From deeply personal patient information to highly sophisticated insurance insights, healthcare is rich with data. As the industry at large continues to focus on individualizing care and improving the patient experience, organizations throughout the healthcare ecosystem are turning to artificial intelligence to manage all that data and automate data-driven processes.
While it’s already commonplace in healthcare to use AI for managing documentation and powering chatbot applications, some healthcare systems are unlocking even more functionality with the help of the latest AI innovation: AI agents — software applications that use AI to autonomously sense their environment, process data, make decisions and take actions in pursuit of specific goals.
Speed often is touted as the biggest untapped benefit of AI agents, according to Raihan Masroor, founder and CEO of Your Doctors Online. However, he thinks AI agents could improve the entire healthcare process by empowering decision isolation.
“We use AI to surface friction points in real time — [for prior authorization] delays, form confusion [and] even questions patients are too embarrassed to ask,” Masroor said. “For example, our benefit-check AI doesn’t just confirm coverage; it compares what patients search before they upload insurance and flags silent concerns, which has cut drop-off during intake by 18%.”
While data privacy and security often are a key focus when implementing AI agents, Masroor said one of the most important safeguards when using AI agents is tracing: Every AI-generated suggestion, routing choice or draft message should be traceable to its logic — for example, logging the model version, source data and confidence score. This creates an audit trail that prevents “black box” decisions and helps humans understand why AI agents make the choices they make.
“The pitfall isn’t AI making mistakes,” Masroor said. “It’s humans assuming it’s right. You get the upside when you treat AI like a colleague who’s fast, smart and needs a second opinion.”
Knowledge Agents for Providers and Staff
Throughout the healthcare process, providers and staff often have questions, including ones about patients, best practices for using equipment or how to conduct a seldom-used process. Because the answers to these questions often depend on the specific context or situation, manuals or documentation aren’t always helpful. AI agents in healthcare can provide personalized information based on real-time data to give staff the answers they need to provide the highest quality of care.
HyScaler specializes in building AI-powered digital transformation solutions for healthcare, including knowledge agents for providers and staff. Digital Marketing Manager Sauravk Panda said the company has implemented these assistants in hospitals and integrated the technology with their electric health record (EHR) systems. They help providers pull patient histories, lab results and clinical guidelines in real time, which reduces time spent on administrative tasks by up to 40%, according to HyScaler.
“One client uses a knowledge agent trained on internal protocols and public medical literature,” Panda said.
“During patient consultations, it provides physicians with differential diagnosis suggestions based on symptoms entered into the EHR.”
Patient Journey Agents
Patients seeking the right care and support often face roadblocks because of all the different providers involved, which creates a disjointed experience. Hindering the patient journey even further is the fact that it’s often challenging to find a single source of data from all providers to make the best possible decisions for patients.
Because he saw too many patient questions falling through the cracks, questions that were important, reasonable and often repeated, Benjamin Caplan, M.D., chief medical officer at CED Clinic, built several AI agents to fill gaps in the patient journey that exist between appointments.
When a patient visits the CED website or app, they can access an AI agent that acts as a front door to the practice, helping patients understand what the practice does and why. Upon becoming a patient, they can access the second agent, which uses data from a database of peer-reviewed literature hand-selected by Caplan to get smart, evidence-based answers to specific questions. For example, many patients seek information about nuanced care — for instance, information about cannabis medicine, where reliable guidance is still hard to come by.
“The chatbot doesn’t get tired or annoyed. It speaks clearly, and it’s there when patients are ready — whether that’s 2 p.m. or 2 a.m. And it gives people structure, reassurance and control, especially when they’re navigating something overwhelming,” Caplan said.
“Because the tool is available around the clock, in some cases even in multiple languages, patients can use it when they’re ready — not just when my schedule allows.”
Caplan said that the AI agent also improves productivity for his staff. When a patient doesn’t need to email the clinic three times to get a simple answer about dosing or product formats, his employees save time — which allows the practice to focus on more complex and meaningful work.
In turn, patients are more informed, more confident and often more engaged in their own care, Caplan noted. “I see AI as a support system — one that frees up space for the real work of connection, storytelling [and] trust,” he said. “Used that way, it actually makes medicine feel more human, not less.”
Insurance Agents
Verifying benefits and managing prior authorizations often stand between patients and the care they need. However, the process is often time-consuming and error-prone. By using AI agents, healthcare organizations and payers can seamlessly work together to get approvals completed quickly and accurately.
Emmanuel Chilengwe, founder of BioWell Space, said using AI agents to verify insurance benefits and manage prior authorizations can eliminate the insurance-related bottlenecks that often delay care.
“Companies like Olive and AKASA have developed automation tools that interface directly with payer portals to instantly verify coverage and initiate preapproval workflows — cutting processing times from days to minutes,” Chilengwe said.
Moving Healthcare Forward with AI Agents
One of the biggest concerns about using AI in healthcare is losing the human touch and compassion. Although he understands those concerns, Caplan insisted that AI will not replace the human part of healthcare.
“[AI] can’t feel,” Caplan said. “It doesn’t recognize when a patient hesitates before answering or pick up on subtle shifts in tone. That’s the part of medicine that’s harder to define — but it’s what makes patients feel truly understood. And that’s something a bot can’t replicate.”
Even so, Caplan thinks AI in the long term will play an even deeper role in healthcare by analyzing patterns across patient populations, adapting care plans in real time and catching small but meaningful changes that even experienced clinicians might overlook.
To be clear: He doesn’t think AI will replace doctors. Rather, he thinks it will function as a second brain that’s tireless, data-rich and always scanning for insights.
“The technology is only as good as the people building it and the values baked into it,” he said.
“If we treat AI like a shortcut to speed or scale, we’ll miss the point. But if we design it to support trust, to make room for presence and listening, then it becomes something far more powerful. I don’t believe AI is going to make medicine cold or robotic. I think it gives us the chance to make it more human again by lifting the noise and letting the real work of care come through.”
Jennifer Goforth Gregory is a contributing writer. Find her on X @byJenGregory.
© 2025 Nutanix, Inc. All rights reserved. For additional information and important legal disclaimers, please go here.