ADL report accuses Steam of "normalizing hate and extreme views in the gaming community"

The Anti-Defamation League (ADL), a US-based organization that fights antisemitism while supporting Israel, published a report in which it claimed that Steam was "rife" with antisemitism, and accused Valve of allowing hateful and extremist content to spread through its "highly permissive content policy."

The ADL's report is based on an "unprecedented platform-wide" analysis, which it describes as a "platform-wide" analysis involving more than 458 millions user profiles, 152,000,000 profile and user group avatars, and 610,000,000 user comments. The ADL Center on Extremism discovered "millions of examples of extremist content, including explicit hatred symbols like sonnenrads, and 'happy retailers,' as well copypastas (blocks that are copied and past to form images or longer-form writing), shaped into swastikas", being shared on the platform.

The ADL stated that "the clear gaps in Steam’s moderation of these content inflict damage by exposing untold numbers of users to hate and harassing, enabling radicalization and normalizing hatred and extremism within the gaming community." Understanding the extent of extremist content and hateful material on the platform is crucial to combating the proliferation of hatred online.

According to the report, Pepe the Frog, and swastikas, respectively, represent 54.6% and 9,1% of detected symbols. The ADL found "tens and thousands of pieces of terrorist-related content" on Steam Community, including more than 15 000 public accounts with profile photos featuring flags or logos from ISIS, Hezbollah Hamas and others. Hundreds of accounts used photos of white supremacist murderers Brenton Tarrant, Anders Breivik and others as their avatars.

The report also touches on game mods. It claims to have found hundreds, including Garry's Mod and "that specifically refer to mass shootings."

The report states "Steam's policy on public-facing content does not include any mention of extremism or hate speech." This isn't exactly accurate. The Steamworks documentation explicitly states that content containing "hate speeches, i.e. Speech that promotes hate, violence, or discrimination towards groups of people on the basis of ethnicity, religion or gender, age, disability, or sexual orientation should not be published. Steam's code does not specifically mention hateful or extreme content, but it does prohibit harassment, disruptive or destructive behavior, as well as the violation of personal rights by other users.

Questions about the effectiveness and application of Steam's moderating policies are more pertinent. The report states that Steam is "technically" capable of moderating content and has done so in the past "selectively" as a response to external pressure. One example would be the removal Pepe the frog emoticons from Steam Marketplace after a DMCA notice filed by Pepe creator Matt Furie. Valve's overall approach is "largely ad-hoc" and it has failed to systematically address extremism and hatred on the platform.

Valve updated their community rules in 2023, providing a more detailed breakdown of what's allowed and what's not. However, as noted at the moment, the real issue was how they would enforce these rules. This has been a problem for Steam since the beginning: we reported on Valve’s “hate group issue” in 2018 and in 2022 US senator Maggie Hassan wrote to Gabe Newell, Valve’s boss, stating that "a significant number of users are displaying and advocating neo Nazi, extremists, racial superiority, misogynistic and other hateful feelings."

The report states that "Valve must make significant changes in their approach to platform governance, both in terms policy and practice, to address the ways hate and extremism has proliferated on Steam."

"As ADL stated previously, policymakers need to demonstrate their commitment in preventing hate and harassment within online multiplayer games. While the government is focusing on the dangers of social media and AI in general, policymakers should also be paying attention to the immediate threats that are prevalent in online gaming environments.

The ADL has also recently put pressure on Twitch, resulting in temporary bans of a group Arab streamers for making jokes at a TwitchCon Panel that the organization interpreted to be antisemitic.

I've contacted Valve to ask for a comment on this report. I will update you if I get a response.