Twitch lets users block Amazon from using their content for AI training
Twitch has introduced an option letting users opt out of having their content used to train Amazon's AI models, following backlash from the platform's community. The setting remains opt-out by default.
Livestreaming platform Twitch has announced that users can now opt out of a default setting allowing owner Amazon to use their streams, clips, images, chat logs and other channel content to train generative AI models.
The Wednesday announcement confirmed that the tech giant, which acquired Twitch in 2014, has indeed been using the platform's content for AI training — a revelation that sparked widespread criticism among users.
Amazon operates a range of AI services and has invested heavily in their development as it competes with Google, Meta and other technology companies.
Opt-out by default
Twitch's chief product officer, Mike Minton, explained during a livestream that AI training data collection is enabled by default, as it is across most content platforms, but said Twitch is respecting users' wishes to disable it if they choose. Asked why the setting wasn't opt-in instead, Minton said: "If it's opt-in, nobody would opt-in. That's the honest answer."
Users who remain opted in could have any of their channel content used for training, though Minton said the collected data would not be resold to other companies. Twitch's head of community, Mary Kish, walked viewers through disabling the feature via channel settings, noting that data collection for AI training has become standard across the industry.
The stream's chat filled with hundreds of critical comments from viewers. Some argued Twitch had missed a chance to set an industry standard by pushing back against default data collection. It remains unclear when Amazon began collecting Twitch users' data for AI training, and Minton acknowledged he did not know whether or how such data had already been used in model training.
