This Privacy Policy explains what personal data SG Tutor Bots collects, why we collect it, how we use it, and the rights you have over your information. It applies to all Telegram-based AI tutoring bots operated under the SG Tutor Bots project, starting with Ms. Lee, our O/N-level English tutor.
By completing the onboarding form for any of our bots, you (and where applicable, your parent or guardian) agree to the practices described in this policy. If you do not agree, please do not use the service.
SG Tutor Bots is an independent project providing free AI-powered tutoring for Singapore secondary school students through Telegram. Our mission is to make personalised, exam-calibrated tutoring available to every student regardless of family income.
For any privacy-related enquiries, please use the contact details in Section 11.
We collect information you voluntarily provide during onboarding, and message content you send to a bot during tutoring sessions. Our bots also build a learning memory over time to improve the quality of tutoring.
| Data | Why collected |
|---|---|
| Name | Personalise messages and build rapport |
| School level (Sec 1–4, N1, N2) | Pitch tutoring at the right exam standard |
| Weakest components (e.g. comprehension, summary writing) | Focus sessions on areas that need the most work |
| Preferred check-in time | Schedule daily motivational messages |
| Telegram User ID | Deliver messages via Telegram |
| Chat history (last 10 messages) | Maintain conversational context for the AI |
| Data | Why collected |
|---|---|
| Known weaknesses | Patterns extracted from past sessions (e.g. "text-lifting in summary") |
| Recurring mistakes | Specific errors seen repeatedly across sessions |
| Recent progress notes | Positive observations to track improvement over time |
| Session count and last session date | Track engagement and tailor returning-student greetings |
| Engagement trend | Help Ms. Lee adapt her tone to the student's motivation level |
Memory data is extracted automatically from conversation content every few sessions. It is used only to improve the tutoring experience for that individual student.
Your data is used exclusively to:
We do not use your data to train AI models, build advertising profiles, sell to third parties, or for any purpose beyond operating the tutoring bot you signed up for.
To improve tutoring quality for all students, we analyse anonymised, aggregated patterns across the student cohort — for example, identifying that a particular grammar mistake is common among Sec 3 students, or that a certain type of summary question causes widespread difficulty.
This cohort analysis uses only anonymised data — no individual student can be identified from it. Insights from this analysis are used solely to improve how the bots teach and what practice questions are generated. They are never shared externally.
Running these bots requires a small number of trusted third-party services. Each receives only the minimum data necessary for its function:
| Service | Purpose | Data shared |
|---|---|---|
| Telegram | Message delivery platform | Your Telegram User ID and message content |
| Anthropic (Claude API) | AI response generation | Your profile data and recent chat history, as needed to generate a tutoring response |
| Google Firebase | Database and cloud infrastructure | Your profile data and conversation history, stored in a private Firestore database in the asia-southeast1 region |
| Voyage AI | Semantic search for practice questions | The text of your messages (anonymised), used to find relevant practice questions. No personal profile data is shared. |
We do not share your data with any other third parties, including advertisers, schools, or government bodies.
Your profile data and session memory are stored for as long as you are an active user. If you request deletion (see Section 8), we will remove all your data within 14 days.
Rolling chat history is limited to the last 10 messages per session — older individual messages are automatically dropped from the AI context, though your learning profile and memory persist until you request deletion.
Your data is stored in Google Firebase (Firestore), which provides encryption at rest and in transit. The database is hosted in the asia-southeast1 (Singapore) region. Access is restricted to authenticated cloud functions only — no public read access is permitted.
Bot tokens and API keys are stored as Firebase environment secrets and are never exposed in client-side code. While we take reasonable precautions, no method of transmission over the internet is 100% secure.
You (and where applicable, your parent or guardian) have the right to:
To exercise any of these rights, contact us using the details in Section 11. We will respond within 14 days.
We recommend that a parent or guardian reviews this Privacy Policy before a student begins using any SG Tutor Bots service. By allowing a student to complete onboarding, the parent or guardian confirms they have read and agreed to this policy on the student's behalf.
We collect only the minimum data needed for effective tutoring. We do not collect sensitive personal information such as home addresses, NRIC numbers, school names, or financial details. No student data is ever used for advertising, profiling, or shared with third parties beyond those listed in Section 5.
Parents or guardians may request access to, correction of, or deletion of their child's data at any time by contacting us using the details in Section 11.
We may update this policy from time to time. When we do, we will update the "Last updated" date at the top of this page. If changes are material, we will notify active users via a Telegram message from the relevant bot.
Continued use of any bot after a policy update constitutes acceptance of the revised policy.
For privacy questions, data requests, or to report a concern:
SG Tutor Bots
Email: learningloons@gmail.com
We aim to respond to all privacy enquiries within 14 days.