Does Regie.ai train on your data?
Regie.ai's policy said it did not use any data to train generalised AI models. The policy rewritten on 4 August 2026 dropped that section, and says nothing about training at all.
Graded B from 14 August 2026 on one sentence: "Regie.ai does not use any data, including data obtained through Google Workspace APIs, to train generalized AI or machine learning models." That was section 15 of a policy dated 6 March 2025, and Regie was written to about it on 27 August. It hasn't replied. The policy Regie serves now is dated 4 August 2026 and is a rewrite, 88 per cent of its paragraphs changed. Section 15 isn't in it. No form of "train" is, and no "machine learning". AI appears only where the service does something to you: a cloned voice made from an enrolled speaker's recording, and call audio and transcripts that "may be processed by automated and AI systems". Whether any of that data trains a model is no longer stated either way. The monitor's shrink guard rejected the shorter page for five weeks as a bad fetch, so the entry carried the old sentence until 23 September 2026. A commitment withdrawn is a finding about the document, and this entry records it: UNCLEAR now, with the sentence it used to rest on kept here.
- Regie.ai privacy policy dated 4 August 2026, captured 2026-09-23, 81 paragraphs (https://www.regie.ai/privacy-policy) · The previous policy dated 6 March 2025, captured 2026-08-14, 125 paragraphs, whose section 15 carried the training exclusion this entry was graded on
- train, training, trains, fine-tune, machine learning, artificial intelligence, AI model, large language model
- 0 of 0 raw hits
- 23 Sept 2026
None of the searched terms appears in the rewritten policy. "AI" on its own appears 6 times, about cloned voices and about call audio and transcripts being "processed by automated and AI systems", which is use, not training. Counted so the zero above is checkable.
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 Regie.ai training
Plan by plan, from what Regie.ai'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 plans
Not graded, so no route is pointed at. Graded B from 14 August 2026 on one sentence: "Regie.ai does not use any data, including data obtained through Google Workspace APIs, to train generalized AI or machine learning models." That was section 15 of a policy dated 6 March 2025, and Regie was written to about it on 27 August. It hasn't replied. The policy Regie serves now is dated 4 August 2026 and is a rewrite, 88 per cent of its paragraphs changed. Section 15 isn't in it. No form of "train" is, and no "machine learning". AI appears only where the service does something to you: a cloned voice made from an enrolled speaker's recording, and call audio and transcripts that "may be processed by automated and AI systems". Whether any of that data trains a model is no longer stated either way. The monitor's shrink guard rejected the shorter page for five weeks as a bad fetch, so the entry carried the old sentence until 23 September 2026. A commitment withdrawn is a finding about the document, and this entry records it: UNCLEAR now, with the sentence it used to rest on kept here.
What this policy has done since we started watching
1 change detected since 14 Aug 2026. Each one is a diff between two archived copies, and each links to the passage that moved.
Everything we read
Vendor response
We wrote to Regie.ai on 27 Aug 2026, quoting the clauses above and inviting a correction. Nothing has come back. That's published because a company declining to comment on its own policy is something a reader should know, and because the invitation should be on the record whether or not it's taken up. Any reply goes here, word for word.
Report a wrong clause