HR-10210-119
Referred to the Committee on Energy and Commerce, and in addition to the Committees on Ways and Means, and Education and Workforce, for a period to be subsequently determined by the Speaker, in each case for consideration of such provisions as fall within the jurisdiction of the committee concerned.
Sponsored by Greg Landsman (D-OH)
What it does
The bill would require group health plans and insurers (group and individual markets) to ensure that any denial of benefits involving clinical judgment, at first review or on appeal, is not issued or dictated by an artificial intelligence system. Such denials would have to be made by a licensed professional with relevant training who conducts an independent evaluation, and denial notices would have to disclose any AI use and the reviewer's name and credentials. The bill would also treat AI use in utilization review as a treatment limitation under mental health parity rules, with added reporting to the Secretary. It would apply to plan years starting on or after January 1 of the first calendar year beginning at least 12 months after enactment, and it amends the Public Health Service Act, ERISA, and the Internal Revenue Code.
Who benefits
Patients and plan enrollees facing coverage denials, who would get review by a licensed clinician, notice of AI use, and access to records. Patients with mental health or substance use conditions, whose benefits would get added parity scrutiny of AI tools. Physicians and other licensed reviewers, who may see more demand for clinical review work. Providers who appeal denials and could use the new documentation.
Who is hurt
Health insurers and self-funded employer plans, which would bear costs for added clinician reviewers, documentation, and compliance systems. Vendors of AI-based utilization review tools, which may lose business. Plan sponsors and enrollees who might see higher administrative costs passed into premiums. Smaller plans that may find compliance harder, and enrollees who could face slower determinations if clinician review capacity is limited.
Supporters argue
Supporters argue that coverage denials based on medical necessity are medical decisions and should rest on a licensed clinician's independent judgment, not an algorithm's output. They contend that reported cases of automated denial tools and high rates of reversal on appeal show a need for human review, and that disclosure and documentation requirements give patients and regulators a way to check whether the rules are followed.
Opponents argue
Opponents argue that AI tools help insurers process large claim volumes quickly and consistently, and that barring AI from driving denials could add cost and delay that may raise premiums. They contend the broad definition of AI, which covers any algorithm or software, could sweep in routine claims-processing tools, and that existing appeal and external review rights already protect patients from improper denials.
Constitutional context
Congress would regulate insurers and employer-sponsored plans under the Commerce Clause and its taxing power through the Internal Revenue Code, as in NFIB v. Sebelius (2012), which upheld federal authority over existing insurance activity. ERISA preemption of state law and the Supremacy Clause would govern how the rules interact with state insurance regulation; no unusually contested constitutional question is apparent.
Checks and balances
The bill would extend regulatory authority to the executive agencies that administer these laws (HHS, Labor, Treasury), including parity oversight by the Secretary, while courts review enforcement and Congress retains oversight.
Historical precedent
The Mental Health Parity and Addiction Equity Act and ACA claims and appeals rules under PHSA section 2719 similarly regulated how plans make and review coverage determinations, and this bill builds on both.