Trump and Kennedy seek to relax safeguards for AI healthcare tools
2026-05-20
The Philadelphia Inquirer
The information displayed in the AIM should not be reported as representing the official views of the OECD or of its member countries.
Kaiser Permanente clinicians in Oakland, California report that AI-powered note-taking software by Abridge often produces inaccurate clinical notes, requiring corrections and risking missed clinical nuances. Proposed regulatory changes by Trump and Kennedy could relax safeguards, increasing the risk of patient harm from flawed AI documentation in healthcare.[AI generated]
Why's our monitor labelling this an incident or hazard?
The AI system (note-taking software) is explicitly mentioned and is in active use, with documented issues causing clinicians to spend extra time correcting errors, indicating indirect harm to healthcare providers and potential indirect harm to patients if errors go unnoticed. The article also highlights concerns from safety researchers about the risk of clinicians relying on inaccurate AI-generated notes, which could lead to patient harm. The proposed regulatory changes to relax safeguards and transparency requirements increase the plausible risk of future harm. Since there is already realized harm (clinicians correcting errors, potential for patient harm) and credible risk of future harm due to weakened regulations, the event meets the criteria for an AI Incident. The regulatory changes and ongoing concerns also represent an AI Hazard, but incidents take precedence in classification. The article is not merely complementary information because it focuses on the harms and risks associated with AI system use and regulatory rollback, not just updates or responses.[AI generated]