PRIVACY
POLICY
Last updated: 11 May 2026
This Privacy Policy describes what data AI Crucible collects when you visit our website, join our waitlist, enrol in a NEURON cohort, or use the Prompt Lab during a live session. We have written it in plain language; if anything is unclear, please email hello@ai-crucible.org and we will rephrase.
WHO WE ARE
AI Crucible is an educational programme based in Sri Lanka. The "data controller" responsible for any personal data you provide is AI Crucible. Our contact email for all privacy questions is hello@ai-crucible.org.
WHAT WE COLLECT
When you join the waitlist
- Your email address (required).
- Your name (optional).
- Which NEURON level and audience track you are interested in (optional).
- The page on our site you signed up from (captured silently).
- Your IP address and browser user-agent string at the moment of signup (captured silently — used only for abuse prevention).
When you enrol in a cohort
- Your full name, phone number, and email address (required for cohort administration, attendance, and certificate issuance).
- Your audience track (student / professional / creator / founder) and any track-specific context you choose to share (e.g. profession).
- Payment records — amount, currency, gateway transaction ID, payment status. We never see your card number, CVV, or expiry. Card entry happens on PayHere's hosted page; we receive only a transaction reference and a masked card number for receipt purposes.
When you use the Prompt Lab
- The prompt drafts you submit to the Prompt Lab during a live session, stored against your participant record so we can show your history and your instructor can review your progress.
- The critique the Prompt Lab returns for each draft.
- A timestamp for each.
The Prompt Lab critiques your prompts; it does not answer them. Your actual conversations with Claude happen in your own Claude Desktop — a consumer account you control, governed by Anthropic's consumer terms. AI Crucible is not a party to that data flow.
HOW YOUR PROMPT LAB DATA IS HANDLED
Anthropic API data is not used to train Anthropic's models by default. The Prompt Lab and the pre-assessment run on the Anthropic API, so your drafts and critiques are not added to any training corpus — a meaningful difference from consumer products like claude.ai.
Anthropic does retain API logs for up to thirty days for safety and abuse review, per their standard API terms. After that, the logs are deleted from Anthropic's systems. Our own database keeps your drafts and critiques for as long as your enrolment is active, so you can see your own history during the cohort.
Every query against your Prompt Lab history is scoped to your participant identifier. Other students cannot see your drafts, and you cannot see theirs — this is enforced at the application level, not at the API level.
HOW WE USE THE DATA
- Waitlist email: a single announcement when enrolment opens. No newsletter, no drip campaign, no third-party marketing.
- Enrolment data: cohort administration, attendance tracking, certificate issuance, refund processing.
- Prompt Lab drafts and critiques: presenting your own history back to you within an active session; pedagogical review by your instructor; aggregated, anonymised analytics about how the cohort is using the Lab (e.g. number of critiques run).
- Payment records: bookkeeping and refund processing, retained for the period required by Sri Lankan accounting and tax law.
WHO WE SHARE DATA WITH
We share data only with the third parties needed to operate the programme:
- Anthropic — operator of the Claude API that powers the Prompt Lab and pre-assessment. Your prompt drafts are transmitted to Anthropic to generate the critique. Anthropic's privacy practices are documented at anthropic.com/legal/privacy.
- PayHere — Sri Lankan payment gateway. Card entry happens on their site; we receive a transaction reference only.
- SMSLenz — used for transactional SMS such as instalment reminders. They receive only your phone number and the message text.
- Our email service provider (Resend or Postmark, depending on the deployment) — used to send transactional email such as waitlist confirmations, enrolment receipts, and certificate notifications.
- Laravel Cloud — the hosting platform our application runs on. Standard server-side logs are retained per their policy.
We do not sell your data, share it with advertisers, or use it for ad targeting. We do not run third-party analytics or marketing pixels.
YOUR RIGHTS
You may, at any time, by emailing hello@ai-crucible.org:
- Request a copy of the personal data we hold about you.
- Ask us to correct anything that is wrong.
- Ask us to delete your data. We will do so unless we are legally required to retain it (e.g. payment records for tax purposes).
- Withdraw from the waitlist or from a newsletter you never signed up for (this is a joke — we don't have one).
COOKIES
We use a single session cookie required to keep you logged in to the app once you are enrolled. We do not use tracking cookies, analytics cookies, or advertising cookies.
CHILDREN
NEURON is open to students aged twelve and older. Participants under sixteen require a parent or guardian to consent to enrolment. We do not knowingly collect data from anyone younger than twelve.
CHANGES TO THIS POLICY
If we change anything material, we will update the "last updated" date at the top of this page and email any active enrollees before the change takes effect.
CONTACT
All privacy questions and requests: hello@ai-crucible.org. We aim to reply within two business days.