Healthcare Workers Uploaded Patient PHI to Unauthorized AI Tools at Two Institutions
Logged · updated · Source: hipaajournal.com ↗
"40% of hospital workers said they were aware of colleagues using unauthorized AI tools, and 20% admitted to using the tools themselves."
Staff at Weill Cornell Medicine and two Washington state mental health facilities sent protected health information to unauthorized AI tools, triggering HIPAA breach notifications. At Weill Cornell, researchers uploaded radiology data for hundreds of patients to unspecified AI and machine learning tools between June 10 and June 20, 2026, as part of a research study. The data included medical record numbers, radiology accession numbers, clinical diagnoses, medication histories, and surgical records. At Lake Burien Transitional Care Facility and Western State Hospital, a staff member used a personal ChatGPT account to draft case-note summaries for mental health patients, exposing names, diagnoses, and medications. Twelve families were notified in the Lake Burien incident; four Western State Hospital patients were also affected.
The incidents are not isolated. A Wolters Kluwer Health survey found that 40% of hospital workers knew colleagues using unauthorized AI tools, and 20% admitted doing so themselves. Consumer AI tools lack the contractual and technical safeguards required under HIPAA, and the risks go beyond privacy: the article also references an ongoing lawsuit against OpenAI after a woman with epilepsy died following a medication-reduction schedule she received from ChatGPT.