Does Blink train on your data?
Blink stores video and audio from cameras in your home. Its policy never uses "train", "AI" or "machine learning" — whether that footage builds anything, the document does not say.
This clause is real, but it doesn't resolve the training question - see the finding below.
"We collect your personal information in order to provide and continually improve the products and services of Blink and its affiliates."
- 31 Aug 2026
- 5 Oct 2026
- All accounts
- us
- Fetched from the company
- 2759F8A2…7549
Why this is still UNCLEAR: Zero forms of "train", "machine learning", "artificial intelligence", "AI", "model" or "algorithm" anywhere in the capture. The nearest the document comes is the clause quoted below — improvement running to "Blink and its affiliates", and Blink's parent is Amazon — but where the policy explains what improvement means, it means analysing performance, fixing errors and usability: engineering, not model-building. The document says Blink holds "images, audio, and video stored in connection with Blink Services" and says nothing about whether a person may ever view them or whether anything is built from them. "Biometric" never appears, from a company selling person detection. This is the same silence the site recorded for Amazon itself, repeated over more intimate data. The capture is the US policy; separate country policies exist at their own URLs and are not covered by this entry.
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 Blink AI training
Plan by plan, from what Blink'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. Zero forms of "train", "machine learning", "artificial intelligence", "AI", "model" or "algorithm" anywhere in the capture. The nearest the document comes is the clause quoted below — improvement running to "Blink and its affiliates", and Blink's parent is Amazon — but where the policy explains what improvement means, it means analysing performance, fixing errors and usability: engineering, not model-building. The document says Blink holds "images, audio, and video stored in connection with Blink Services" and says nothing about whether a person may ever view them or whether anything is built from them. "Biometric" never appears, from a company selling person detection. This is the same silence the site recorded for Amazon itself, repeated over more intimate data. The capture is the US policy; separate country policies exist at their own URLs and are not covered by this entry.
What this policy has done since we started watching
The record starts here. Blink's policy was first captured on 31 Aug 2026, and there's no earlier copy to compare it against, so this verdict has no history behind it yet. From now on every re-read is diffed against that copy, and anything that moves appears here.
Everything we read
Vendor response
We haven't yet written to Blink 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