Twitch has introduced an option allowing users to stop Amazon from using their streams, clips, images, chat logs and other channel content to train its generative artificial intelligence models, the livestreaming platform announced Wednesday. The move followed widespread backlash after Twitch confirmed that Amazon, which has owned the platform since 2014, had been collecting user content for AI training by default.

A Default Setting Users Can Now Disable

Twitch said the AI training collection setting had been switched on automatically for users, mirroring the approach taken by most major content platforms, but that the company would honor requests from users who wish to opt out. Data collected from users who remain opted in could include any content from their channel, and the company said that data would not be resold to other companies.

Twitch’s chief product officer, Mike Minton, addressed the default setting directly during a livestream on the platform, acknowledging that few users would choose to participate if given the option upfront. “If it’s opt-in, nobody would opt-in. That’s the honest answer,” Minton said.

A Tutorial Delivered With Little Enthusiasm

During the same broadcast, Twitch’s head of community, Mary Kish, walked viewers through the steps needed to disable generative AI training in their channel settings. Kish characterized the practice of collecting user data for AI training as now standard across the industry, while acknowledging the unpopularity of the policy among Twitch’s user base. “We don’t expect you to be happy or excited about this. I don’t expect anyone to react to this favourably,” she said.

The livestream’s chat function was flooded with critical comments as the announcement unfolded. One viewer wrote that “nobody would opt in because nobody wants to feed AI with our creativity and content,” while another suggested Twitch had missed a chance to set a different industry standard by resisting the practice altogether rather than merely offering an opt-out.

Unanswered Questions on Data Use

Minton said during the stream that he did not have clarity on whether user data had already been used to train AI models before Wednesday’s announcement, telling viewers he was uncertain what Amazon “has done in terms of model training and what they’ve used and not used.” The exact date Amazon began collecting Twitch user data for AI training purposes has not been disclosed.

Twitch’s Role Within Amazon’s Broader AI Push

Twitch is one of the largest livestreaming platforms globally, with a user base concentrated among gamers and younger audiences, and its most prominent creators command tens of millions of followers. Amazon, which operates a range of AI-related services and products, has invested heavily in AI development as it competes with rivals including Google and Meta for position in the sector.