For generations, healthcare has often been reactive: symptoms appear, a patient seeks care, tests and referrals follow and treatment begins. A wearable can now flag an irregular heartbeat before the first symptom appears. AI can help identify subtle signs of lung cancer on a chest X-ray. Advances in early detection could one day make late-stage cancer diagnoses less common by finding disease when it is more treatable and survivable.
This is the promise of AI-enabled healthcare: moving medicine from reacting to disease to predicting and preventing it. At the same time, AI is changing how consumers navigate their healthcare: 41% of consumers are already incorporating AI and/or social media into their search for healthcare information, and that rises to 57% of Gen Z and 53% of millennials.
But adoption does not equal trust. Among consumers who use open web-based digital assistants for health information, only 33% say they are very satisfied with these tools. Additionally, 85% verify the health information they find on any AI or social media tool they use. To realize the transformative benefits AI innovations make possible, it’s important to ensure AI systems are safe, trustworthy and transparent. Earning trust requires strong safeguards, smart national oversight and policies that allow responsible innovation.
As AI becomes more deeply embedded in both healthcare delivery and consumer actions, getting the rules right becomes increasingly urgent. Policymakers can create a consistent, risk-based federal AI framework that protects patients while giving innovators room to develop and deploy lifesaving technologies, or they can allow a fragmented patchwork of state laws to limit access to better care.
This year alone, more than 240 health AI bills have been introduced across 43 states. AI doesn’t stop at state lines. These bills could force companies to navigate hundreds of different definitions, disclosure rules and oversight requirements. Without a consistent national framework for AI regulation, this growing patchwork of bills will slow the development and deployment of solutions that can expand access, diagnose diseases earlier and save lives.
Delayed access has real consequences. Today, more than 77 million Americans live in areas with a shortage of primary care providers and more than 122 million live in places without enough mental health professionals. Digital health tools, remote monitoring and AI-enabled diagnostics technology are critical in expanding access to healthcare and early detection.
What we need now is risk-based, smart oversight that supports innovation. Unlike many traditional medical technologies, AI can continue to evolve after deployment, but our federal regulatory system is built for medical interventions that remain the same once approved—like a drug, vaccine or pacemaker.
But approval is not the finish line. Post-market monitoring means watching for declining performance, unintended consequences and changes that affect how well a tool works in clinical care once it is in use. It means tailoring that scrutiny to the application’s risk and to the vulnerability of the target population, whether pediatric or geriatric patients. It means tracking performance by age and demographic group, because generalized metrics can conceal failures in specific subgroups.
That’s the focus of a new standard created in a consensus-based process with more than 75 organizations. It provides risk-based guidance to help organizations monitor performance, detect model drift, and require greater transparency around performance. Voluntary standards show what workable oversight looks like and federal policy should build on.
Trust also depends on the health data that is powering these technologies. We need modernized, nationally consistent privacy protections that match oversight to actual risk. Currently, there are at least 20 different state laws, creating a system in which patients’ rights can change depending on where they live. Washington, Nevada and Connecticut have each passed laws that require specific opt-in consent before using their health data. Most patients assume this is covered by HIPAA, but it isn’t. Patients and their doctors also need to be able to securely access and share health information across different devices, apps and healthcare systems.
But technology only improves healthcare if patients can access it and there is a clear path to reimbursement. Today, nearly 60% of providers say that digital health solutions help to ease the burden of an overstretched healthcare system. That impact diminishes if the tools are not covered by insurance.
Over the past two years, Congress has watched state regulations grow increasingly complex and fragmented. Every delay in federal action creates more uncertainty for the companies building AI-enabled health tools and more barriers for the patients and providers who could benefit from them. Health AI is advancing rapidly and our policy framework must advance with it. Smart regulation should protect patients while allowing responsible innovation to scale, bringing us closer to a healthcare system that predicts and prevents disease rather than simply reacting to it.
Kinsey Fabrizio is president and CEO of the Consumer Technology Association.