Privacy and consent for AI-moderated interviews — practical checklist

Why AI interviews need explicit care

Conversational interviews can elicit richer personal detail than a three-question form. That is the point — and the responsibility. Product and CX teams should treat Insight sessions like research: purpose limitation, transparency, and minimization.

Checklist before you publish

• State why you are interviewing and how findings will be used
• Add your privacy policy link in participant appearance settings
• Avoid prohibited topics (health, politics, etc.) unless you have a lawful program
• Keep blueprint fields to decisions you will actually make
• Use CRM invites only for contacts you may lawfully message
• Limit who in admin can export transcripts and notes

Consent patterns that work

For website embed, show welcome copy that feedback is voluntary and used to improve the product. For CRM waves, align with your existing research or marketing consent. For employees, follow HR policy and local employment rules.

What KogniFeed provides

• Privacy policy link on the participant experience
• Objectives guardrails for prohibited topics
• Workspace controls for who runs studies and sees analysis
• Owned-audience distribution so you are not buying stranger panel data by default

What you still own

Legal basis, DPIAs, retention schedules, and vendor reviews remain your responsibility. Use this guide as an operational checklist — not legal advice.