Glass Box Medicine

When people hear “healthcare AI” their minds often jump to high-stakes clinical models–and indeed, clinical models have immense potential to improve human health and save lives.

Most readers of this blog know me as a physician scientist and entrepreneur who writes a lot about health AI.

In the age of AI, electronic health records (EHRs) need a makeover.

Ghosts of dead projects haunt the valley between health AI research and health AI implementation.

Last week, a frontier model at OpenAI hacked into Hugging Face to cheat on an evaluation.

During graduate school, I was obsessed with barefoot shoes—those thin-soled foot gloves with a separate pocket for each toe.

Three recent studies have highlighted safety limitations of LLMs in direct patient interactions.

I was the Founder and CEO of Cydoc, a health AI startup, for seven years (April 2018 to August 2025).

This article explains acid-base analysis—specifically, how to interpret an Arterial Blood Gas (ABG) and Basic Metabolic Panel (BMP) to understand a patient’s acid-base status.

HealthBench is a recently released benchmark to evaluate large language models in healthcare.

A recent New York Times article titled AI Chatbots Defeated Doctors at Diagnosing Illness covers a study recently published in JAMA Network Open.

This post explains how to set up an AWS Deep Learning Amazon Machine Image, or AWS DLAMI.