Privacy Policy

1. Who this policy is from

This Privacy Policy explains how TeachBoard (an Indian sole-proprietorship, the "Data Fiduciary" / "Controller", referred to here as "we", "us") handles personal data when you use teachboard.app and related services (the "Service"). It applies to all users worldwide.

For the purposes of the EU/UK GDPR we are the controller of your personal data. For the purposes of India's Digital Personal Data Protection Act, 2023 ("DPDP Act") we are the Data Fiduciary.

2. Data we collect

2.1 If you use the board anonymously

You can open teachboard.app and use the board — drawing, shapes, media import, recording yourself and Studio — without an account. In that mode:

• We do not create an account record for you.
• Your board is stored in your browser's local storage, and your recordings in your browser's IndexedDB, on your own machine — we never see either.
• A random identifier is kept in your browser so one visit reads as one session in our product-activity log (§2.7). It is not linked to any name or email unless you later create an account.
• Standard server logs (§2.6) and the pageview counter (§2.7) still apply.

2.2 If you sign in

There are two ways to sign in, and both create an account record holding your email address.

Google. When you sign in via Google OAuth, Google sends us — and we store — the following from your Google profile:

• your Google account email address;
• your display name;
• your profile picture URL;
• your Google subject ID (a stable identifier we use to associate your data with you).

We do not receive your Google password and we do not request access to your Drive, Gmail, contacts, calendar or any other Google scope.

Email. You can instead sign in with just an email address. We email you a one-time code to prove you own the address, and store the address plus a timestamp. You may optionally set a password for faster sign-in afterwards; we store only a salted hash of it, never the password itself.

2.3 Content you create, record or upload

When you use the cloud features of TeachBoard (signed in), we store the projects you save:

• your pages: strokes, shapes, text, tables, charts, diagrams, motion scenes, speaker notes and embedded media (images, sprites, videos, PDFs you imported);
• AI chat history within a project (prompts and AI responses), and the lesson's chosen language and design theme;
• math-render history within a project (prompts and rendered MP4/PNG files);
• saved assets in your asset library (the thumbnails and element data you've explicitly saved).

Microphone and camera. When — and only when — you start a recording with them enabled, your browser captures your microphone and, if you switch it on, your camera. The same applies to a screen share you start mid-recording. These streams are mixed into the recording in your browser; they are never streamed to us live, and we do not access either device at any other time. You can tell the difference by your browser's own recording indicator, which is on only while you are recording.

Where recordings live. Finished takes are stored locally in your browser (IndexedDB) so Studio can work on them offline. They reach our servers only when you choose to: pressing Download or Share uploads the take so it can be converted to a standard MP4 (and, for Share, hosted). Recordings you never download or share never leave your machine.

2.4 Share links & public videos

If you copy a share link for a video, that video and a small metadata file (title, duration, creation date) are uploaded to our storage and served from our CDN at a public, unguessable URL, together with a public watch page. That means:

• Anyone with the link can watch it — there is no sign-in wall, by design. Treat a share link the way you'd treat an unlisted video link.
• The link stays live until you ask us to remove it. To have a shared video deleted, email us with the link (see §14).
• We do not index shared videos, list them publicly, or show them to anyone who does not have the link.

Downloading the MP4 instead of sharing it does not create a public URL: the file is converted and returned to your browser, and the uploaded copy is deleted as part of that step.

2.5 Subscription & billing data

If you subscribe to TeachBoard Pro, Dodo Payments processes the payment. We receive only:

• your subscription status (active / cancelled / past-due);
• the plan you chose;
• the next billing date;
• a Dodo customer/subscription identifier so we can look you up.

We never see or store your full card number, CVV, UPI ID or bank credentials. Those live with Dodo Payments.

2.6 Technical & log data

Like any web service, our servers (and our infrastructure providers — see §5) automatically log:

• your IP address (used for rate-limiting, abuse prevention and approximate region detection);
• the date / time / URL of your request;
• HTTP status, user-agent and referrer (for debugging);
• error stack traces and request IDs when something goes wrong.

2.7 First-party site analytics

So we can see which pages are visited and roughly how the Service is used, every page on teachboard.app includes a small first-party script (/track.js) that sends a single request per pageview to our own servers. For each pageview we record:

• the page path you visited (e.g. /features);
• the referring host, if any (e.g. google.com — query parameters are stripped);
• your browser language, time-zone and a coarse screen-width bucket (xs/sm/md/lg/xl);
• a derived device / browser / OS family from your User-Agent (e.g. "mobile / Chrome / Android");
• a synthetic visitor id: SHA-256(secret salt || today's UTC date || your IP || your User-Agent), truncated to 16 hex characters. The same visitor on the same day always hashes to the same id (so we can count unique daily visitors), but because the hash is salted with the day, the same visitor on a different day gets a different id — we cannot follow you across days.

We do not store your raw IP address, raw User-Agent, cookies, or any other cross-session identifier as part of this analytics. We do not share this data with any third party — it is collected by, stored on, and read from our own AWS infrastructure only. The script respects the browser's Do Not Track signal and does not run if you have it enabled.

2.8 Product-activity events

So we can see where the product breaks or confuses people, the app records a small stream of coarse product events — for example: a page was drawn on, Create video was opened, a render started or failed, a quota limit was hit, a prompt was sent and what kind of thing the AI built. Each event carries the event type, a timestamp, a coarse browser family (e.g. "Chrome"), and short technical detail such as an error message.

Events from a signed-in user are attached to that account. Events from a visitor with no account are attached to the random per-browser identifier described in §2.1; if that visitor later creates an account, the two are joined so one person's session reads as one story. This log is used only by us, to run and fix the product — it is never shared, sold or used for advertising, and it does not record the content of your pages.

2.9 What we don't collect

• We do not run third-party advertising or marketing trackers.
• We do not run third-party web-analytics (no Google Analytics, no Facebook pixel, no Hotjar, etc.). Our pageview counter is the first-party one described in §2.7.
• We do not access your location.
• We do not access your microphone or camera except while you are recording with them enabled (§2.3), and we never receive those streams separately from the recording you choose to upload.
• We do not sell, rent or trade your personal data.

3. How we use your data

We use the data described in §2 to:

• provide the Service to you — letting you sign in, sync projects across devices, build pages with AI, render narrated video, host a video you chose to share, and access your subscription;
• process payments and manage subscriptions (via Dodo Payments);
• communicate with you about your account — receipts, billing problems, security issues, important service changes;
• protect the Service from abuse, fraud and security incidents (rate-limiting, anomaly detection);
• debug, monitor and improve the Service;
• comply with our legal obligations (e.g. tax records, responding to lawful requests).

4. Legal basis (DPDP / GDPR)

For users in jurisdictions that require it (e.g. India under the DPDP Act, the EU/UK under the GDPR), we rely on the following grounds:

• Performance of a contract — to deliver the features you've signed up for.
• Consent — for AI features that send your prompts to Anthropic, for storing your Google profile, and for any optional features you explicitly opt into. You can withdraw consent at any time by deleting your account.
• Legitimate interests — to keep the Service secure, detect abuse, and operate the business. We balance these interests against your rights and freedoms.
• Legal obligation — to comply with applicable laws (tax, accounting, lawful disclosure requests).

5. Third-party services we share data with

We rely on the following sub-processors / providers. Each has its own privacy policy; we only share the minimum data needed for the feature to work.

5.1 Google (sign-in)

Used for OAuth sign-in, when you choose that route. We receive your email, name, profile picture and a stable Google ID. We do not request additional scopes. If you sign in with an email address instead, Google is not involved at all.

5.2 Anthropic (AI)

When you ask the AI for anything — a lesson, a page, a chart or diagram, a math render, an edit — your prompt, attached files (if any) and a summary of the relevant part of your board are sent to Anthropic's Claude API to produce a response. Anthropic, by default on the API, does not use API inputs to train their models. See Anthropic's privacy policy.

If you are on Pro and you supply your own Anthropic API key, those requests are billed to your own Anthropic account. We store your key encrypted and use it only to make the calls you trigger; you can remove it at any time from Settings.

5.3 Amazon Web Services (AWS)

We host the Service on AWS (S3, CloudFront, Lambda, DynamoDB, etc.). Your account data, projects, math renders, shared videos and server logs are stored on AWS. AWS acts as our data processor.

5.4 Amazon Polly (AI narration)

When you render a video with the AI voice, the speaker notes for each page are sent to Amazon Polly — a text-to-speech service within AWS — which returns the spoken audio and the word timings we use to place captions. Only the narration text is sent; your drawings and images are not. Amazon Polly is covered by our AWS data-processing terms and acts as a processor; AWS states that content processed by Polly is not used to develop or improve its services where that opt-out is in effect, and we do not send Polly anything other than the narration you asked to be spoken.

5.5 Dodo Payments (billing)

Dodo Payments processes your subscription as merchant of record. They handle card data, tax calculation, invoicing and recurring billing. See Dodo's privacy policy on their website.

5.6 Freepik (icon & image search)

When you search for an icon or image from inside the board, your search query is sent to Freepik's API. Results shown in the panel are loaded directly from Freepik's CDN, which may log your IP and request metadata under their privacy policy.

5.7 Pexels (stock video search)

When you use the stock-video search, your search query is sent to Pexels' API. Video previews stream from Pexels' CDN, which may log your IP and request metadata under their privacy policy.

5.8 Other recipients

We may also disclose personal data: (a) to comply with a binding legal request or court order; (b) to enforce our Terms; (c) to protect the rights, property or safety of TeachBoard, our users or the public; or (d) in connection with a business transfer (merger, acquisition, asset sale), in which case we will give notice before personal data becomes subject to a different privacy policy.

6. How AI features handle your data

Because AI features are the most data-sensitive part of the Service, here is exactly what happens:

1. You type a prompt (or attach a file) in the AI panel.
2. Our backend forwards that prompt — together with the minimum context needed, such as a summary of your pages or a snapshot of the selected elements — to Anthropic's Claude API. Building a lesson makes several such calls: one to work out what you asked for, one to plan the storyboard, and one per page.
3. Anthropic returns a response; our backend forwards it to your browser and stores it in your project history so you can refer back to it.
4. For a math render, the model writes a Python script which is then executed on our server-side Manim + LaTeX renderer (AWS Lambda). The resulting MP4 / PNG is stored on AWS S3 and served back to you.
5. For a narrated video, the speaker notes of each page are sent to Amazon Polly (§5.4), which returns the audio and word timings. The audio is mixed and the video is recorded in your browser — the finished video is not sent anywhere unless you download or share it (§2.3, §2.4).

We do not use your prompts, attachments or outputs to train any AI model. We do not give your content to Anthropic for model training; Anthropic's API default policy applies.

AI output is generated probabilistically and can be wrong. Check anything you are going to teach from before you publish it.

7. Cookies & local storage

We use a minimum of storage in your browser:

• An authentication token / session marker that keeps you signed in after a Google or email sign-in.
• localStorage: the in-progress state of your board, your UI preferences (theme, frame, last-used tool, narration voice, etc.), recently-used icons, the random session identifier described in §2.1, and similar non-personal app state.
• IndexedDB: your recorded takes and the Studio timelines built from them, so Studio works without uploading anything. Clearing your browser's site data deletes them permanently — we hold no copy unless you downloaded or shared the video.
• sessionStorage: a single flag (tb_seen) used by our first-party analytics script (§2.7) to distinguish the first pageview of a session from subsequent ones, plus a held prompt if you typed one before signing in. Wiped automatically when you close the tab.

We do not use cookies for advertising, marketing or third-party analytics.

8. Data retention

• Account data (email, name, profile picture, Google ID) — kept while your account is active. Deleted within 30 days of account deletion, except where retention is required by law.
• Your projects, pages, AI / math-render history, assets — kept while your account is active. Deleted within 30 days of account deletion (or sooner if you delete the project directly).
• Recordings you never shared or downloaded — never leave your browser; we hold nothing to retain. They live in your browser's storage until you delete them or clear site data.
• Shared videos — kept on our storage and CDN for as long as the link is meant to work. Email us the link and we will delete it, normally within 7 days. A video you only downloaded is deleted from our storage as soon as the conversion finishes.
• Subscription & billing records — retained for the period required by Indian tax and accounting law (typically up to 8 years from the financial year of the transaction). Dodo Payments has its own retention schedule for the underlying transactions.
• First-party site analytics (§2.7) — individual pageview records are automatically deleted after 90 days by a DynamoDB time-to-live policy. After that point only aggregate counts (if any) survive.
• Product-activity events (§2.8) — kept while your account is active and deleted with it; events from visitors with no account are deleted within 90 days.
• Server logs — retained for up to 90 days for security and debugging, then deleted or aggregated.
• Backups — encrypted backups roll on a 30-day window; after that, deleted data is no longer recoverable.

9. Security

We take security seriously:

• All traffic to and from the Service is encrypted in transit (HTTPS / TLS).
• Data at rest in AWS S3 is encrypted using AWS-managed keys.
• API keys for third-party providers are kept in environment variables, never exposed to the browser.
• Access to production systems is restricted and audited.

No system is 100% secure. If we discover a personal-data breach that is likely to result in significant harm, we will notify affected users and the relevant authorities in accordance with applicable law (including the DPDP Act and, where relevant, the GDPR's 72-hour notification requirement).

10. International data transfers

TeachBoard is operated from India. Our infrastructure providers (notably AWS, Anthropic and Dodo Payments) operate in multiple regions, including the United States and the European Union. This means your personal data may be transferred to, stored in or processed in countries other than your own.

Where we transfer personal data outside your jurisdiction, we rely on the safeguards offered by those providers (such as Standard Contractual Clauses for GDPR transfers) and limit the data shared to what is strictly necessary.

11. Your rights

Depending on where you live, you may have the right to:

• Access a copy of the personal data we hold about you;
• Correct data that is inaccurate or incomplete;
• Erase your account and associated personal data;
• Restrict or object to certain processing;
• Withdraw consent where processing is based on consent;
• Port your data — receive a copy in a structured, machine-readable format;
• Nominate another person to exercise these rights on your behalf in the event of death or incapacity (DPDP Act).

To exercise any of these rights, write to us at xeonaicontact@gmail.com from the email associated with your account. We will respond within 30 days, or sooner where the law requires it.

If you believe we've mishandled your data, you also have the right to complain to your local data-protection authority — for users in India, that's the Data Protection Board of India once constituted under the DPDP Act.

12. Children

The Service is not directed at children. We do not knowingly collect personal data from anyone under 18. If you believe a child has provided us with personal data, please email xeonaicontact@gmail.com and we will delete it.

13. Changes to this policy

We may update this Privacy Policy from time to time. The "Last updated" date at the top reflects the latest revision. For material changes, we will give reasonable advance notice — for example by email or via an in-app notice — before they take effect.

14. Contact & data requests

For any privacy question, data-access request, deletion request or grievance under the DPDP Act:

Email

xeonaicontact@gmail.com

Subject prefix

[Privacy] — helps us route faster

Operator

TeachBoard (Indian sole-proprietorship)

Website

teachboard.app