When you use experimental Labs Models, your data and outputs are used for AI training by default, and any opt-out settings you configured for other Mistral AI products do not carry over. You must separately activate zero data retention within Labs Models to prevent this.
This analysis describes what Mistral AI's agreement states, permits, or reserves. It does not constitute a legal determination about enforceability. Regulatory applicability and practical outcomes may vary by jurisdiction, enforcement context, and individual circumstances. Read our methodology
This provision creates a product-siloed consent architecture where a business that has opted out of training for its primary Mistral AI products may unknowingly contribute training data through Labs Models unless it takes a separate, affirmative action per product.
The updated terms redefine when Mistral AI will use customer data and outputs for AI model training. Previously, the terms distinguished between free tiers (where training was opt-out) and Labs Models (where training always occurred). The revised language consolidates experimental models as 'Labs or Preview Models' and explicitly states that training opt-out preferences applied to other products do not carry over to these experimental models. Additionally, the terms now specify that customer data includes 'access to data' and 'coding environment,' expanding what qualifies as trainable material. The liability cap was narrowed to remove the carve-out protecting confidentiality breach claims, meaning liability limits apply more broadly. You can review which Mistral products are set to opt-in versus opt-out by default and avoid Labs or Preview Models if you do not want your data used for training.
View change record →Organizations that use Labs Models for testing experimental AI capabilities may have their Customer Data and Outputs used for model training even if they have opted out of training elsewhere, because the terms state that opt-out preferences from other products do not apply to Labs Models.
How other platforms handle this
The Gemini Apps Privacy Notice does not apply, and your Opal data, including feedback you provide, is processed per the Google Privacy Policy.
we may share data between our affiliates for the safety and security of our users and may take necessary actions if we believe you have violated these Terms, including banning you from our Services and/or our affiliates' services...
Whenever we transfer personal data internationally, we use tools and transfer agreements to: make sure the data transfer complies with applicable law; and help to give your data the same level of protection as it has in the EU...
"Mistral AI may provide its commercial Customers access to experimental or pre-release models identifiable by the prefix 'labs' in AI Studio ('Labs Models') for testing purposes. Labs Models are exclusively available to Mistral AI's commercial Customers and are restricted to professional use only. By using Labs Models, you acknowledge that (i) Mistral AI may use Customer Data and Outputs generated from Labs Models to train its artificial intelligence models, unless you have activated zero data retention, and (ii) the opt-out preferences you selected for other Mistral AI Products does not apply to Labs Models. If you do not want your data or outputs used for training, do not use Labs Models.Excerpt from Mistral AI's Commercial Terms
(1) REGULATORY LANDSCAPE: This provision engages GDPR requirements on granular, specific consent and transparency, particularly where different product contexts have different data processing implications.
Enforcement risk, jurisdiction flags, contract triggers, and due diligence action items.
ConductAtlas detected a major restructuring of Meta’s privacy policy that removed detailed consumer rights disclosures and relocated them to separate documents.
Your genetic data may be transferred to a new owner as a business asset. Here is what the Terms of Service actually say and what you can do right now.
Get the research letter
Companies change their terms quietly. We read every version and catch what actually changed. One email a week on the changes that matter and what they mean.
This provision creates a product-siloed consent architecture where a business that has opted out of training for its primary Mistral AI products may unknowingly contribute training data through Labs Models unless it takes a separate, affirmative action per product.
Organizations that use Labs Models for testing experimental AI capabilities may have their Customer Data and Outputs used for model training even if they have opted out of training elsewhere, because the terms state that opt-out preferences from other products do not apply to Labs Models.
ConductAtlas has identified this type of provision across 288 platforms. See the full comparison.
No. ConductAtlas is an independent monitoring service. We are not affiliated with, endorsed by, or sponsored by Mistral AI.