Does DeepL train on your data?
DeepL trains its neural networks on what free users type. Pro subscribers are told their texts "will not be used to improve the quality of our services". The difference is the price.
Opt-out is sold, not offered
"We process the content you upload and their translations or improvements for a limited period of time to train and improve our neural networks and algorithms."
- 1 Oct 2026
- 5 Oct 2026
- DeepL Translator / Write (free version)
- eu
- Fetched from the company
- 631CE232…E2AD
How we read it: The policy splits by section, and the split is the finding. Section 3 governs the free version and says the content you upload, its translations, and even the corrections you make to DeepL's suggestions are processed to train its neural networks. Section 4 governs Pro and says the opposite. There is no setting in between: the way to stop training on a free account is to stop being on a free account. That is what this site grades E rather than F — the control exists, it is real, and it is sold.
"When using these services, your texts will not be used to improve the quality of our services."
- 1 Oct 2026
- 5 Oct 2026
- DeepL Translator Pro / API Pro / Write Pro
- eu
- Fetched from the company
- 631CE232…E2AD
How we read it: Unhedged, and paired with a retention commitment: submitted texts "will not be permanently stored" and are deleted after the service is performed. Read against section 3, this is the same company describing two different products, and the difference between them is a subscription.
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.
- Paid upgrade required
- Not stated
- 75 / 100
- Not stated
- Unknown
- Nothing to switch off
- Not stated
- 19 / 100
- Not stated
- Unknown
How to opt out of DeepL AI training
Plan by plan, from what DeepL'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.
DeepL Translator / Write (free version)
Only by moving to a higher-priced plan. DeepL puts the opt-out behind a paid tier.
DeepL Translator Pro / API Pro / Write Pro
Nothing to switch off: DeepL's policy says training is off here.
What this policy has done since we started watching
2 changes detected since 30 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 DeepL on 1 Sept 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