On September 1, 2026, OpenAI announced that healthcare organizations can now connect their electronic health record systems and other data sources directly to ChatGPT, enabling clinicians to query patient records and public health data via Epic integration and a Healthcare Public Data plugin.
Short answer: On September 1, 2026, OpenAI announced that healthcare organizations can now connect their electronic health record systems and other data sources directly to ChatGPT, enabling clinicians to query patient records and public health data via Epic integration and a Healthcare Public Data plugin.
On September 1, 2026, OpenAI announced that healthcare organizations can now connect their electronic health record systems and other industry data sources directly to ChatGPT. The release includes a new Epic integration that brings authorized patient information from Epic environments into ChatGPT for Healthcare, as well as a Healthcare Public Data plugin that provides structured access to nine official public datasets such as PubMed, DailyMed, CMS Coverage, ClinicalTrials.gov and RxNorm. Together these tools let clinicians ask questions about a patient’s recent changes, lab results, medication adjustments or follow-up needs and receive a summarized answer that points back to the original chart entries.
The Epic integration offers two complementary ways to use the AI. In the first approach, users pull authorized patient context from the EHR into ChatGPT to review history, spot trends and prepare for appointments. In the second approach, ChatGPT can be embedded directly into the EHR layout so that clinicians can interact with the model without leaving the patient chart. A pilot partner, UCSF Health, said the integration helps clinical teams quickly grasp what has changed across a complex record, cuts down the time spent synthesizing data and creates more opportunity for face-to-time with patients. Frontline staff are being involved to validate the usefulness of the features in real workflows and to shape where they add the most value.
Beyond patient records, the Healthcare Public Data plugin lets teams work with specific fields, identifiers and versions from trusted sources. A research group might use ClinicalTrials.gov to locate actively recruiting studies and compare eligibility criteria side by side. A pharmacy team could check DailyMed for the most recent label and safety warnings on a medication. A population health group designing a diabetes-prevention program could bring together research articles, active trial information and Medicare coverage details in a single source-backed view, making it easier to evaluate options and spot remaining questions.
To ensure the model behaves reliably when handling this connected data, OpenAI partnered with hundreds of physicians from 60 countries, speaking 49 languages and representing 26 medical specialties. These clinicians have already reviewed more than 700,000 model responses that reflect everyday healthcare questions, using their feedback to improve the system’s behavior. For the EHR-focused capabilities, physicians evaluated responses across 27 clinical scenarios such as pre-visit reviews, medication checks, timelines and handoff summaries. Out of 4,363 ratings, 99.1 percent were judged safe across all use cases. In a separate two-round assessment, reviewers examined hundreds of ChatGPT answers to nuanced clinical questions drawn from large U.S. health datasets. For each of the five data sources tested, more than 93 percent of the responses received a rating of good or better accuracy.
The same governed workspace that protects healthcare information also supports broader business uses. Clinical and business teams can turn insights into reports, analyses, presentations and plans with ChatGPT Work, while technical teams can employ Codex to build or refine software that aids care delivery and operations. Approved plugins for platforms like Microsoft SharePoint, Google Drive, Salesforce and Slack expand the available business context without overriding existing permissions. At AdventHealth, the chief AI officer explained that putting tools such as ChatGPT Work, Codex and connected business data into the hands of staff helps reduce routine tasks and speeds the translation of practical innovations into action, ultimately giving caregivers more time for the human connection that underpins whole-person care.
Getting started depends on the type of account. Existing ChatGPT for Healthcare customers should ask their workspace administrator to turn on the Epic integration and the Healthcare Public Data plugin. ChatGPT Enterprise customers need to reach out to their OpenAI account team to confirm eligibility and the proper setup for a Regulated Workspace. Individual clinicians who qualify for the U.S. ChatGPT for Clinicians offering can install the Healthcare Public Data plugin on their own, although the Epic integration remains unavailable for personal accounts. With a valid Business Associate Agreement in place, all of these features can be used in a HIPAA-compliant environment.
It lets them pull authorized patient data from Epic EHR into ChatGPT to review history, spot trends, and prepare for appointments, or embed ChatGPT directly into the EHR layout so clinicians can interact without leaving the chart.
The plugin provides structured access to nine official sources including PubMed, DailyMed, CMS Coverage, ClinicalTrials.gov, and RxNorm, enabling teams to retrieve specific fields, identifiers, and versions-for example, locating recruiting trials, checking medication labels, or combining research articles, trial data, and Medicare coverage for diabetes-prevention planning.
Physicians from 60 countries reviewed over 700,000 responses; for EHR-focused scenarios, 99.1 % of 4,363 ratings were judged safe across all use cases, and for five data sources tested, more than 93 % of answers received a good or better accuracy rating in a two-round assessment.
Existing ChatGPT for Healthcare customers should ask their workspace administrator to turn on the Epic integration and the Healthcare Public Data plugin to begin using the features.
Source: OpenAI
On September 1, 2026, Google DeepMind introduced an agentic video-understanding feature for Gemini models that reduces token usage by up to 88%, cuts operating costs by up to 66%, and improves output quality by up to 7% across Gemini 3.7 Flash, 3.6 Flash and 3.5 versions.
On August 31, 2026, Instagram announced it is tightening rules on accounts that present AI-generated personas as real people, replacing the "AI creator" label with a clearer "AI-generated profile" tag and reducing the reach of profiles that do not display the label.
On August 31 2026 the European Commission designated OpenAI’s ChatGPT, Reddit and Roblox as very large platforms under the EU’s Digital Services Act after each surpassed 45 million monthly users in the bloc, triggering stricter content-removal, minor-protection and algorithm-transparency obligations and fines of up to 6 % of global turnover.
NO COMMENTS YET
Comments are open. Have a thought or a question? Share it below.