Twitch will be using machine learning to combat users circumventing bans. The platform announced its new Suspicious User Detection system to flag users who may be creating new accounts to evade bans from a streamer’s channel.
The system, which will be turned on by default for all Twitch users, employs two classifications for moderating users potentially in violation of bans: “likely” and “possible.”
While “likely” ban evaders will still be able to visit the streamer’s channel, their messages won’t be sent to the chat. The user will also be visible to Creators and Mods so that they can simply monitor or choose to ban the account.
For those only in “possible” violation, their account will be flagged for closer moderation, but they’ll still be able to send chat messages.
In addition to the new system, Creators can now opt to restrict messages from both likely and possible ban evaders.
Twitch explained in its announcement that its machine learning tool may have some faults at first:
“One thing to prepare for, particularly around launch, is that no machine learning will ever be 100% accurate, which means there is a possibility of false positives and false negatives,” the company said. “That’s why Suspicious User Detection doesn’t automatically ban all possible or likely evaders.”
The company noted that the accuracy of the tool would improve over time as it learns from the decisions that users make against ban evaders.
Suspicious User Detection, powered by machine learning, is here to help you identify and restrict suspected channel ban evaders from chatting before they can disrupt your stream.
Learn more here: https://t.co/01cCwnQZfw pic.twitter.com/QWVSnRPg1X
— Twitch (@Twitch) November 30, 2021
In other tech news, you can now display your NBA top shot NFTs in an Infinite Objects video frame.