OpenAI adds agents and plugins to ChatGPT for Healthcare
OpenAI says ChatGPT for Healthcare now brings healthcare teams’ existing tools and medical evidence into one platform. The announcement adds computer-use agents and more than 20 plugins for clinical care, care coordination, and business operations.

Key takeaways · 4
- 01
Check whether your workflow could benefit from agents that move documents, process referrals, or assemble care-team information.
- 02
Organizations can choose which applications agents access and which steps require human review.
- 03
The plugin ecosystem has more than 20 additions, with tasks including record review, prescription preparation for clinician review, outreach coordination, and insurance-eligibility checks.
- 04
OpenAI says healthcare content is not used to train its models, but organizations must use eligible services covered by their BAA and apply required security and retention settings.
Agents work across existing systems
OpenAI says its computer-use agents can read screens, click, type, and move between existing healthcare applications.[1] The company says the agents include more capable, safe, and affordable models for clinical and administrative workflows.[1] Listed tasks include processing referrals, moving documents between systems, and assembling information for care teams.[1] In one described workflow, an agent reviews a chart and drafts an order and a patient reply for review.[1] This setup is aimed at work that crosses software systems, while leaving review steps under organizational control.[1]
Plugins add workflow options
OpenAI says it expanded its healthcare plugin ecosystem with more than 20 plugins from companies in clinical care, care coordination, and business operations.[1] The plugins are available in ChatGPT for Healthcare, including through ChartGPT, which brings ChatGPT into the EHR workflow.[1] OpenAI says supported work includes reviewing patient records and hospital protocols, preparing prescriptions for clinician review, coordinating outreach, and checking insurance eligibility.[1] OpenAI also says ChatGPT can find, evaluate, and cite medical evidence.[1] Organizations considering these workflows should distinguish assistance and preparation from completed clinical decisions.
Controls and evaluation results
Organizations can specify which applications agents may access and which steps need human review, according to OpenAI.[1] The company says organizations must use eligible services covered by their business associate agreement with OpenAI and apply required security and retention settings.[1] OpenAI says content shared with ChatGPT for Healthcare is not used to train its models.[1] For its clinical-search evaluation, OpenAI says it gathered more than 1,300 physician reviews and assessed factual correctness, safety, actionability, and source authoritativeness.[1] Physicians rated the updated experience more highly across all eight dimensions evaluated.[1]
Reported performance and deployments
OpenAI says physician-shaped tasks test GPT-6 Astra’s clinical reasoning over longitudinal patient context and its navigation of complex interfaces.[2] It reports that 99% of 325 reviewed clinical tasks were rated safe, and physicians rated GPT-6 Astra very or extremely helpful in 93% of 646 assessments.[2] On Stanford researchers’ HealthAdminBench, GPT-6 Astra earned 94.3% of available rubric points and GPT-6 Luna earned 91.8%; OpenAI says GPT-6 Luna cost 8.5 cents per task, 98% less than an unspecified comparator.[2] CVS Health says one agent cleared a months-long document-sending backlog over a weekend, while another handles system steps for tens of thousands of consults each day.[2] Cigna Healthcare says computer use can bring relevant information together in seconds and reduce time spent searching.[2]
For healthcare professionals, the announcement points to a way to connect AI assistance with existing systems while keeping selected steps subject to human review. Teams assessing it should weigh the reported evaluation results alongside access, security, retention, and workflow requirements.
Why it matters
AI training for your team
kju.ai is built for teams: daily sessions, team analytics, and rollout support for your organization.
How this developed
20 November 2026
OpenAI adds agents and plugins to ChatGPT for Healthcare