How to stop Ring training AI on your data

All accounts

No form of "train", no "machine learning" and no "artificial intelligence" appears anywhere in Ring's privacy notice; the only "AI" is in the US-served render, in the title of a link to a "Generative AI Development Disclosure". What the document does say is what Ring holds: it processes and stores "video or audio recordings, live video or audio streams, images, comments, and data our products collect from their surrounding environment", with the stated purpose "To provide you with these services" — and no purpose beyond improve-our-services boilerplate is stated for any of it. "Biometric", "voice" and "recognition" never appear either; the one facial-features sentence is about the camera owner's obligations to visitors, not about anything Ring does with faces. The notice read first was the UK-served render of 31 August, with a GDPR legal-bases section naming "legitimate business interests" as the basis for improving Ring's products; the US render of 7 September adds the link. The disclosure itself answers 404 from outside the United States, so it was read from the Internet Archive's copy of 23 June 2026. It is Amazon's template: for its generative AI services Ring says "we train and test on a range of data intended to enhance our services' capabilities" — "licensed and proprietary datasets, synthetic datasets, open-source datasets, and publicly available content" that "may contain public domain content, rights-protected material, and in some cases, personal information or aggregate consumer information". Doorbell video is not named. Whether it is among the proprietary data is the one thing neither document says, so the finding stands. Ring is Amazon's; Amazon's own consumer notice, graded separately, reads the same disclosure and stands in the same place.