AI Usage Policy

How we use artificial intelligence

The app uses artificial-intelligence models from a third-party provider to power several features: a conversational assistant that explains health topics and helps you prepare for appointments; a background safety check that reviews each response before it is shown; assistants that offer to save a question for your doctor or to add an item to the record for your approval; and a reader that extracts details from a document or photo you provide.

The models are accessed through a secure server that holds our access credentials. Those credentials are never present in your browser or on your device.

What is sent to the AI

When you ask a question, your message and the relevant part of the record for the person you are currently viewing (such as their conditions, medications, allergies, results, care team, visits, and related notes) are sent to that provider to generate a response. When you use the document reader, the image you provide is sent to be read.

To run the background safety check and the assistants that offer to save a question or update the record, we also send the resulting response and the recent messages in the same conversation, along with the names of items already on that person's record, to the same provider. It is handled the same way, and none of it is used for training.

What is not sent

Your password is never sent. Information about other people in your account, or records you are not currently viewing, is not sent. Only what is needed for the feature you are using is included.

How AI data is handled

The provider processes what we send only to generate the response we request. The app reaches it through a developer API under commercial terms that govern that use, and under those terms what we send and what comes back are not used to train the provider's models.

If we ever change provider, or the terms we hold with one change, this policy changes with them before anything else does.

How to interpret AI responses

Treat every AI response as general information to discuss with a professional, never as a decision. AI responses are not medical advice, are not reviewed by a clinician before you see them, and can be inaccurate or incomplete.

The app is designed to explain and prepare, not to diagnose, prescribe, or tell you what to do about a specific person's care. It routes those judgments to your doctor or pharmacist, and so should you.

Limits of AI in a health setting

Artificial intelligence has real limits that matter more here than almost anywhere:

• •It can state something false with complete confidence.
• •It does not know anything about a person beyond what has been entered, which may be partial or wrong.
• •It cannot examine anyone, cannot account for the full clinical picture, and cannot replace a professional's judgment.
• •It is not a medical device and is not intended to diagnose, treat, cure, or prevent any disease.

You stay in control

For these reasons the app keeps a person in control. It never changes your record on its own, and anything it suggests adding is shown to you to approve or decline.

Safety measures

The app is built to hold a clear line. A separate automated check reviews each response before it is shown and redirects or withholds content that would cross into a personal clinical verdict.

Messages that suggest an emergency or a risk of self-harm receive an immediate, fixed response pointing to 911 or the 988 Lifeline, with no AI involvement.

These measures reduce risk but do not eliminate it, which is why the guidance to confirm with a professional always applies.

Compliance, and not a substitute for professional advice

We use the provider's services in accordance with their usage policies, including the requirements that apply to consumer-facing and health-related uses, such as keeping a person in control of decisions and telling you clearly that you are interacting with AI.

Above all: the AI in this app is a tool to help you understand and prepare. It is not a substitute for professional medical advice, diagnosis, or treatment. Always seek the advice of a qualified health professional, and never disregard or delay it because of something the app produced.