Does Ring train on your data?
Ring's notice never says "train". Its US render links a disclosure: generative models trained on datasets that "in some cases" hold personal information. Whether your video is in them, neither says.
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.
- Ring privacy notice, captured 2026-08-31, UK-served render (https://ring.com/privacy-notice) · Ring privacy notice, captured 2026-09-07, US-served render, which links the disclosure (https://ring.com/privacy-notice) · Ring Generative AI Development Disclosure, Internet Archive copy of 2026-06-23; the live page answers 404 outside the US (https://ring.com/gen-ai-disclosure)
- train, training, trains, fine-tune, machine learning, artificial intelligence, AI model, large language model
- 10 of 11 raw hits
- 16 Sept 2026
Ten matches, all in the disclosure: what the generative AI services are trained and tested on, and the safeguards around that training. None says whether the video and audio Ring holds about its own customers is in those datasets, which is the question this entry asks, so they do not resolve it. The one hit in the notice itself is hardware: "your Ring product's model, serial number and software version".
What each plan can refuse
The same questions, asked of every plan we could identify. Every cell comes from a quoted clause or a documented search - see the evidence above.
- Not established
- Not stated
- —
- —
- Not established
How to opt out of Ring AI training
Plan by plan, from what Ring's own policy says about refusing. Where it publishes the steps, they're listed here with the limits they carry. Where there's nothing to switch off, or nothing graded, this says so instead of inventing a menu path.
All accounts
Not graded, so no route is pointed at. 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.
What this policy has done since we started watching
No change detected. We've held a copy of Ring's documents since 23 Jun 2026 and re-read them since, across 5 captures of 2 documents. The text hasn't moved in a way that touches this verdict. That's a fact about the documents, and it's only knowable by having kept the earlier copies.
Vendor response
We haven't yet written to Ring about this verdict, and they haven't disputed, clarified or announced a change to any clause quoted above. When either happens it's published here, word for word.
Report a wrong clause