OpenAI is expanding its healthcare offerings with computer-use agents, improved clinical search and more than 20 new healthcare plugins. The additions are intended to help care and administrative teams work with information held in separate records, scheduling systems and insurance portals.
Agents in existing systems
Computer-use agents can read screens, click, type and move between applications, allowing them to carry out multistep tasks in software teams already use. OpenAI says the advances include more capable, safer and more affordable models, as well as access controls and audit logs for clinical and administrative deployments.
OpenAI is also working with Abridge, Ambience, Cigna Healthcare and CVS Health on deploying the agents. Uses include processing referrals, moving documents between systems and gathering information for care teams. In one demonstration, a clinician uses a voice request to seek help following up on an MRI request; the agent reviews the patient record and drafts an order and a patient reply for review.
Organizations can specify which applications agents may access and which steps require human review. For workflows involving protected health information that comply with the US Health Insurance Portability and Accountability Act (HIPAA), organizations must use eligible services covered by their Business Associate Agreement (BAA) with OpenAI, apply required security and retention settings, and extend protections to connected systems. OpenAI says content shared with ChatGPT for Healthcare is not used to train its models.
Clinical search and plugins
OpenAI has improved how clinical search in ChatGPT for Healthcare and ChatGPT for Clinicians finds, evaluates and cites medical evidence. The aim is to give clinicians evidence relevant to their questions, with inspectable citations and practical next steps.
According to the company, physicians conducted a blind evaluation of answers to real-world clinical questions across multiple specialties, including multistep conversations. Across more than 1,300 reviews, they rated the updated experience higher on all eight dimensions assessed, including factual correctness, safety, actionability and source authoritativeness.
The more than 20 new plugins support clinical care, care coordination and business operations. They are available in ChatGPT for Healthcare, including through ChartGPT, which brings ChatGPT into electronic health record workflows. The plugins can help teams review patient records and hospital policies, prepare prescriptions for clinician review, check insurance eligibility and analyze costs within a conversation.
OpenAI’s examples include using Luma Health to find appointments awaiting confirmation and TriZetto to check insurance eligibility. A hospital discharge summary found through Metriport can be used alongside UpToDate recommendations to follow up on medication changes. Other tasks can combine hospital policies from AIVA with on-call information from Amion, or a procedure code from IMO Health with insurance details from Stedi and rates from Turquoise Health. Data from Assort Health and Ferry Health can help teams review referral trends and outstanding scheduling work.
Availability
ChatGPT for Healthcare provides a governed workspace for clinical, research and administrative teams. ChatGPT for Clinicians is free for eligible, verified clinicians in the United States and can be used to review medical evidence and prepare clinical documentation. Both products offer a BAA with OpenAI to support HIPAA-compliant workflows.
For healthcare organization customers, computer-use agents in ChatGPT Work and Codex and improved clinical search are available now, subject to workspace settings. A workspace administrator can enable the new plugins. Individual clinicians can also use the agents and search, as well as select plugins. Developers can access computer-use capabilities through the OpenAI API, ChatGPT Work and Codex. ChatGPT for Healthcare can be purchased directly or through an AWS Marketplace private offer.



