Privacy, confidentiality, and compliance
An overview of how Feelpath handles consent and sensitive data for research collaborations.
Quick summary
Consent-first
Participants choose what to save and share at session start.
HIPAA posture
BAAs are in place with core vendors used to run the platform.
Audit logging
Access and system activity is logged for accountability.
How we handle sensitive data
Privacy & compliance
This page is written to be clear and honest. We separate what is true today from what is proposed or planned for research partners.
What we do today
BAAs with core vendors
TodayWe have BAAs in place with the vendors we use to run Feelpath (including Zoom, AWS, Kinde, and OpenAI).
Consent choices at session start
TodayAt the start of a session, participants choose whether to save a transcript for insights and whether to share it (based on the options shown in the join flow).
HIPAA-grade logging and accountability
TodayWe maintain HIPAA-compliant audit logging to support accountability and review.
Proposed study modes (US-only)
Mode A: Clinical / PHI studies
ProposalStudies that use identifiable clinical data. This mode is stricter by design and would use tighter access controls and partner-specific governance.
Mode B: De-identified research datasets
ProposalStudies that use datasets prepared to be safer to share and analyze across sites. This mode is designed to reduce exposure while supporting research workflows.
Abridged guide
Web-friendly version
This is a short version of the guide for quick reading. The PDF is the full document you can share with your team.
What we do today
- BAAs are in place with core vendors used to run Feelpath (Zoom, AWS, Kinde, OpenAI).
- Participants choose what to save and what to share at session start (based on the join flow).
- We maintain HIPAA-compliant audit logging for accountability and review.
Proposed study modes (proposal)
- Mode A: clinical / PHI studies (stricter controls and partner-specific governance).
- Mode B: de-identified research datasets (designed to reduce exposure and support research workflows).
- US-only for now; international support is on the roadmap.
How research data works
- What we collect: session metadata, consent choices, transcripts (when enabled), AI outputs (when enabled).
- When we collect it: consent at session start; transcript during session; insights after session (if enabled).
- Today: exports are oriented around the session speaker. Proposal: study exports with a clear field list (data dictionary).
Study templates (proposal)
- IRB-ready study description (starter text).
- Consent language blocks (starter text).
- Data dictionary for exports (what each field means).
Download the full guide (PDF)
A longer version you can share with your team (US-only for now).