The healthcare industry is experiencing a transformation through agentic AI—artificial intelligence systems that can work independently to solve problems and complete tasks with minimal human oversight. This week revealed several breakthrough implementations of these powerful tools across different areas of healthcare and different countries.

The most significant development came from the United States, where the FDA launched a secure, agentic AI platform available to all agency employees. This represents a major shift in how government agencies approach their work. Rather than just using AI for simple tasks, the agentic AI system can handle complex workflows independently. FDA staff members can now rely on this intelligent assistant to manage multiple responsibilities simultaneously. The platform demonstrates that even large government institutions are ready to trust agentic AI with important responsibilities.

Cancer care in the United States is being revolutionized by a new partnership between GE HealthCare and Mayo Clinic, called GEMINI-RT. This project uses agentic AI to personalize cancer treatment for each patient individually. The concept is called 'twinning the patient, personalizing the beam'—essentially creating a digital copy of each patient's situation so AI agents can recommend the most effective radiation therapy. Cancer is a massive problem globally, with 19.3 million new cases reported in 2022 alone. In the United States, over 2 million patients annually receive radiation therapy. The challenge has always been that every patient is different, and what works for one person might not work for another. The GEMINI-RT agentic AI system uses patient imaging, artificial intelligence, and real-time monitoring to continuously adjust and improve treatment recommendations based on how patients are responding.

Microsoft Research contributed to the agentic AI revolution by releasing GigaTIME, an open-source, multi-modal AI system that can work across different types of medical information. This agentic AI can examine pathology slides—tiny samples of tissue viewed under microscopes—and automatically convert them into virtual images. The system then independently analyzes these images against patient medical records from Providence Healthcare. This creates what researchers call a 'virtual population' that agentic AI agents can study to understand connections between cell conditions and important health markers. Researchers don't have to manually compare each slide to each patient's records; the AI agents handle this automatically and find patterns humans might miss.

Medical device innovation reached a milestone when CeriBell, Inc. received FDA clearance for its next-generation Clarity algorithm—an agentic AI system designed to detect seizures in newborns and patients of all ages. This is particularly important because approximately 9% of patients in newborn intensive care units (NICUs) develop seizures, yet research suggests up to 90% go completely undetected without special monitoring. When high-risk newborns have seizures lasting more than 13 minutes per hour, their chances of serious problems increase 8 times over. The agentic AI system works continuously and independently to monitor brain activity and alert medical teams to seizure activity that human observers might miss. The FDA clearance was supported by data from over 700 patients—the largest validation dataset ever used for a neonatal seizure detection system.

The United Kingdom has been implementing large-scale agentic AI systems through its NHS Federated Data Platform, which uses intelligent AI agents to help hospitals work more efficiently. This system currently operates in 77 NHS Trusts (hospital networks), with 73 more signed up and 41 additional Integrated Care Boards planning to join. The results have been impressive: the agentic AI platform has helped complete an additional 80,000 operations that might not have happened otherwise and reduced discharge delays by 15%. The United Kingdom government predicts that by the end of this decade, this agentic AI system will deliver £150 million in annual benefits. Palantir Technologies has partnered with Multiverse to train NHS staff on these agentic AI systems, with the first training groups beginning in February 2026. This training helps healthcare workers understand how to work alongside AI agents effectively.

These developments show that agentic AI is moving from experimental projects into real-world healthcare delivery worldwide. From the FDA in the United States to the NHS in the United Kingdom to cancer centers and research institutions, healthcare providers are successfully deploying AI agents that work independently to improve patient care, detect problems earlier, and help healthcare teams accomplish more with their limited time and resources.