AI in Healthcare Raises Privacy Gaps and Patient-Safety Risks

AI-driven healthcare tools are expanding rapidly, but legal and security protections for patient data often lag behind their clinical ambitions. Reporting highlighted that consumer-facing medical chatbots and AI health offerings from OpenAI, Anthropic, and Google may fall outside HIPAA obligations in many common use cases, meaning sensitive health information shared with these services may not receive the same statutory protections as data handled by regulated healthcare providers; experts warned that terms-of-service promises are not equivalent to regulated safeguards and that non-HIPAA consumer health data can be sold or shared with third parties, including data brokers.

Separately, an investigation summarized from Reuters described patient-safety concerns tied to “AI-enhanced” medical devices, citing lawsuits and FDA adverse-event reporting that allege AI-related changes contributed to serious surgical injuries. One example involved an AI-updated sinus surgery navigation system where reported malfunctions increased sharply after an AI “enhancement,” though the reporting noted FDA incident data is incomplete and does not by itself prove causation; the same coverage also pointed to a higher recall rate for FDA-authorized medical AI devices versus baseline and described FDA capacity constraints in reviewing AI-enabled devices due to staffing losses in relevant technical teams.