Twitch Faces Backlash Over Amazon AI Training

Aug 14, 2026

Twitch is facing a wave of criticism from its community after introducing a setting that allows Amazon to use Twitch channel content to train generative AI models — with the option enabled by default.

The change surfaced on August 12 and quickly drew attention from streamers and viewers concerned about how their broadcasts, clips, chat activity, and other channel content could be used for AI development. Twitch is owned by Amazon, which acquired the livestreaming platform in 2014.

Creators aren’t locked into the program, however. Twitch has provided an option to disable generative AI training through its privacy settings. The controversy largely centers on the company’s decision to make participation opt-out rather than opt-in.

What Is Twitch Allowing Amazon to Use?

The new setting is labeled “Training for Generative AI.”

According to reporting on the policy, leaving the setting enabled allows channel content to be used to train Amazon generative AI models capable of generating or synthesizing media such as text, images, audio, and video.

The content potentially covered includes material associated with Twitch channels such as streams, VODs, clips, chat and other channel content.

That makes the issue particularly relevant to the gaming community. Twitch streams frequently contain hours of gameplay alongside streamer commentary, webcam footage and interactions with viewers.

Why Are Twitch Streamers Angry?

Much of the backlash isn’t simply about Amazon developing AI. It’s about consent.

Instead of asking creators whether they want to participate, Twitch introduced the control with AI training enabled by default. Anyone who doesn’t want eligible channel content used for this purpose has to manually disable the setting.

The reasoning behind that decision attracted even more attention after Twitch Chief Product Officer Mike Minton discussed it during a livestream.

When addressing why Twitch didn’t simply make the program opt-in, Minton said:

“If it was opt in, nobody would opt in.”

That comment has become one of the central talking points surrounding the controversy, particularly among creators who believe participation in AI training should require affirmative consent.

Community reaction has been substantial. A discussion about the setting on Reddit’s Twitch community accumulated thousands of upvotes, with many commenters objecting specifically to the default-on approach and sharing instructions for disabling it.

How to Opt Out of Amazon AI Training on Twitch

Twitch users who don’t want their channel content used to train Amazon’s generative AI models can disable the setting.

On Twitch:

Open Settings.
Select Security and Privacy.
Scroll down to the Privacy section.
Find Training for Generative AI.
Turn the setting off.

The setting has been reported as easier to locate through Twitch’s desktop website. Users may also want to verify afterward that their preference has been saved.

Turning It Off Doesn’t Stop Every Use of AI

There is an important distinction creators should understand.

Disabling Training for Generative AI prevents channel content from being used for the specific purpose of training Amazon’s generative AI models covered by the setting. It does not opt users out of every AI-related feature or every other use of their content described in Twitch’s Privacy Notice.

Twitch and Amazon can still use channel content for other platform purposes, including AI-supported Twitch features involving recommendations, streamer growth and monetization tools, and community safety systems such as AutoMod.

In other words, switching off the setting is specifically about generative AI model training, not a universal “no AI” option for Twitch.

There’s Another Complication for Twitch Chatters

The policy also raises an interesting issue for people who don’t stream themselves but regularly participate in Twitch chats.

According to reporting on Twitch’s policy, interactions on another creator’s channel are governed by that channel’s AI-training preference. That means opting your own channel out doesn’t necessarily control how an interaction you make elsewhere on Twitch is handled.

For a platform built heavily around interaction between streamers and their communities, that distinction could prove especially controversial.

The Gaming Industry’s AI Debate Isn’t Going Away

Twitch’s decision arrives amid a much larger debate over generative AI throughout gaming and other creative industries.

Game developers, artists, voice actors, writers and content creators have increasingly questioned where AI training data originates, whether creators should be compensated and whether companies should be required to receive explicit permission before using creative work for model training.

Twitch adds another dimension to that conversation because livestreams can combine several different types of material simultaneously.

A single gaming broadcast could include a streamer’s voice and appearance, viewer conversations, original graphics and overlays, gameplay footage, music and intellectual property belonging to a game publisher.

That makes questions surrounding exactly what content is being used — and under what rights — considerably more complicated than simply feeding plain text into an AI model.

Twitch’s Opt-Out Approach May Be the Bigger Story

For some users, the existence of AI training itself will be enough to object to the policy.

But Twitch’s decision to enable it automatically may ultimately be what keeps this controversy alive.

Giving creators a choice theoretically provides users with control. Making the choice opt-out, however, means users who never discover the setting could remain enrolled without actively deciding that they want their content used for generative AI training.

And Minton’s acknowledgment that an opt-in system could result in very few people volunteering is unlikely to quiet those concerns.

For Twitch creators who don’t want their channel content contributing to Amazon’s generative AI models, the immediate solution is straightforward: check your Twitch Security and Privacy settings and disable “Training for Generative AI.”

The broader argument over whether creators should have needed to do that in the first place is likely far from over.