Twitch Deploys Ban Evasion System

Makes it harder to get on alt accounts.

Twitch; it’s a streaming service where people can stream themselves streaming. If there’s any topic that’s been dominating the Twitch landscape this past year it’s the problem of ‘hate raids,’ a growing trend of people mostly using bot accounts to harass streamers. Which streamers? Mostly of marginalized groups. Because bigots ruin everything.

This week Twitch announced the deployment of a new machine learning system to detect and stop ban-evaders in their tracks. Twitch’s machine learning isn’t expected to be perfect, which is why the system ultimately puts the onus on banning the user in the hands of the streamer and their moderators. Ban evasion detection is turned on by default for all channels and can be shut off at the streamer’s convenience.

When you ban someone from your channel, they should be banned from your community for good. Unfortunately, bad actors often choose to create new accounts, jump back into Chat, and continue their abusive behavior. Suspicious User Detection, powered by machine learning, is here to help you identify those users based on a number of account signals.

The system will place suspected users into “possible” and “likely” categories. By default “possible” flagged accounts are allowed to talk normally but are flagged for moderators. Likely accounts will need to be approved before their chat appears. Both of these options can also be changed to the streamer’s preference.

It’s a start.

%d bloggers like this: