Twitch has enabled users to prevent their streams, recorded videos, clips, chat logs, and channel images from being used to train Amazon’s generative AI models. This change applies specifically to future training for systems designed to synthesise text, audio, images, or video. The setting is located within the account preferences menu under the privacy section. Selecting the option ensures that the specific data associated with that channel will not contribute to the development of these artificial intelligence tools. Other AI-supported features such as live captions and safety filters will continue to operate normally even when this selection is active. However, participation in chat on another streamer’s broadcast is subject to that broadcaster’s chosen privacy settings. If a streamer opts out, the chat data from their channel cannot be used for training purposes regardless of whether the viewer has opted in. This update addresses growing concerns among content creators regarding data ownership and commercial usage of their digital footprint.
- Opt-out applies only to generative AI training, not standard AI features like captions.
- Chat data usage depends on the streamer’s preference, not the viewer’s.
- Settings are managed via the Twitch account privacy menu.
