Who is this article for?

Users and admins who want to learn more about AI prompts in Ideagen Healthcare Guardian

Access to the AI prompts is required.

This article explains how to use the ‘Do not send to AI’ setting in Ideagen Healthcare Guardian to safeguard sensitive data—such as personally identifiable information (PII)—when using AI-powered prompts in questionnaires.

How AI accesses questionnaire data

The AI in Ideagen Healthcare Guardian only processes the data explicitly sent to it from a questionnaire. It does not have access to any other system data.

By default, all questionnaire data is included when running an AI prompt. However, some questions or sections may contain sensitive information that should not be shared with the AI.

Using 'Do not send to AI' settings

To protect sensitive data:

• You can mark individual questions or entire sections as ‘Do not send to AI’.
• Any data in these marked areas will be excluded from the AI request, ensuring it is never processed or stored by the AI.

This helps prevent accidental sharing of sensitive information such as PII.

Priority over source selection

The ‘Do not send to AI’ setting takes precedence over any source questions or sections selected in the AI prompt configuration. This means:

• Even if a question is selected as a source, it will be excluded if marked as ‘Do not send to AI’.
• This ensures maximum confidence in your data security and compliance.

Benefits of this feature

• Data protection: Prevents sensitive information from being shared with AI.
• Compliance support: Helps meet data privacy and regulatory requirements.
• User control: Gives administrators full control over what data is used in AI processing.