YourVoice.Now Summary
Your MoneyWorkers & JobsCivil LibertiesTransparency & AccountabilityCriminal Justice & Due ProcessHealth workers could turn down AI advice about your care and keep their jobs, with fines up to $769,870 for bosses who punish them.
Your Money
Employers that use AI decision software would need a written policy saying it never replaces a nurse's or doctor's own judgment. Staff could set the AI aside whenever they think a different call is better for the patient.
Health plans count as employers here, and their reviewers count as health workers. A reviewer deciding whether your treatment gets approved could reject what the AI recommends without risking their job.
Workers & Jobs
Punishing a worker for a proper override would be against the law. The banned punishments also include pay cuts, worse schedules, lost promotions, and pulled hospital privileges.
The Labor Department could fine an employer up to $76,987 for each violation, and up to $769,870 for repeat ones. The amounts would rise each year to keep pace with prices.
Each employer using the software would set up a review board that meets at least every three months. At least half the seats would go to non-managers, and the workers' union would have a place on it.
Employers would have to teach staff how the software works, when to override it, and where it can be wrong or biased. The training would also cover how the tool was built and what data it used.
Civil Liberties
Employers could not pass around records showing how often a named worker went against the AI, or data that makes the person easy to identify. They could still explain a care decision to the patient and hand over records in a lawsuit.
A worker could take the employer to court without waiting for the government to act. A court could order back pay, up to triple damages, set damages from $5,000 to $100,000, legal fees, and the job back.
Many workers sign contracts that send disputes to a private arbitrator instead of a court. Those clauses would not hold for claims under this bill.
Some work contracts bar staff from joining together in one case. For these claims, workers could still sue as a group.
Staff at state-run hospitals and clinics often cannot sue their employer. A state that takes federal money for the program would give up that protection for these claims.
A nurse or doctor who overrides the software is still answerable for the care given. The bill does not block a patient's malpractice or negligence claim.
Transparency & Accountability
Workers who report a suspected violation, ask a union for help, or testify would be shielded from payback. The protection reaches as far as discussing a possible violation with a co-worker.
Employers would have to tell health workers and their representatives that AI decision software is in use. They would also have to explain the right to override it.
A state attorney general or state privacy regulator could take an employer to court for residents. They would notify the federal agency first, and it could join the case.
Criminal Justice & Due Process
Complaints about the override policy rules would go to the Health and Human Services Department's civil rights office. They could not be sent to the Justice Department for criminal charges.
More about this bill
A nurse or doctor could say no to AI advice about your care and keep their job. The bill covers software that scores, predicts, or advises on care. Hospitals, clinics, home-care agencies, and health plans that use it would need a written policy. The policy must let staff set aside the AI when they judge that is better for you. It must also let them set it aside to follow the law. Health plan staff who review coverage requests are covered too. So a nurse checking your prior approval could reject an AI denial. Employers could not share data that shows how often one worker went against the AI. Your employer could not fire, demote, or punish you for a proper override. A second rule would protect any worker who reports a problem with the AI. It covers even a quiet talk with a co-worker about a rule that may have been broken. Employers would have to train staff on the tools and where they fall short. They would also have to say when AI is in use at work. Each employer would set up a review board that meets four times a year. Half the seats or more would go to non-managers, and unions could join. The Health and Human Services Department would handle complaints about the policy rules. The Labor Department would handle the job protection rules. Fines could reach $76,987 each time, or $769,870 for repeat cases. Workers could also take an employer to court on their own. A court could order back pay, up to triple damages, and give the job back. Set fines would run from $5,000 to $100,000, based on which rule was broken. Signed arbitration clauses and class-action waivers could not block these cases. States that take federal money could be sued by their own health staff. State attorneys general could also sue on behalf of local residents. No complaint about the care policy rules could be sent for criminal charges. The rules would be a floor. Stronger state laws and union contracts would still stand. Going against the AI would not protect a worker from a malpractice claim.
Congressional Summary
Right to Override ActThis bill requires employers of health care professionals to allow such professionals to override artificial intelligence clinical decision support systems. It also prohibits employers from retaliating against those who override these systems. The bill defines artificial intelligence clinical decision support systems as technology that supports decision-making through the use of algorithms or models that are based on clinical practice guidelines or training data and that produces predictions, recommendations, evaluations, or analysis.Health care facilities, health plans, and other entities (including government entities) that employ health care professionals and use these systems must adopt policies that allow health care professionals to use their independent judgment to override outputs from these systems. Such entities must also provide training and establish a committee to advise the entity on these systems. The Department of Health and Human Services (HHS) must enforce these requirements, including by imposing civil penalties for violations.Also, the bill prohibits these employers from taking adverse employment actions, discriminating, or retaliating against those who override these systems in accordance with employer policies. The Department of Labor must enforce these protections, which may include civil penalties specified in the bill. Individuals alleging violations of these protections may (1) submit an administrative complaint to Labor, or (2) commence a civil action. State programs receiving federal funds do not have immunity regarding such civil actions brought by employees. States may also bring civil actions against employers for violating the bill’s requirements upon notice to HHS or Labor, as appropriate.
Details
- Congress
- 119th
- Chamber
- Senate
- Status
- summarized
- Action
- Introduced in Senate
- Action Date
- 2025-10-09
- Date Added
- 2026-07-09
- Source
- Congress.gov →
Like reading a bill in plain English?
We're building an app that does this for every bill in Congress and lets you tell your reps how you want them to vote. We're a small team getting ready to launch, and we're trying to show investors that real people want this. Be one of them. Help us get it built. Leave your email and we'll tell you the moment the app is ready.