Privacy

Plain-language summary of what the app records and how that data is handled.

What we record

What we do not record

Where the data goes

Data is stored in a Cloudflare D1 database operated by the research team. Cloudflare is the platform host; we do not share data with any other party. Researcher-facing exports are anonymized (no recruitment identifiers in the analysis dataset).

AI calls

The reading coach runs on Cloudflare Workers AI (the Qwen model), invoked server-side from within the same Cloudflare account. Your selected passage and the focal paragraph are sent to the model as context for each request. No persistent profile is built from your inputs.

Recruited studies

If you arrived through CloudResearch, SONA, or another recruitment platform, the platform's own consent and privacy terms apply alongside this notice. The recruitment identifier (e.g. a worker_id) is recorded so payment can be confirmed, and is removed from the analysis dataset.

Your control

Studies using this tool are reviewed by the University of Alabama IRB. The specific consent text for a recruited session appears at the start of the Study 1 procedure and supersedes this general notice where they differ.