Twitch's AI opt-out shipped after the training: consent as a retrofit
The toggle that stops Amazon training generative models on your streams is off by default, and Twitch's product chief says that is deliberate: if it were opt-in, nobody would opt in.
The Product Desk
What happened
- Twitch updated its account settings to let streamers opt out of having their content used to train the generative AI models of Twitch's parent company, Amazon.
- The setting is off by default, which means the AI training is on by default; creators are opted in unless they act.
- What the training covers is close to everything on a channel: streams, past broadcasts, clips, chat messages, and the images and text on a channel page.
- A Twitch support page states that if a streamer opts out, their streams, on-demand versions, clips, stream chats and other text and images on their channel will not be fed into Amazon's models in the future; Twitch says the purpose of these models is "to generate or synthesize text, audio, images or video."
- Twitch's own example is that a streamer's voice could train speech-to-text models, improving captions on Twitch and Amazon's AI products elsewhere.
Why it matters
Twitch has added an account setting that lets streamers stop Amazon from training generative AI models on their channel content, and the setting sits off by default, which means the training is on [1][2]. The toggle is new, the practice is not, and that ordering is the part operators should read closely [10]. The scope is broad: streams, past broadcasts, clips, chat messages, and the text and images on a channel page [3]. According to a Twitch support page cited by Engadget, opting out keeps those out of Amazon's models "in the future," with the models defined as ones whose purpose is "to generate or synthesize text, audio, images or video" [4]. Twitch's own worked example is a streamer's voice training speech-to-text, improving captions on Twitch and on Amazon AI products elsewhere [12]. The control stops at the channel boundary: messages you post in another streamer's chat are governed by that streamer's setting, not yours [13]. And it only stops training. AI keeps running for chat moderation, ad selection and recommendations [5], and Twitch says some safety functions are not individual settings because "turning those off for one person would weaken safety and security for everyone else on the platform" [6]. The reason for the default was stated out loud. On a livestream with head of community Mary Kish, chief product officer Mike Minton answered the question the chat was spamming: "If it was opt-in, nobody would opt in." [7] TechCrunch put the live audience at nearly 3,000, many of them posting anti-AI messages [8]. Minton added that he is comfortable with AI training and would not opt out himself [21]. TNW's read is that a default chosen because users would refuse it is the textbook definition of a dark pattern, and that the unusual thing here is an executive describing the reasoning without dressing it up [19]. Nobody at Twitch would say how long this has been running. Asked whether a user's videos had already been used, Minton said, "I don't actually know the answer to that question because I don't know what Amazon [...] has done in terms of model training and what they've used and not used." [9] Wired reports it remains unclear when the practice began [11]; TNW reports the platform has been feeding the pipeline for a couple of years [10]. Twitch's terms of service, in effect since March 2024, already granted Twitch and its sublicensees the right to use, reproduce, modify, adapt and create derivative works from user content, but did not explicitly cover generative AI training [22]. Because the opt-out only applies going forward and the company's own product chief cannot say what has already been ingested, the setting offers no remedy for anything already used [20]. The distribution of this news matters as much as its content. Twitch framed the change as adding a setting rather than announcing that training would begin [16]; the practice surfaced when streamer Zach Bussey found the toggle and a screenshot spread [15]; Wired counted more than 16,000 creators registering opposition in a forum [14]. Asked why Twitch had not emailed streamers, Kish said not everyone reads their emails [24]. Kish also framed the opt-out itself as a response to community sentiment [17], and Minton argued the broader practice is beyond Twitch's control, saying it is "quite reasonable to assume that almost any publicly available content is used to train models in one way or another, with or without permission" [23]. Three things to watch. Whether Twitch ever publishes a start date and a list of which Amazon models consumed which content, since the current answer is that its own executives do not know [9]. Whether the March 2024 terms get amended to name generative AI explicitly, which would tell you whether the toggle is a policy change or a UI change [22].
Clarity's read
What the record supports and how the coverage leans. The claims behind it follow.
Reality
- Evidence72
- Adoption
- Not scored
- Hype gap+15
- Incentives78
- Confidence70
Perspective Coverage
8 publishers- Builder
- Builder 24%
- Operator
- Operator 56%
- Investor
- Investor 20%
Claim ledger
Ranked by verification strength, evidence, and original report placement.
- [1]
Twitch updated its account settings to let streamers opt out of having their content used to train the generative AI models of Twitch's parent company, Amazon.
- [2]
The setting is off by default, which means the AI training is on by default; creators are opted in unless they act.
- [3]
What the training covers is close to everything on a channel: streams, past broadcasts, clips, chat messages, and the images and text on a channel page.
- [4]
A Twitch support page states that if a streamer opts out, their streams, on-demand versions, clips, stream chats and other text and images on their channel will not be fed into Amazon's models in the future; Twitch says the purpose of these models is "to generate or synthesize text, audio, images or video."
- [5]
Opting out only stops generative AI training; AI continues to run on Twitch for chat moderation, ad selection and recommendations, and the change does not touch that.
- [6]
Twitch said some AI functions help protect the whole community and that "turning those off for one person would weaken safety and security for everyone else on the platform, so they are not individual settings"; it also said AutoMod does not retain users' data or use it to generate material.
- [7]
Asked on a livestream why the setting is not opt-in, Twitch chief product officer Mike Minton said: "Why is it not opt-in? That's what everybody is spamming in chat. I get it... Well, there's an honest answer... If this was opt-in, nobody would opt in. That's honestly the answer."
- [8]
The livestream on the official Twitch channel featured Head of Community Mary Kish and Chief Product Officer Mike Minton addressing a live audience of nearly 3,000 aggrieved users, many posting anti-AI sentiments in chat.
- [9]
When a user asked whether their videos had already been used for training, Minton responded: "I don't actually know the answer to that question because I don't know what Amazon [...] has done in terms of model training and what they've used and not used."
- [10]
TNW reports that the toggle is new but the practice is not, and that the platform is reported to have been feeding the training pipeline for a couple of years.
- [11]
It remains unclear exactly when Twitch began using the posts, streams and videos of its users to train Amazon's systems.
- [12]
Twitch's own example is that a streamer's voice could train speech-to-text models, improving captions on Twitch and Amazon's AI products elsewhere.
- [13]
Opting out protects content on your own channel, but messages you send in another streamer's chat can still be collected if that streamer has not opted out; the other streamer's settings govern.
- [14]
In a forum dedicated to the topic, more than 16,000 creators expressed opposition to having their content used by default to train Amazon's AI systems, a practice that only came to light after the settings update.
- [15]
Minton made the comments on a livestream this week alongside Mary Kish, after streamer Zach Bussey found the toggle and a screenshot of it spread.
- [16]
Rather than telling the community that Amazon would begin training on user content, Twitch framed the change as "[adding] a setting that lets you opt out of having your channel content used to train generative AI content models across Amazon."
- [17]
Kish said that including an opt-out option at all is "a reflection of reacting to this community's voice that you are not wanting to train Gen AI models, and we want to give you that option."
- [18]
Meta uses public content from Facebook and Instagram to train its own AI models; users in the UK can opt out, while elsewhere the only way to opt out is to use private settings, which is not feasible for creators who monetize their accounts.
- [19]
TNW argues that a default chosen because users would refuse it is the textbook definition of a dark pattern, and that what is unusual is hearing an executive describe the reasoning without dressing it up.
- [20]
Because the opt-out applies only to future use while Twitch's chief product officer says he does not know what Amazon has already used for model training, the setting provides no remedy for content already ingested.
- [21]
Minton said he is comfortable with AI training and would not opt out himself if he were a streamer.
- [22]
Twitch's Terms of Service, in effect since March 2024, stipulate that users grant Twitch and its sublicensees the right to use, reproduce, modify, adapt, distribute and create derivative works from their content, but until now have not explicitly stated that such materials could be used to train generative AI models.
- [23]
Minton said the mechanisms are not unique to Twitch and that other AI developers likely extract content from Twitch and other services: "I don't know for sure, but I think it's quite reasonable to assume that almost any publicly available content is used to train models in one way or another, with or without permission. So I think we also need to acknowledge that there's a lot here that's beyond even our direct control."
- [24]
Asked on the same stream why Twitch had not simply emailed people about the change, Kish said not everyone reads their emails.
- [25]
TNW notes that none of the coverage says how the Twitch default works in Europe, and that opt-out consent for AI training has been challenged there before: Meta had to pause an identical approach in 2024 before regulators let it resume.
Sources
8 independent publishers whose own reporting we read for this story.
- arstechnica.comTwitch content has trained Amazon AI for years, but users can opt out now
1 article · August 12, 2026
- dexerto.comTwitch addresses backlash after streamers find AI is being trained with their content
1 article · August 13, 2026
- engadget.comTwitch streamers can now refuse to let Amazon train its genAI models on their content
1 article · August 12, 2026
- gizmodo.comTwitch Adds Setting Letting Users Opt Out of AI Training Amid User Backlash
1 article · August 13, 2026
- techcrunch.comAmazon will train on Twitch streamers’ content by default, unless they opt out
1 article · August 12, 2026
- thenextweb.comTwitch will train Amazon’s AI on your streams unless you find the setting
1 article · August 13, 2026
- theverge.comTwitch streamers can now opt out from training Amazon’s AI
1 article · August 12, 2026
- wired.comAmazon Can Use Your Twitch Content to Train Its AI—Unless You Opt Out
1 article · August 15, 2026
Topics and entities
Topics
- Default consent settings
- AI training data
- Live Streaming Platforms