Twitch recently sparked a quiet revolution by letting creators opt out of having their streams, VODs, and chats used to train AI. While users were initially upset to learn their content was being scraped in the first place, Twitch's simple opt-out toggle actually makes it an outlier.
Most major platforms are quietly feeding your digital life to generative AI models—and stopping them is incredibly difficult.
The Platforms Allowing You to Opt Out
A few platforms give you direct control over your data, though the settings are often buried deep within your privacy menus:
- Twitch & LinkedIn: Both companies have active opt-out toggles. LinkedIn (owned by Microsoft) trains on your resume, posts, and comments by default, but you can turn this off under Settings & Privacy > Data Privacy > Data for Generative AI Improvement.
- X (formerly Twitter): Elon Musk's platform trains its Grok AI on your public interactions. To stop it, head to Settings > Privacy & Safety > Grok and uncheck the data-sharing box.
- YouTube: While you cannot stop Google from training its own models on your uploaded videos, creators can block third-party AI scraping in the YouTube Creator Studio under Channel > Advanced Settings > Third-party training.
Where Opting Out is Nearly Impossible
For other dominant platforms, your consent is treated as a given the moment you log in.
- Meta (Facebook, Instagram, Threads): Meta uses almost everything—posts, photos, and public interactions—to train its Llama models. Unless you live in the EU (where local laws force Meta to comply), you cannot easily opt out. Your only real defense is setting your account to private or deleting it entirely.
- TikTok: ByteDance's privacy policies are notoriously complex. Reports suggest TikTok even trains on private videos and unsaved drafts, with no simple toggle to stop it.
- Reddit: Reddit has lucrative data-licensing deals with Google and OpenAI. There is currently no user setting to opt out of this training, and third-party web crawlers frequently scrape the public forums anyway.
How to Protect Your Privacy
If you want to limit how tech companies use your content, audit your account privacy settings immediately. Toggle off AI sharing on LinkedIn, X, and Twitch. For Meta platforms and TikTok, consider switching your accounts to private. Ultimately, the most foolproof way to keep your data out of AI training models is to avoid posting it publicly altogether.
