By Leigh Burchell

By the middle of 2025, almost every state in the country had passed or was drafting new rules for AI in healthcare. Federal agencies, too, are discussing standards for transparency and patient safety that will shape how health systems adopt and use AI. Some new laws are holding health systems to a high standard, demanding that if an algorithm is giving advice, there must be proof it doesn’t miss certain groups or make assumptions that could lead to mistakes and negative health outcomes. Additionally, Congress recently held several hearings on AI, and it is clear that policy makers are working urgently to keep up with the rapid evolution of the technology itself.

But ask any nurse or doctor and you’ll find that policy doesn’t always align with reality and, depending on how it’s written for AI, could either make delivering healthcare a little easier or a lot more tangled.
Legislation meets the real world
Why the sudden legislative interest in AI for healthcare? The most obvious answer is that because health technology, whether devices or EHR software, can have a bearing on the patient’s care and outcomes and introduce possible risk. While summarizing charts, drafting clinical notes, flagging care gaps and helping patients find answers in the middle of the night are ideas that can ease current friction points in healthcare, their application in the real-world may not be quite as straightforward. It’s no longer a test or a novelty, but something with real implications for real people.

This is where policy and its application get tricky. Some policies are clear: tell people if a chatbot is answering them or share which datasets underlay the AI logic. Others get deep into the weeds, spelling out who’s responsible if an AI bot makes a bad recommendation or influences medical decisions in the wrong direction. And although there are significant upsides to using AI in healthcare, they can’t come at the cost of patient safety or trust in the health system.

AI can make a difference in healthcare, if we can navigate the red tape
AI is already making a real and positive difference in healthcare by supporting real needs. When it pulls and summarizes a patient’s medical history before a visit, it helps a clinician know what to expect when they walk into the exam room. If the doctor doesn’t have to wade through dozens of PDFs, it saves time that can instead be used to do what matters most: have a real, human conversation with the patient. And when ambient listening takes notes, doctors can spend more time listening and engaging—things that patients not only notice, but that make a difference in the quality of care. However, with new regulations, there are questions about whether these tools have been vetted for privacy and where the protected health data really goes.

U.S. Healthcare Homepage