Mistral AI faces CNIL complaint over Le Chat free tier data practices
A French lawyer filed a complaint with the CNIL against Mistral AI, alleging that the free version of its Le Chat chatbot collects personal data from user queries to train AI models by default, and requires paid subscriptions for users to opt out. Mistral AI states it complies with GDPR and that users can opt out by email.
What happened
Mistral AI launched a mobile version of its conversational agent Le Chat in early February 2025. Lawyer Me Jérémy Roche filed a complaint with France's CNIL against Mistral AI, alleging that the free tier of Le Chat collects user data by default to train AI models, and users must pay for a subscription to opt out of data collection.
Reported impact
The complaint alleges that Mistral AI's practice of requiring payment to disable data collection for model training violates Article 12 of the GDPR, which prohibits companies from charging users to refuse personal data sharing.
Index analysis
Mistral AI's privacy policy states that data from user queries is used to 'develop and train our general-purpose AI models.' Users of the free tier can only disable data collection by subscribing to the paid 'Pro' or 'Team' plans, which automatically deactivate machine learning from user data.
Governance controls
Mistral AI stated that users can opt out of data collection by sending an email to privacy@mistral.ai, and that the company has 'always allowed users to refuse the use of information contained in requests sent to the Chat.' The company says it is in compliance with GDPR regulations.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.