Published June 15, 2026

Workshop open for registration: Simulation and causal models for AI evaluation and regulation

ARPA-H is convening a small group of researchers, clinicians, and technologists to explore the current state of the science regarding simulation, causal inference and foundation models and their potential ability to rigorously evaluate clinical artificial intelligence (AI) for regulatory use.

AI is entering healthcare faster than our ability to evaluate it. A new generation of clinical AI — conversational, agentic, multimodal, and continuously learning — is emerging without a clear evaluation framework.

The current standard for clinical AI evaluation – the clinical study or trial – is expensive, slow, often unblinded, and inherently static. It evaluates a fixed, locked version of a technology that may already have evolved by the time results are published. For software that continuously updates, this creates a fundamental mismatch between how clinical AI is developed and how it is evaluated.

Several fields are now advancing in ways that suggest an alternative evaluation process may be possible. Causal inference methods are moving from theory into validated clinical use, but use cases are very restrictive. Clinical data exists at unprecedented scale but is not structured for causal reasoning; it captures what happened, not what would have happened under different decisions. Continuous wearable data captures physiology between visits, and ambient clinical documentation captures the reasoning behind decisions, the alternatives considered, and uncertainties weighed — absent from structured records. Medical foundation models are being developed using data from hundreds of millions of patients, but we are hitting the limits of scaling, given the limitations of existing electronic health data.

The goal of this workshop is to identify the hardest problems and explore opportunities for solutions that existing academic, industry, and government incentives are unlikely to produce on their own.

Workshop details

Event date: July 16, 2026, 12:00 – 3:30PM ET

Event location: Virtual

Interested individuals and organizations may submit inquiries to ai.simulation@arpa-h.gov.