Practical AI in Healthcare S1, E17: Kathy Roe & Kenny White: The Legal Realities of AI in Healthcare
Dec 21, 2025
Kenny White, an insurance veteran who helps health systems spot and finance risk. Kathy Roe, a health lawyer advising on regulation, privacy, and care delivery. They unpack liability differences across AI types. They explore HIPAA and de-identification limits. They cover intellectual property, contractual risk, and how insurers and courts are adapting to AI in healthcare.
AI Snips
Chapters
Transcript
Episode notes
Define Risk Broadly Before Acting
- Risk means the possibility of harm and includes legal, financial, and moral dimensions in healthcare AI.
- Organizations must identify specific risks before trying to mitigate or transfer them.
Dual Sources Of Liability
- Liability can arise from both the AI product and the clinician's use of it, creating overlapping legal theories.
- Courts will need to parse where the tool's action ends and the clinician's action begins to assign responsibility.
De-Identification Faces New Limits
- HIPAA's minimum-necessary and de-identification rules strain under AI's appetite for large datasets.
- Safe harbor de-identification may no longer assure anonymity as re-identification techniques improve.