OpenAI has added an Epic EHR integration and a public health data plugin to ChatGPT for Healthcare, with San Francisco-based UCSF Health serving as a pilot partner, confirming a significant ChatGPT for Healthcare Epic integration expansion announced Sept. 1.
How the ChatGPT for Healthcare Epic Integration Actually Works
The Epic integration lets authorized clinicians at healthcare organizations with supported Epic environments pull together patient information, including appointment notes, lab results, medications, and specialist documentation, inside ChatGPT or directly within Epic workflows, OpenAI told Becker’s on Sept. 1. The company said the integration is read-only and does not write information back to the patient record, supporting two use cases: reviewing authorized patient data within ChatGPT, and embedding ChatGPT directly into an EHR layout for in-chart AI assistance.
Why the Read-Only Design Matters
Keeping the integration strictly read-only, with no ability to write information back into the patient record, addresses a critical safety boundary for AI tools operating within clinical documentation systems, ensuring the technology can synthesize existing information without introducing a pathway for AI-generated errors to enter the official medical record.
UCSF Health’s Role Within This ChatGPT for Healthcare Epic Integration Pilot
“As a pilot partner, we’re exploring how the new EHR integration with ChatGPT for Healthcare can help clinical teams understand what has changed and what matters most across a complex patient record,” UCSF Health President and CEO Suresh Gunasekaran said in an OpenAI blog post. He said the technology has the potential to reduce time spent synthesizing data and give clinicians more time with patients.
Beyond UCSF’s Pilot Status
UCSF Health is OpenAI’s featured pilot customer for the integration, and other health systems are exploring the tool, according to an OpenAI spokesperson, indicating the company anticipates broader adoption beyond this initial featured partnership as more health systems evaluate the capability.
The Healthcare Public Data Plugin Alongside This ChatGPT for Healthcare Epic Integration
The Healthcare Public Data plugin connects ChatGPT to nine official public sources, including ClinicalTrials.gov, CMS Coverage, RxNorm, DailyMed, and PubMed. OpenAI said physicians rated 99.1% of ChatGPT responses safe across 4,363 ratings spanning 27 clinical use cases involving connected EHR context.
How Accuracy Varied Across Different Data Sources
Accuracy ratings for the public data connectors ranged from 93.2% for CMS Coverage to 98.6% for DailyMed, a notable spread suggesting that not all nine connected public data sources currently perform with equal reliability, with coverage-related information showing comparatively lower accuracy than medication reference data.
Governance and Compliance Within This ChatGPT for Healthcare Epic Integration
Administrators control which integrations are enabled, and existing user permissions govern what information each person can access. ChatGPT for Healthcare includes role-based access controls, single sign-on, and audit logs, and customers with an applicable business associate agreement can use ChatGPT Work, Codex, apps, and connectors in the same workspace to support HIPAA-compliant workflows.
How This Builds on OpenAI’s Broader Healthcare Footprint
The update builds on ChatGPT for Healthcare’s existing clinical search tool, which draws from thousands of medical sources, as OpenAI continues expanding its footprint among hospitals and health systems following earlier launches of ChatGPT for Clinicians and ChatGPT Health.
What This ChatGPT for Healthcare Epic Integration Means Going Forward
With UCSF Health serving as the initial featured pilot and other health systems already exploring the tool, this Epic integration positions ChatGPT for Healthcare to compete more directly with Epic’s own in-house AI capabilities for synthesizing complex patient records at the point of care. Given the read-only design and layered governance controls, including role-based access and audit logs, health systems evaluating this integration may find the compliance framework addresses many of the data security concerns that typically slow enterprise AI adoption in clinical settings.
What to Watch Going Forward
As more health systems beyond UCSF Health begin exploring this Epic integration, industry observers will likely watch whether the accuracy gaps seen across different public data connectors, ranging from 93.2% to 98.6%, narrow as OpenAI continues refining the underlying data sources. Given OpenAI’s continued expansion following ChatGPT for Clinicians and ChatGPT Health, this ChatGPT for Healthcare Epic integration may represent another significant step in the broader competitive dynamic between OpenAI’s healthcare AI products and Epic’s own expanding AI portfolio for health systems already running the dominant EHR platform.
For more healthcare industry updates, insights and news, visit DistilINFO. Click here to subscribe to stay informed.
