OpenAI Connects Epic EHR Data and Public Health Sources to ChatGPT

OpenAI has launched an Epic electronic health record integration and a Healthcare Public Data plugin, bringing authorized patient context and nine official healthcare datasets directly into ChatGPT for Healthcare.

The move puts ChatGPT inside one of the most controlled information environments in the software industry. Patient records, medical evidence, and public healthcare data have traditionally lived in separate systems, and OpenAI is betting that a governed workspace connecting all three will attract healthcare organizations that have so far approached clinical AI cautiously.

What the Epic integration does

Healthcare organizations can now connect their Epic environments to ChatGPT. Instead of searching across appointment notes, laboratory results, medications, and specialist documentation, clinicians can ask questions like: What has changed since this patient's last visit? Which recent lab results should they review before today's appointment? ChatGPT then pulls together relevant information from the authorized patient record, summarizes important developments, and points back to supporting chart information.

The integration supports two complementary experiences. The first, "EHR context in ChatGPT," brings authorized patient information from a supported EHR into the chatbot to review patient history, identify changes, and prepare for appointments. The second, "ChatGPT in the EHR workflow," embeds ChatGPT directly into an EHR layout in supported deployments, enabling AI-assisted workflows without leaving the patient chart.

The EHR integration is not available for individual accounts. ChatGPT for Healthcare customers need to ask their workspace administrator to enable it, while ChatGPT Enterprise customers must contact their OpenAI account team to confirm eligibility and the right configuration for their Regulated Workspace.

Nine official public data sources

The second capability, the Healthcare Public Data plugin, gives teams direct, structured access to official healthcare datasets. It includes dedicated connectors to nine sources: ClinicalTrials.gov, CMS Coverage, RxNorm, DailyMed, PubMed, and others. Teams can work with specific records, fields, identifiers, and versions while focusing the task on these authoritative sources.

OpenAI says this makes it easier to compare and verify precise information — trial eligibility criteria, medication identifiers, coverage policy versions, provider records — without searching each source separately. A research team could use ClinicalTrials.gov to identify actively recruiting trials and compare eligibility criteria. A pharmacy team could use DailyMed to confirm the latest label and warnings for a medication. A population health team planning a diabetes-prevention program could combine research, active trials, and Medicare coverage information in a source-backed view.

The plugin builds on clinical search already available in ChatGPT for Healthcare, which the company says spans thousands of medical sources. Eligible U.S. users of ChatGPT for Clinicians, the individual-tier product, can also install the Healthcare Public Data plugin.

The evaluation numbers

OpenAI backs the release with two evaluation rounds. According to the company, physicians rated 99.1% of responses safe across all use cases, based on 4,363 ratings covering 27 clinical use cases that include pre-visit review, clinical timelines, medication review, and handoff summaries. Those tests specifically measured ChatGPT's performance when working with connected EHR context.

In a separate two-round evaluation, physicians reviewed hundreds of ChatGPT responses to nuanced clinical questions based on large U.S. healthcare datasets. For each of the five connected data sources tested, more than 93% of responses were rated as having "good" or better accuracy.

The ratings come from a physician network OpenAI describes as a core part of its healthcare effort. The company partners with hundreds of physicians across 60 countries, 49 languages, and 26 medical specialties to define, measure, and improve health responses in ChatGPT. To date, these physicians have reviewed more than 700,000 model responses across examples that reflect real-world healthcare questions.

OpenAI frames these evaluations as measuring how ChatGPT handles healthcare context end to end — surfacing relevant information, citing supporting evidence, and preparing work that clinicians and staff can review. The company says it continues to invest in dedicated training and evaluation for healthcare.

Governance and compliance

The regulatory stakes are central to the product design. ChatGPT for Healthcare combines healthcare-specific capabilities with enterprise controls including role-based access, single sign-on, and audit logs. With an applicable Business Associate Agreement, customers can use ChatGPT Work, Codex, apps, and plugins in the same workspace to support HIPAA-compliant workflows.

The same workspace extends beyond clinical use. Clinical and business teams can use ChatGPT Work to turn connected information into reports, analyses, presentations, and plans for review. Technical teams can use Codex to build and improve software supporting care delivery and business operations. Plugins for Microsoft SharePoint, Google Drive, Salesforce, Slack, and other enterprise systems expand the approved business context available in ChatGPT while preserving existing permissions, OpenAI says.

Why it matters

Epic dominates the U.S. hospital EHR market, and an official integration with the vendor gives OpenAI a distribution channel into thousands of health systems that Microsoft, Google, and other AI competitors are also courting. The 99.1% safety rating and the physician review network signal OpenAI's effort to address the reliability concerns that have slowed clinical AI adoption, where errors in summarizing patient records carry direct patient-safety consequences.

OpenAI positions the release as incremental by design: organizations can start with the capabilities that fit their needs and expand over time, bringing more of the information central to care and operations into one governed workspace. Whether health systems move beyond pilot deployments will depend on how the integrated workflows hold up in day-to-day clinical use — the evaluations released today cover the model's responses, not live deployment outcomes.