Skip to main content

Twitch overhauls its rules around harassment and hateful conduct

Twitch overhauls its rules around harassment and hateful conduct


They’ll go into effect on January 22nd

Share this story

Illustration by Alex Castro / The Verge

Today, Twitch is announcing changes to its policies around harassment and hateful conduct. The reason, the company says, is because marginalized people — “women, members of the LGBTQIA+ community, Black, Indigenous, and people of color,” the company wrote in a blog post — are disproportionately harassed on the site. The new policy is far more specific about what constitutes harassment and abuse. The larger changes include stricter protections around sexual harassment, the policy extending off platform to an extent, and that enforcement in cases that are harder to judge will rely on external signals that the behavior was unwanted (like channel timeouts).

A bigger overhaul has come to Twitch’s policy regarding sexual harassment; it’s now much more stringent in what is and isn’t permitted, and it now has its own category in the terms of service. The policy now prohibits “[r]epeatedly commenting on someone’s perceived attractiveness, even in what you believe to be a positive or complimentary manner,” if there’s an indication the behavior is unwelcome (if you’ve been asked to stop, been banned from the channel, or timed out); it also forbids “[m]aking lewd or explicit comments about anyone’s sexuality or physical appearance,” specifically stating that the company will not make an exception for public figures.

Twitch says that new rules will apply to content created on January 22nd — the day it goes into effect — and afterward. The company also says that while the policy is more stringent in some ways, it will also dole out punishments according to severity. Enforcement will also depend on signals from targeted individuals in cases where potentially offending behavior is more subjective. “In the event of a more subjective situation, for example competitive banter or trash talk, will rely on signals from the person being targeted, or their mods, so that we can take appropriate action and avoid punishing users in situations where that behavior is welcome,” the company wrote in a blog post.

Stream sniping gets its own line

Mostly, it seems that the new policy is aimed at plugging the holes in the old one. For example, the Confederate flag is now banned in chat, and promoting “hateful viewpoints under the guise of education or comedy” will lead to a suspension. Stream sniping gets its own line; it’s banned. And broadcasters who don’t make an effort to curtail “incitement or organization of abuse” in their own channels will be banned.

Twitch says it has worked with its Safety Advisory Council and industry experts for months on the new rules, and the effort shows. Streamers and audiences should read the new guidelines here, because they’re worth reading in full. Twitch is also hosting three events over the next month to familiarize broadcasters and their communities with the new rules: a “Creator Camp” discussion of the policy on December 11th, a town hall on December 16th covering enforcement, and a final Creator Camp on January 6th to review the policy again and answer frequently asked questions.

All told, it looks like Twitch has decided to take serious action against its most toxic users. But as with any policy, its effectiveness depends on the company’s willingness to enforce its own rules.