DeepL — privacy & data policy

Provider: DeepL. PromptPrivacy score: 79/100. Last verified August 26, 2026.

What happens to your data?

If you use the free tier, your submitted texts and corrections are temporarily stored and used by DeepL to train and refine its AI translation models. If you use a paid DeepL Pro plan, your content is encrypted, processed only to provide the translation, and immediately deleted without being saved or used for AI training. DeepL also collects account, billing, and diagnostic metadata to manage subscriptions and prevent fraud.

Does it train on your data?

Texts, documents, and corrections entered into the free versions of DeepL Translator and DeepL Write are stored and processed for a limited time to train and improve neural networks and algorithms. For paid DeepL Pro subscribers, submitted content is never used to train or improve models.

Can you opt out?

You cannot opt out of AI training while using the free version of DeepL; avoiding training requires using a paid DeepL Pro plan. Users can opt out of non-essential tracking cookies via the website cookie manager.

Data retention

For paid DeepL Pro tiers (Translator Pro, API Pro, Write Pro, and Voice), submitted text, documents, audio, and their translations are only kept temporarily during processing and are deleted immediately upon completion. Technical metadata associated with document translations is retained for 14 days for error analysis before automatic deletion. Free tier inputs are kept for a limited period for algorithm training, while customer service call recordings are retained for two years.

Encryption & security

Data transmission between the user's client device and DeepL's servers is encrypted in transit, preventing intermediary networks such as Cloudflare from accessing customer translation content. Specific encryption mechanisms at rest or security certifications are not specified in the text.

Is it safe for work?

Yes, but ONLY with a paid DeepL Pro subscription. DeepL Pro provides contractual no-training guarantees, immediate deletion of submitted documents and texts, Single Sign-On (SSO), and GDPR Article 28 Data Processing Agreements (DPAs). The free tier is NOT safe for confidential workplace data, as DeepL's terms explicitly prohibit translating personal data on the free tier and use free inputs for model training.

Source policy: https://www.deepl.com/en/privacy