ChatGPT for Healthcare puts agents inside hospital software: reading screens, clicking, and moving records while clinicians keep review
Synopsis
OpenAI is bringing computer-use agents to ChatGPT for Healthcare that carry out multi-step work across existing healthcare software, alongside improved trusted clinical search and more than 20 new healthcare plugins. Healthcare teams previously had to piece together information by hand from the EHR, hospital policies, scheduling systems, and payer portals; the agents read screens, click, type, and move between applications, with access controls and audit logs. In a blind evaluation spanning multiple specialties and multi-turn conversations, physicians rated the updated clinical answers more highly across all eight dimensions over more than 1,300 reviews.
Interpretation
ChatGPT for Healthcare gains computer-use agents that carry out multi-step work inside the software hospitals already use, such as reviewing a chart and drafting an order and patient reply for human review. Healthcare teams previously had to piece together information by hand from the EHR, hospital policies, scheduling systems, and payer portals to prepare documentation, coordinate care, and complete administrative tasks. OpenAI also describes work with healthcare organizations deploying agents, including Abridge, Ambience, Cigna Healthcare, and CVS Health, to process referrals, move documents between systems, and bring together the information care teams need.
The agents work by operating existing software interfaces: reading screens, clicking, typing, and moving between applications. Computer-use agents are the capability introduced here, and organizations can start with a workflow that spans existing applications and then set which applications agents may access. In the demo, a clinician uses Voice in the ChatGPT app to ask for help following up on an MRI request, and the agent reviews the chart and drafts an order and patient reply for review.
Improvements to trusted clinical search let ChatGPT find, evaluate, and cite medical evidence with citations clinicians can inspect, and connect that evidence to practical next steps. The updates cover trusted clinical search in both ChatGPT for Healthcare and ChatGPT for Clinicians. In a blind evaluation, physicians reviewed responses to real-world clinical questions across multiple specialties, including multi-turn conversations, and in over 1,300 reviews rated the updated experience more highly across all eight dimensions, including factual correctness, safety, actionability, and source authoritativeness.
Perspective
The demonstrated scope is computer-use agents, improved trusted clinical search, and more than 20 healthcare plugins in ChatGPT for Healthcare and ChatGPT for Clinicians, aimed at clinical, research, and administrative teams; workflows involving protected health information require eligible services covered by a BAA, the required security and retention settings, and protections extended to connected systems.
OpenAI describes the agent models as more capable, safe, and affordable without giving a corresponding quantitative comparison; the higher clinical-answer ratings come from physician blind review, not from clinical outcomes or patient-level results.
