The Anti-Defamation League, a US organization that aims to combat antisemitism and support Israel, has published a report alleging that Steam is "rife with extremism and antisemitism," and accusing Valve of allowing the spread of hateful and extremist material through a "highly permissive approach to content policy."
The ADL said its report is based on what it described as an "unprecedented, platform-wide" analysis of public data on Steam, including more than 458 million user profiles, 152 million profile and user group avatar images, and 610 million user comments. The ADL's Center on Extremism found "found millions of examples of extremist and hateful content, including explicit hate symbols like sonnenrads and 'happy merchants,' as well as copypastas (blocks of text that are copied and pasted to form images or long-form writing) shaped into swastikas" being shared on the platform.
"The clear gaps in Steam's moderation of this content inflict harm by exposing untold users to hate and harassment, enabling potential radicalization and normalizing hate and extremism in the gaming community," the ADL wrote. "Understanding the extent of extremist and hateful content on the platform is key to fighting the proliferation of hate online."
Pepe the Frog and swastikas are the most common extremist symbols found on Steam, according to the report, respectively representing 54.6% and 9.1% of detected symbols. The ADL also found "tens of thousands of pieces of terrorism-related content on Steam Community," including more than 15,000 public accounts with profile pictures featuring the flags or logos of ISIS, Hezbollah, Hamas, and others. Hundreds of accounts were found using photos of white supremacist mass murderers Brenton Tarrant and Anders Breivik as their avatar images.
Game mods are also touched on in the report, which claims to have found hundreds of mods for games, most notably Garry's Mod, "that specifically reference mass shootings."
The report says "Steam's public-facing content policy includes no mention of hate or extremism," which isn't entirely accurate: The Steamworks documentation specifically states that content with "hate speech, i.e. speech that promotes hatred, violence or discrimination against groups of people based on ethnicity, religion, gender, age, disability or sexual orientation" should not be published on the platform. Steam's code of conduct does not make specific reference to hateful or extremist content, but does forbid harassment, disruptive or damaging behavior, and the violation of the personal rights of other users.
More on-point are questions about the application and effectiveness of Steam's moderation policies. The report says Steam appears to be "technically capable" of moderating content, and has done so "selectively" in the past in response to external pressure: One example of that would be the removal of Pepe the Frog emoticons from the Steam Marketplace following a DMCA takedown noticed filed by Pepe creator Matt Furie. But its overall approach has been "largely ad hoc, with Valve failing to systematically address the issue of extremism and hate on the platform."
Valve updated its community rules in 2023 to provide a much more granular breakdown of what is and isn't allowed, but as noted at the time, the real question was how it would go about enforcing those rules. Because this has been an ongoing problem for Steam: We reported on Valve's "hate group problem" in 2018, and in 2022 US Senator Maggie Hassan sent a letter to Valve boss Gabe Newell alleging "a significant presence of users displaying and espousing neo-Nazi, extremist, racial supremacist, misogynistic, and other hateful sentiments," and asking what Valve was doing about it.
"Valve needs to make significant changes to their approach to platform governance both in terms of policy and practice to address the ways in which hate and extremism have proliferated on the Steam platform," the report states.
"As ADL has stated previously, policymakers must demonstrate their commitment to disrupting hate and harassment in online multiplayer games. While government is necessarily focused on the dangers posed by social media and AI, policymakers must also pay attention to the immediate threats pervasive in online gaming environments."
The ADL also put pressure on Twitch recently, leading to temporary bans for a group of Arab streamers who made jokes at a TwitchCon panel that the organization interpreted as antisemitic.
I've reached out to Valve for comment on the report and will update if I receive a reply.
