Ethan Goh, MD

Focused on how frontier AI models are safely translated into real world care.

Engaged in multi-institutional efforts shaping the evaluation and alignment of medical AI systems.

Advises healthcare, life sciences, technology teams, and investors on clinical AI strategy, translational pathways, and commercialization.

Current Initiatives

ARISE
Stanford-Harvard research network advancing evaluation and real-world deployment of AI systems in healthcare
Medical AI Superintelligence Test (MAST)
Unified evaluation framework for frontier models across clinical reasoning, safety, and decision-making
Physician-AI Collaboration Teaming (PACT)
ARPA-H-funded project evaluating how physicians and generative AI work together on high-risk clinical tasks, using real-world EHR data

State of Clinical AI Report
Synthesis of themes across high-impact clinical AI research, translating latest findings into implications for real-world deployment
HAI AI Index: Science & Medicine
Benchmarking major advances in AI within biomedical research and clinical care
FDA public comment: measuring AI medical device performance
Invited commentary on real-world evaluation frameworks, including LLM-as-judge approaches for post-deployment oversight

How physicians actually use AI
Training doctors to use AI matters less than task specific workflow and product design
Omission harms are the dominant safety risk in clinical AI
The overlooked danger of AI failing to suggest critical actions
Why AI agents inside the EHR remain unreliable
Technical barriers to autonomous clinical actions in electronic health records

Select advisory and board roles with healthcare, life sciences, and technology organizations.