AI Transparency
“What Milo receives, what we do with it, and what we don't.”
WHAT MILO DOES
Milo is designed to guide your child rather than hand over the solution: he asks questions, recalls the concept, gives hints, and turns the lesson into flashcards.
WHAT MILO RECEIVES
Only what your child sends: the text of their messages, the photo of their homework and, if they use the microphone, their voice. Each sending is triggered by the child, and before the first one a screen names this data and its recipient, with the option to refuse.
WHAT WE DON'T DO WITH IT
No selling. No targeted advertising. No advertising profile of your child.
WHERE THE DATA GOES
To Google Vertex AI (Gemini models), in a European region, as a processor (GDPR article 28). Any transfers outside the EU are covered by Standard Contractual Clauses and the Data Privacy Framework.
HOW LONG
Homework photos: deleted within 30 days. Audio recordings: deleted after transcription. Learning data: 24 months. Technical logs: 12 months. Details are in the privacy policy.
MODEL TRAINING
No. What your child writes, photographs or says is not used to train AI models, neither ours nor Google's.
THE LIMITS
Milo can make mistakes, especially on very specialised exercises. That's why he makes your child reason instead of dictating an answer. He does not replace a teacher.
YOU STAY IN CONTROL
The account belongs to the parent, your child signs in with a PIN you choose, and you're alerted if a conversation goes off track. To access or delete your data: privacy@mesdevoirsfaits.app.
Last updated: October 4, 2026