
What happened
OpenAI introduced an Epic EHR integration for ChatGPT for Healthcare, letting clinicians ask questions about patient history, labs, and medications. It also launched a Healthcare Public Data plugin connecting to nine official sources like PubMed, DailyMed, and CMS Coverage.
Why it matters
In physician evaluations across 27 clinical use cases, 99.1% of 4,363 responses were rated safe. For five connected data sources tested, over 93% of responses were rated 'good' or better accuracy, suggesting the tool can reliably support clinical tasks.
What to watch
The integration is available to ChatGPT for Healthcare and Enterprise customers, but not individual accounts. Watch whether healthcare organizations adopt it and whether the high safety ratings hold in real-world use.
Summaries like this, in your inbox every morning.
OpenAI is expanding ChatGPT for Healthcare beyond simple question answering. By integrating with Epic, a major EHR system, the tool can pull in authorized patient data directly into a chat interface. This move reduces the need to search across separate systems, potentially saving time for clinicians. The new Healthcare Public Data plugin aims to make it easier to access up-to-date medical information from trusted sources without toggling between databases.
The safety and accuracy results are a critical step for building trust with healthcare organizations. With 99.1% of physician-rated responses deemed safe and over 93% accuracy across five data sources, there is evidence the system performs well in controlled tests. However, these metrics come from evaluations designed with physician input, not live clinical use. The true test will be how the tool handles real-world patient cases and whether it can consistently support decisions without errors.
As with many AI health tools, adoption depends on governance and trust. OpenAI is addressing this by offering enterprise controls like role-based access and HIPAA-compliant workflows. Whether hospitals and clinics embrace this integration will hinge on how well it fits existing workflows and how confidently clinicians can rely on its outputs. The coming months will likely reveal whether the high safety ratings translate into practical value.
Pick your industry and the AI tools you use, and get news related to your work every day.
Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.
ElevenLabs completed a $300 million employee tender at a $22 billion valuation, and co-founder Mati Staniszews…

On October 15, OpenAI announced the Rosalind Academic Plan, giving approved US university labs free access to…

A Qiita article compared automated red-teaming methods for LLM apps — GCG, PAIR, and TAP — using mock implemen…

Liquid AI released d1-ultra on October 15, 2026, an upgraded open-weights decision model that averaged 76.7 ac…

Circlia started a proof-of-concept for TASTE SIGNAL: visitors to Holland Village Cafe FUKUOKA taste cookies an…

In a webinar hosted by French Mathematical Society president Isabelle Gallagher, Fields Medal winner Hugo Dumi…
