism for checking names linked to extremist ideologies or inappropriate behavior.

Usernames also tend to be the most obvious indication of extremism on an online platform. A study published by the Journal of Computers in Human Behavior in 2022 analyzed usernames on Reddit and found that those who had offensive or sexually-explicit usernames were more likely to harass other users and produce harmful content.

ADL researchers examined five online multiplayer games: League of Legends, PUBG: Battlegrounds, Fortnite, Overwatch 2 and Call of Duty, all of which ranked among the top 25 games on Tracker Network, a website that monitors player count on popular games. The games, which were released between 2009 and 2022, have tens to hundreds of millions of players, according to Tracker Network.

Moderating hateful and extremist language in online games is challenging because the context of words frequently changes, new offensive terms can be introduced and extremists often use coded words. The report said companies should consult with experts and keep track of new words to tackle this issue.

Online multiplayer games have struggled in the past to moderate audio-based hate speech and harassment among users, who can communicate through headsets and microphones while playing together in a game.

“Game companies need to do more to address the spread of hateful and extremist ideologies on their online game platforms. At the same time, we need more transparency about how they are working to address these issues,” the ADL’s Kelley said.

Each game has a code of conduct or rules that prohibit offensive language or hate speech. Some games have specific policies on usernames, although the report calls into question how effective these policies are given the inconsistent moderation of usernames.

A spokesperson for Activision Blizzard, which developed Call of Duty: Warzone, and is the parent company of Blizzard Entertainment and the publisher of Overwatch 2, said in an email to NBC News, “We are constantly working to improve our moderation policies and tools to make our games safer and more inclusive for players. Our games are enjoyed by millions of players around the world, and the majority of them do not engage in offensive or hateful activity.”

Fortnite’s developer, Epic Games, did not immediately respond to a request for comment, and neither did Krafton, the publisher of PUBG:Battlegrounds.

Riot Games, the developer of League of Legends, did not respond to a request for comment.

In two games, researchers said they tested registering various offensive usernames from each of five categories: antisemitism, misogyny, racism, anti-LGBTQ+, ableism and white supremacism.

The findings showed inconsistent moderation of offensive terms based on each game’s policy. Researchers got mixed responses from each game’s registration system. For example, on Call of Duty: Warzone, one anti-LGBTQ+ username was allowed while another anti-LGBTQ+ username was rejected.

Overwatch 2 had the least offensive usernames, the report said. Some possible reasons the report cited were that Overwatch 2 was the newest of the five games, and its company Blizzard Entertainment has invested in online safety initiatives.


Copyright dell’immagine: Jung Yeon-je / AFP via Getty Images file