Turn on cookies to sign in
Signing in saves your progress to your Curie account. We can only do that with cookies on — turn them on to continue.

Most healthcare organizations are experimenting with AI but can't seem to move beyond small pilots. Johns Hopkins professor Ritu Agarwal explains why scaling AI across health systems is so hard — and what leaders need to do differently. From clinician trust to ROI measurement, the path forward requires more than just good technology.
Healthcare has spent decades digitizing its infrastructure, but that hasn't translated into AI working at scale. Ritu Agarwal, a professor at the Johns Hopkins Carey Business School, says most health systems are still stuck in pilot mode — unable to embed AI meaningfully into clinical and administrative workflows across entire organizations. The culprits? Model drift (when AI trained on one patient population underperforms in another), unclear workflow integration, and a lack of clinician buy-in.
Agarwal argues that physicians need to be leading AI development, not just consulted on it. She also cautions health system executives against chasing immediate ROI — instead, she recommends setting clear measurement baselines and focusing on specific disease conditions before trying to scale broadly.
Key Takeaways:
Why it matters: AI holds real promise for reducing healthcare costs, improving access, and addressing inequities — but only if health systems move beyond experimentation and build the governance, trust, and measurement frameworks needed to deploy it responsibly at scale.