Privacy Policy

This site has no accounts and no forms that store what you type. This site does not use analytics cookies or advertising trackers. Search on this site runs in your browser, and what you type is not sent anywhere.

Faculty Twin is a course assistant for students in my courses. It collects:

• Questions. Each question is logged so I can improve the assistant. Email addresses, numbers and recognizable names are removed before it is stored. No account, name, email address or student ID is collected.
• A session cookie with a random id, so the page does not ask for the course passcode again. Rate limits use a hashed form of it.
• Usage counts such as tokens used and voice characters spoken, with no question text and no ids.

To answer a question, its text is sent to the AI providers that power the assistant: the narration model (Anthropic, OpenAI or OpenRouter), Voyage AI for search, and ElevenLabs or Microsoft for the voice. It is hosted on Vercel and stored in Supabase. Question logs can be deleted on request.

Faculty Twin alerts, from Prof Collier AI, send a text message to one person only: me, Ben Collier, the account owner, when a student reports a technical problem such as an API key out of credit, a broken submission or a broken quiz. The only phone number used is my own, which I opted in myself by texting START (see how to opt in). Students' phone numbers are never collected and students never receive texts. The texts are sent through Twilio.

We do not sell or share your SMS opt-in data or personal information with third parties for marketing purposes. SMS opt-in data and consent are not shared with any third party, except the providers that deliver the messages.

Reply STOP to any alert to stop SMS messages. For any privacy question, see the contact page.