Protecting patient privacy in clinical foundation models: Technical and legal perspectives
cs.AI, cs.LG
Submitted: 2026-08-07
Updated: 2026-09-15
Comments: 14 pages, 2 Figures, 2 Tables
License: http://creativecommons.org/licenses/by/4.0/
The gist: Clinical foundation models trained on large-scale patient data are increasingly used for decision support, screening, and public health.
Terminology
Abstract
Clinical foundation models trained on large-scale patient data are increasingly used for decision support, screening, and public health. As deployment expands, privacy risk increasingly arises from model-mediated leakage, yet its prevalence and severity remain poorly quantified. Models can disclose sensitive training artifacts, enabling patient re-identification in ways not captured by data-handling controls alone. Existing frameworks, including HIPAA and GDPR, offer limited guidance for such indirect threats. We propose a practical framework for assessing privacy risk in clinical foundation models and illustrate realistic leakage scenarios across deployment settings, map them to legal regimes, and outline complementary technical and legal mitigations. Our analysis provides a context-aware risk assessment grounded in realistic usage to preserve the value of medical foundation models while rigorously safeguarding patient privacy.
Sources
- MedGemma Technical Report
- MedAlpaca -- An Open-Source Collection of Medical Conversational AI Models and Training Data
- Generalization through Memorization: Nearest Neighbor Language Models
- The Inadequacy of Offline LLM Evaluations: A Need to Account for Personalization in Model Behavior
- Quantifying Memorization Across Neural Language Models
- On Memorization in Diffusion Models
- An Investigation of Memorization Risk in Healthcare Foundation Models
- The Canary's Echo: Auditing Privacy Risks of LLM-Generated Synthetic Text
- Safety Layers in Aligned Large Language Models: The Key to LLM Security
- Transferability in Machine Learning: from Phenomena to Black-Box Attacks using Adversarial Samples
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems