OpenAI for Healthcare aims to streamline clinical workflows and reduce administrative burden

As OpenAI for Healthcare, the company’s latest addition artificial intelligence (AI), deployed in leading healthcare organizations nationwide, experts emphasize the need for careful adoption to meet ethical concerns and bias.

New OpenAI solutions aim to support clinicians and improve care delivery

Last Thursday, OpenAI introduced OpenAI for Healthcare, a suite of secure AI products designed to help healthcare organizations deliver higher quality care, reduce administrative burden, and power personalized clinical solutions while maintaining Health Insurance Portability and Accountability Act (HIPAA) compliance by protecting patient data.¹ Offerings include the OpenAI application programming interface (API) and ChatGPT for Healthcare, both powered by GPT-5.2 models and developed through ongoing research and real-world evaluation to reflect how clinicians use AI in practice.

The OpenAI API platform allows developers to create tools and products using the company’s latest models and integrate AI directly into healthcare systems and workflows; OpenAI noted that eligible consumers can apply for a Business Associate Agreement (BAA) to meet HIPAA compliance requirements. In practice, teams use OpenAI APIs to develop applications for healthcare team coordination, patient record summarization, and output workflows.

Meanwhile, ChatGPT for Healthcare enables clinical teams to synthesize medical evidence as well as institutional guidance and apply it to specific patient contexts. The tool can also be used to write clinical and administrative documentation and adapt patient educational materials for readability and translation. According to OpenAI, these features can help reduce time spent on administrative tasks, promote shared standards of care, and provide a better patient experience.

Responses generated by ChatGPT for Healthcare draw on millions of peer-reviewed studies, public health advice, and clinical guidelines, with citations including article titles, journals, and publication dates. OpenAI said this supports source verification and helps clinicians reason through cases with greater confidence, allowing patients to receive accurate diagnoses and begin treatment sooner.

ChatGPT for Healthcare can also integrate with enterprise tools, such as Microsoft SharePoint, allowing responses to incorporate a facility’s approved policies, clinical pathway and operational guidance. Additionally, the platform supports shared templates for common tasks, such as writing discharge summaries, patient instructions, and supporting prior authorizations; this helps teams spend less time rewriting and researching while improving care transitions and clarity for patients.

The platform also provides a centralized workspace with role-based access controls and organization-wide user management, providing healthcare organizations with better governance and visibility when deploying AI across clinical, research and administrative teams. Finally, patient data and protected health information remain under the control of each organization, with data residency options, audit logs, customer-managed encryption keys, and a BAA with OpenAI to support HIPAA-compliant use. OpenAI stressed that content shared with ChatGPT for Healthcare is not used to train its models.

ChatGPT for Healthcare became available last Thursday and is already rolling out to major healthcare settings, including AdventHealth, Memorial Sloan Kettering Cancer Center and Cedars-Sinai Medical Center.

“At AdventHealth, we are working with OpenAI to apply AI in ways that directly benefit our consumers: reducing friction, improving access to clear, timely information, and giving our care teams more time to focus on what matters most: providing whole-person care with compassion and consistency,” Rob Purinton, AdventHealth’s chief AI officer, said in a press release.

ChatGPT Health helps patients navigate care with personalized information

The launch of ChatGPT for Healthcare came a day after the introduction of ChatGPT Health, a consumer tool designed to combine health information with ChatGPT capabilities to help patients feel more informed, prepared and confident as they navigate their care.² OpenAI reported that more than 230 million people around the world ask health and wellness-related questions on ChatGPT every week. Building on this interest, ChatGPT Health personalizes responses using patients’ health information and context and can securely connect to medical records and wellness apps. These features can help users understand recent test results, prepare for doctor visits, receive advice on diet and exercise, and evaluate insurance options based on individual health care patterns.

“(ChatGPT) Health is designed to support, not replace, medical care,” OpenAI said in a press release. “It is not intended for diagnosis or treatment. Instead, it helps you answer everyday questions and understand trends over time – not just moments of illness – so you can feel more informed and prepared for important medical conversations.”

OpenAI encourages patients interested in ChatGPT Health to join the waiting list. The company is initially providing access to a small group of early adopters while continuing to refine the tool, with plans to expand availability to all users on the web and iOS in the coming weeks.

Balancing innovation and accountability when integrating AI into healthcare

In recent years, experts have weighed the potential benefits and harms of integrating AI into healthcare as its use becomes more widespread. Jason Spangler, MD, MPH, FACPM, then president and CEO of the Center for Innovation & Value Research, and John Nosta, president of NostaLab, explored both ends of the spectrum in a 2024 contributor article published on AJMC.com.³

The experts highlighted “exciting areas” where AI shows promise, including in the generation and interpretation of evidence. Traditionally, evidence is gathered through clinical trials or large-scale observational studies, which can take years and cost millions of dollars. AI, however, has the potential to streamline this process, enabling faster and more accurate evidence generation.

They particularly highlighted the role of large language models in improving perspective and data analysis. In the context of evidence generation, the researchers explained that AI can learn from each piece of information, refining its approach over time, which could revolutionize future clinical trial designs.

“Imagine trials that dynamically adapt to new data inputs, changing protocols or endpoints in real time based on statistically robust predictive information from AI,” Spangler and Nosta wrote. “The implications for personalized medicine are profound, offering the possibility of truly individualized treatment plans based on an ever-changing data set. »

Despite these benefits, experts have emphasized that the adoption of AI in healthcare must be approached with caution, particularly due to ethical challenges. Bias in AI algorithms is a major concern, as faulty systems can directly impact patient outcomes. They explained that using biased data or poorly designed algorithms can inadvertently reinforce existing health disparities. For this reason, Spangler and Nosta emphasized that integrating AI into healthcare systems requires thoughtful regulation, rigorous validation, and ethical oversight.

“The broader story of AI’s role in healthcare is not just about revolutionizing systems, but ensuring that this revolution leads to a more equitable, transparent and patient-centered future,” the experts concluded. “The promise is immense, but delivering on that promise requires a delicate balance between innovation and accountability. »

References

1. Introducing OpenAI for Healthcare. Press release. OpenAI. January 8, 2026. Accessed January 12, 2026. https://openai.com/index/openai-for-healthcare/
2. Presentation of ChatGPT Health. Press release. OpenAI. January 7, 2026. Accessed January 12, 2026. https://openai.com/index/introducing-chatgpt-health/
3. Spangler J, Nosta J. Contributor: The promise of AI in healthcare. AJMC^®. September 30, 2024. Accessed January 27, 2026. https://www.ajmc.com/view/contributor-the-promise-of-ai-in-healthcare