The National Academy of Medicine (NAM) officially launched its "Patient Safety in the Era of AI" initiative on March 4, 2026. This massive cross-sector effort aims to create a national strategy for integrating artificial intelligence into the US healthcare system. The goal is to use machine learning to predict risks and prevent medical errors, moving closer to a "zero harm" vision for patients.
Medical leaders from the Mayo Clinic and CommonSpirit Health are co-chairing the steering group. They argue that while AI is already being used for diagnosis, a lack of unified strategy could lead to new safety gaps. The initiative will open a public comment period to allow clinicians and patients to weigh in on how these tools should be governed.
The project is backed by a coalition including the Elevance Health Foundation and the California Health Care Foundation. Researchers hope that AI can specifically target persistent issues like diagnostic delays and medication errors. By standardizing AI use, the NAM hopes to ensure that technology serves as a safety net rather than a source of confusion.
Experts believe this is a pivotal moment for American medicine as digital health tools proliferate. The initiative will eventually produce a strategic framework that outlines the responsible use of generative AI in clinical settings. This framework is expected to influence how hospitals and private practices implement new software over the next decade.
NAM launched a national initiative to govern AI in healthcare safety.
The project aims to reach "zero harm" by using AI to predict clinical risks.
It involves leaders from major health systems like Mayo Clinic.
A public comment period will allow stakeholders to shape the final strategy.
The initiative focuses on bridging safety and equity gaps in modern medicine.
HISTORY
The push for AI regulation in healthcare accelerated significantly in 2024 and 2025 following the rapid adoption of large language models. Early implementations showed promise in administrative tasks but raised concerns regarding algorithmic bias and "hallucinations" in clinical settings.
2024: The FDA issued its first comprehensive framework for AI-based medical devices.
2025: Major US hospital systems began using generative AI for real-time patient charting.
Mid-2025: Reports of "black box" algorithms causing misdiagnosis led to calls for NAM intervention.
Late 2025: The White House issued an Executive Order on the Safe Use of AI in Healthcare.
Jan 2026: NAM announced the formation of the steering group to address these emerging threats.