AI Disclosure

Models we use

We use Google Gemini 2.5 Flash for live assistant chat, and Gemini 2.5 Pro for deeper analysis when needed. All requests are routed via Lovable's AI gateway — your data never trains third-party models.

What we send

Only the structured summary needed to answer your question: detected subscription names, amounts, billing cycles, and your own message text. We never send raw email contents, headers, or attachments to AI models.

Retention

AI request inputs are not retained by model providers under our gateway terms. Assistant conversation history is stored in your account so you can review it, and you can delete it at any time from Settings.

Your controls

• Disable AI insights from Settings → Notifications.
• Clear assistant history from /assistant.
• Export or delete all data from /account/export and /account/delete.