S-4916-119
Read twice and referred to the Committee on Health, Education, Labor, and Pensions.
Sponsored by Mark Kelly (D-AZ)
What it does
This bill would direct the Secretary of Health and Human Services to contract with the National Academies of Sciences, Engineering, and Medicine to conduct a study on how AI-enabled systems — including chatbots, voice assistants, health monitors, and robotic devices — affect adults aged 65 and older. The study would examine both benefits (such as reducing social isolation and supporting caregiving) and risks (such as fraud, financial exploitation, and health misinformation). The bill would also add a new provision to the Public Health Service Act directing the National Institute on Aging to facilitate ongoing multidisciplinary research and disseminate best practices on AI use by older adults, with findings reported to Congress within one year. Funding is capped at $2 million from existing HHS appropriations for fiscal years 2027–2028.
Who benefits
Adults aged 65 and older, who would gain a formal evidence base informing safer AI use. Family members and unpaid caregivers of older adults, whose role in AI-assisted caregiving would be studied. Geriatricians, psychologists, and other clinicians who treat older patients and would receive research-backed guidance. Older adults with disabilities or cognitive impairments, whose specific accessibility needs would be explicitly examined. Fraud prevention organizations and consumer protection agencies (e.g., the FTC) that could use findings to target elder financial exploitation. AI developers and technology companies seeking clarity on best practices for designing products for older users. Researchers and academics in aging and technology fields who would benefit from federally coordinated research priorities.
Who is hurt
AI companies whose products are found to pose risks to older adults could face reputational harm or future regulatory pressure based on study findings. Taxpayers would bear the cost of up to $2 million in redirected HHS funds. Older adults in underserved or rural areas may find that study recommendations do not translate into accessible products or services. Competing research institutions not selected for the contract would lose a funding opportunity. To the extent the study's findings prompt future regulation, AI developers serving the elder care market could face compliance costs.
Supporters argue
Supporters argue that the 65-and-older population is the fastest-growing demographic in the United States and is increasingly targeted by AI-enabled scams and financial fraud, with the FBI reporting that older adults lost over $3.4 billion to fraud in 2023 alone. They contend that no comprehensive federal evidence base currently exists on how AI systems affect this population across cognitive, behavioral, and social dimensions, and that a National Academies consensus study — the gold standard for independent scientific review — is the appropriate first step before any regulatory action. Supporters also argue the bill's bipartisan sponsorship and narrow, research-only scope make it a low-cost, high-value foundation for evidence-based policymaking.
Opponents argue
Opponents argue that the bill authorizes a study without any binding follow-up requirements, meaning findings could be shelved without producing concrete protections for older adults — a pattern critics call "study and delay." They contend that existing agencies such as the FTC, the Consumer Financial Protection Bureau, and the National Institute on Aging already have authority and ongoing programs to address elder fraud and AI risks, making a new $2 million contract duplicative. Opponents may also argue that the one-year study timeline is too slow given the rapid pace of AI deployment, and that Congress should move directly to regulatory action rather than commissioning additional research.