The Suspicious User Detection tool will identify users that are “likely” or “possible” to have evaded bans.
A Way to Fight Harassment
Twitch is one of the most popular streaming platforms these days, boasting more than 9.5 million creators.
Streaming, however, is not always fun and games. The platform has had issues with harassment for a while now. As part of its efforts to deal with the problem, it is introducing a machine learning feature.
The Suspicious User Detection tool will flag potential trolls. It will separate them into two categories: “likely” and “possible”.
Those under “likely” won’t get access to chats at all. Creators and Mods, however, will be able to see them and choose between monitoring the user, restricting them, or banning them.
“Possible” ban evaders will be able to participate in chats, but Creators and Mods will be notified that they pose a threat.
In the end, it will be up to the streamer and their helpers if they wish to band the person. That’s because it’s impossible to create a 100% accurate tool and mistakes might happen.
When someone is banned from a channel, they won’t be able to reach the community at all.
A Twitch spokesperson explained: “The tool is powered by a machine learning model that takes a number of signals into account — including, but not limited to, the user’s behavior and account characteristics — and compares that data against accounts previously banned from a Creator’s channel to assess the likelihood the account is evading a previous channel-level ban.”
The Suspicious User Detection is on by default.