Privacy & Data Handling

Account and onboarding data: when you register, we collect your email address and display name. During onboarding we also collect your clinical role, country, work setting, and organisation so the app can tailor outputs to your context.

Browser-stored workspace data: the app saves your in-progress assessment, analysis output, follow-up chat, selected medications and management actions, custom templates, role/country/location cache, and certain interface preferences in browser storage on your device.

Clinical content you choose to process: if you use AI analysis, follow-up questions, or document export, the relevant assessment content is sent to our backend endpoints so we can generate the requested output.

Usage analytics: we use Vercel Analytics and Vercel Speed Insights to understand traffic and performance.

Firebase Authentication: manages sign-in, password resets, and Google sign-in.

Firestore: stores your user profile record. Firestore rules in this application restrict profile access to the signed-in user who owns that document.

Browser storage: stores the working assessment session and templates locally on your device so the workflow can resume if you navigate away and return.

No server-side assessment database: this codebase does not store assessment drafts in Firestore or another ClinicalAssist database. Assessment content is processed when you invoke AI or export features, but it is not persisted as a patient record in the application database.

When you ask ClinicalAssist to analyse an assessment, answer a follow-up question, or generate a response that depends on model output, the relevant request content is sent to our `/api/chat` endpoint and then routed through Vercel AI Gateway.

ClinicalAssist currently uses OpenAI and Anthropic models via the Vercel AI Gateway. Based on the providers' published documentation as of 8 March 2026:

• OpenAI says customer data sent via the API is not used to train OpenAI models by default. OpenAI also says API data may be retained for abuse and misuse monitoring for up to 30 days unless a separate agreement applies.
• Anthropic says it does not train on customer data from the Anthropic API by default and says prompts and outputs are deleted from its backend within 30 days unless a longer period is required for security, legal, or contractual reasons.
• Vercel documents AI Gateway as using zero data retention by default, meaning prompts and responses are deleted after the request completes. Downstream model providers may still apply their own retention policies.

Because AI requests leave your device when you use these features, you should avoid entering direct patient identifiers unless your organisation's governance process explicitly allows it.

User profile data: retained while your account remains active, subject to support, legal, and operational requirements.

Browser-stored workspace data: remains on your device until you clear browser storage, overwrite it with a new session, or remove saved templates and preferences.

AI request and provider retention: follows the provider policies described above.

Deletion requests: there is not currently a self-serve account deletion tool in the app. If you need your account/profile data deleted, contact us at hello@clinicalassist.io.