Skip to content
logo The magazine for digital lifestyle and entertainment
Amazon Artificial intelligence News Streaming All topics
New Attitude

Twitch Users Must Act–or Amazon’s AI Will Step In

A smartphone in landscape mode against an abstract background in pink, purple, and blue. The screen displays the Twitch logo in violet on a white background. The back features curved neon lines and gradients in various shades of violet, pink, and blue.
The use of stream content for AI models is enabled by default on Twitch. Photo: Getty Images
Share article

August 16, 2026, 6:14 am | Read time: 3 minutes

Twitch has introduced a new setting that many streamers might not like. The platform allows its parent company, Amazon, to use content from channels to train generative AI models. This change is already sparking discussions because the corresponding permission is enabled by default. Those who do not want their content used for AI purposes must take action now.

The platform justifies the change by stating that channel content can contribute to improving generative AI systems. Various components of a channel are affected, including livestreams, video recordings, clips, chat messages, as well as images and texts.

The Real Issue Is the Default Setting

As reported by online magazine “TechCrunch,” the focus of criticism is not the use of AI, but the way it is implemented. Many streamers are particularly bothered by the fact that Twitch has automatically activated the permission. Users must actively opt out if they do not want their content to be used.

Many users would have preferred an opt-in model. In this case, content could only be used after explicit consent. This is exactly what numerous creators are now demanding. Just a few hours after the announcement of the new feature, over 15,000 users voted for a change on the Twitch support page.

One user writes: “Creative content or performance media should not be used for training generative AI without the explicit consent of the author or artist by default. That is simply theft of others’ works.” Many also fear that numerous users might not even notice the change and thus participate unintentionally.

Additional frustration was caused by a statement from Twitch’s product chief, Mike Minton. When asked why Twitch did not implement an opt-in, he replied that with voluntary consent, hardly anyone would participate. This openness was poorly received by the community.

More on the topic

How to Disable the Feature

Users who do not want their content to be used for training generative AI models can disable the feature in their account settings. On the Twitch website, the option is found under “Settings” and “Security and Privacy.” There, you will find the “Training for Generative AI” section.

In the app, navigate through “Profile,” “Settings,” and “Security and Privacy.” According to Twitch, it is sufficient to disable the setting once. The change then applies to the entire account and does not need to be made separately on each device.

Also of interest: Chinese AI also bypasses security tests

But beware, there are limits. According to Twitch, opting out of the use for generative AI only prevents the affected content from being used for future AI models. Other systems based on machine learning remain unaffected. These include, for example, functions for moderation, recommendations, or automatic subtitles.

An Important Question Remains Unanswered

For some streamers, another point might be even more significant: It is currently unclear whether and to what extent Twitch content has already been used for AI projects in the past.

According to the tech magazine “arstechnica,” there have been indications for years that Amazon has been using Twitch data for the development of AI systems. However, the company has not yet published a detailed breakdown of which information was used and since when this has been happening.

This is precisely why part of the criticism is not only directed at the new setting. Many users mainly criticize the lack of transparency. The current debate thus revolves around a fundamental question: How clearly must platforms communicate when their users’ content is incorporated into AI systems? At Twitch, this discussion has just begun.

This article is a machine translation of the original German version of TECHBOOK and has been reviewed for accuracy and quality by a native speaker. For feedback, please contact us at info@techbook.de.

You have successfully withdrawn your consent to the processing of personal data through tracking and advertising when using this website. You can now consent to data processing again or object to legitimate interests.