The U.S. Department of Health and Human Services (HHS) is accelerating efforts to integrate artificial intelligence (AI) into medical care, aiming to transform diagnosis and treatment through autonomous AI systems. This multifaceted initiative involves federal agencies such as Medicare and Medicaid Services (CMS), the Food and Drug Administration (FDA), and the Advanced Research Projects Agency for Health, and has garnered significant attention for its potential to expand access to healthcare and reduce costs.
Vinod Khosla, a Silicon Valley venture capitalist with personal ties to the healthcare AI sector through his son Neal Khosla’s company, Curai Health, has played a prominent role in discussions with top HHS officials. He envisions a future in which AI largely replaces human doctors in providing primary care, describing the technology as the “holy grail” for improving the healthcare system. Curai Health is among several companies participating in Medicare pilot programs designed to reimburse AI-based medical services, which currently cover over 60 million Americans. These pilots also include projects like Limbic, which offers AI-driven cognitive behavioral therapy for Medicare beneficiaries with depression and anxiety.
Despite enthusiasm among some government officials, the rapid adoption of AI in healthcare has raised concerns among medical professionals and regulatory experts. Dr. John Whyte, CEO of the American Medical Association (AMA), cautions that AI models designed to independently assess and treat patients are "not ready for prime time." He argues that healthcare AI development should be led by clinical experts rather than the technology industry, warning against the disproportionate influence of Silicon Valley investors in shaping federal health policy.
The FDA, responsible for regulating medical devices, has authorized over 1,500 AI-based tools for specific diagnostic tasks such as tumor detection. However, newer, more autonomous AI systems—like chatbots that can prescribe medications—do not yet fit within existing regulatory frameworks. The agency is actively reviewing how to update its oversight to account for these evolving technologies, especially in light of risks such as incorrect or harmful medical advice generated by some AI models.
Internal tensions have surfaced within HHS regarding oversight and regulation. Jared Seehafer, recently appointed as deputy commissioner for technology and AI, advocates for centralizing AI governance within senior leadership. This approach has reportedly been met with resistance from career officials focused on patient safety, including Dr. Michelle Tarver, head of the medical devices division. Meanwhile, Robert F. Kennedy Jr., the health secretary, has publicly acknowledged both the risks and transformative potential of AI in medicine.
Pilot programs under Medicare aim to evaluate AI performance and cost-effectiveness before broader implementation. For example, the Advocate project, funded with about $60 million, seeks to deploy AI-supported care for heart failure patients by remotely monitoring health data and adjusting treatments under physician supervision. Similarly, Curai Health’s autonomous AI primary care system is undergoing FDA review and Medicare pilot testing. Preliminary studies show clinicians agreed with its diagnoses in the majority of cases, though some disagreements highlight lingering safety concerns.
The debate over AI in healthcare reflects a broader conversation about balancing innovation, patient safety, and regulatory responsibility. While proponents emphasize AI’s ability to expand access to underserved populations and reduce system burdens, critics urge caution, emphasizing the importance of clinical judgment and the need for evidence-based validation. As deployment progresses, stakeholders continue to grapple with how to ensure that AI technologies deliver measurable benefits without compromising patient care.