Twitch has confirmed it will use streamers’ content to train generative AI models, with the setting enabled by default rather than requiring consent. The move has prompted substantial criticism from creators, who argue they should be able to choose before their streams are used; an opt-out option is available, but Twitch’s decision highlights wider concerns about platform control over user-created material.

Users can disable generative-AI training through Security and Privacy settings on either the web or mobile app, and changing it once applies across both. Twitch says opting out will not affect discoverability, but it will still use data for other AI-supported functions such as AutoMod, captions and transcription. Chief Product Officer Mike Minton said an opt-in system would attract little participation, while noting that other services also train AI on user content.