Twitch has quietly updated its privacy settings, giving streamers and viewers a new way to opt out of having their content used to train Amazon’s AI systems. The change, spotted by users earlier this week, adds a toggle in the platform’s privacy controls that allows Twitch accounts to block Amazon from using their data for artificial intelligence development.
The update comes amid growing scrutiny of how major tech companies collect and use user-generated content to build and improve AI models. Amazon, which acquired Twitch in 2014 for nearly $1 billion, has been expanding its AI ambitions across its broader ecosystem—making the opt-out option a notable concession to user privacy concerns.
To opt out, Twitch users can navigate to their account’s privacy settings and disable the option that allows Amazon to use their content for AI training purposes. The setting applies to both past and future content, though the full scope of what data is covered—chat logs, stream recordings, clips—has not been fully clarified by the company.
Twitch has not made a formal public announcement about the change, and the lack of communication has drawn criticism from the streaming community. Many users only learned about the toggle through posts shared on Reddit and X, raising questions about transparency and whether the default setting should be opt-in rather than opt-out.
Digital rights advocates have welcomed the move, but note that burying the control deep within privacy settings limits its practical impact. For a feature to be meaningful, they argue, users need to know it exists.
Amazon has not commented publicly on the update. For now, Twitch users concerned about how their content is being used are encouraged to review their privacy settings directly and to keep an eye out for any further clarification from the platform on exactly what the new toggle covers.