Amazon-owned Twitch has quietly implemented an automatic default setting that uses creators’ streams, VODs, clips, and chat logs to train internal generative artificial intelligence models. The policy change has sparked swift backlash from prominent streamers and the creator community over lack of affirmative consent.
Streaming platform Twitch has introduced a platform-wide update that automatically enrolls creators in training artificial intelligence models for its parent company, Amazon.
What the Opt-Out Setting Covers and Leaves Intact
The new toggle, titled Training for Generative AI,
is located within the Security and Privacy tab of an account’s channel settings.
However, the setting includes specific limitations. The documentation clarifies that turning off the toggle does not opt you out of Twitch and Amazon using your channel content for other purposes described in the Twitch Privacy Notice.
AI systems powering streamer growth tools, viewer discovery systems, and community safety features like AutoMod continue to process content regardless of the generative training toggle. Furthermore, chat messages are governed by the channel owner’s preference rather than the individual chatter’s setting, meaning users who opt out but participate in an opted-in channel have no separate protection for their messages.
Leadership Defense and the Default-On Controversy
The update became public knowledge when streaming reporter Zach Bussey shared a screenshot of the new toggle, prompting viral reactions across X and Reddit, as noted by Kotaku. During a live creator town hall addressing the update, Twitch Chief Product Officer Mike Minton and Head of Community Mary Kish faced an audience of nearly 3,000 aggrieved users whose anti-AI messages flooded the live chat.
“Why is it not opt-in? That’s what everybody is spamming in chat. I get it. ‘Let me opt in versus making me opt out,’ … Well, there’s an honest answer… If this was opt-in, nobody would opt in. That’s honestly the answer.”
Mike Minton, Chief Product Officer at Twitch
Kish acknowledged that leadership expected a negative reception, stating flatly, I don’t expect anyone to react to this favorably.
Preexisting Data Usage and Legal Context
While the user-facing toggle is a recent addition, Ars Technica notes that the practice of utilizing Twitch content for artificial intelligence training is not entirely new. Back in 2024, an executive confirmed during an event held by publication The Information that Amazon was already leveraging Twitch content for training purposes.
During the recent town hall, Minton admitted he did not know whether Amazon had already ingested Twitch livestreams prior to the toggle’s rollout, noting he was not personally responsible for Amazon’s training efforts. Additionally, the policy explicitly states that the new opt-out mechanism is not retroactive, leaving untouched any historical content already incorporated into model development before a user flips the switch.
Wider Creator Backlash and Demands for Reform
Prominent streaming partners, including HasanAbi, Pokimane, Asmongold, and creator Legundo, voiced sharp criticism online, urging users to disable the setting and condemning the lack of financial compensation or affirmative consent.
Keep reading
