Artificial intelligence is becoming a significant part of how Medicare works behind the scenes. From screening prior authorization requests to flagging potentially unnecessary services, AI tools are increasingly involved in the process that determines whether your care gets approved or denied. These changes affect both Original Medicare and Medicare Advantage, and they carry real implications for your access to treatment.
This article explains what is happening, what protections are in place, and what you can do to stay informed and advocate for your care.
How AI Is Being Used in Medicare Today
AI in Medicare takes several forms. Insurance companies and government contractors use algorithms and machine learning models to analyze medical records, predict patient outcomes, and make recommendations about whether a service should be covered. In some cases, these tools help speed up approvals. In others, they have been used to justify denying care.
There are two main areas where AI is playing a growing role:
- Prior authorization screening, where AI reviews requests for services before they are provided and flags those that may not meet coverage criteria
- Claims review, where algorithms analyze submitted claims for patterns of overuse, waste, or fraud
The goal, according to the Centers for Medicare & Medicaid Services (CMS), is to reduce wasteful spending while maintaining timely access to medically necessary care. However, the rapid adoption of these tools has raised concerns among patients, doctors, and lawmakers about accuracy, transparency, and fairness.
The WISeR Model: AI Comes to Original Medicare
One of the most notable developments in 2026 is the launch of the Wasteful and Inappropriate Service Reduction (WISeR) Model, a CMS pilot program that introduces AI-assisted prior authorization to Original Medicare for the first time.
What It Is
WISeR is a six-year pilot program that began on January 1, 2026, and will run through December 31, 2031. It operates in six states: Arizona, New Jersey, Ohio, Oklahoma, Texas, and Washington. The program uses AI and machine learning, combined with human clinical review, to screen prior authorization requests for select medical services.
Which Services Are Affected
The WISeR model currently requires prior authorization for 17 procedures, including:
- Knee arthroscopy for knee osteoarthritis
- Skin and tissue substitutes
- Electrical nerve stimulator implants
- Certain incontinence control devices
If you live in one of the six participating states and receive Original Medicare, your provider may need to submit a prior authorization request before performing one of these services.
How the Process Works
CMS selected six technology companies as model participants to help run the program: Cohere Health, Genzeon Corporation, Humata Health, Innovaccer, Virtix Health, and Zyter. These contractors use AI tools to screen prior authorization requests against existing Medicare coverage policies.
Key safeguards built into the WISeR model include:
- Human clinical review is required for all denials. If the AI system recommends that Medicare not pay for a service, an appropriately licensed clinician must review that recommendation using standardized, transparent, and evidence-based procedures before any denial is issued.
- Decisions must be made within set timeframes. Standard prior authorization decisions are expected within 72 hours, and expedited requests within 48 hours.
- A "gold carding" exemption program is planned for 2026 and beyond. Providers and suppliers with strong compliance records may eventually be exempt from WISeR review requirements.
Concerns About WISeR
The program has drawn criticism from physicians and some members of Congress. Doctors worry that adding prior authorization requirements to Original Medicare could delay care, particularly for patients who need timely procedures. Others have raised concerns about the financial incentives built into the model, noting that the selected technology vendors are paid based in part on how much money they save Medicare by identifying unnecessary services.
CMS has stated that the program is designed to reduce waste and protect taxpayers while maintaining access to appropriate care. Beneficiaries in the six pilot states should be aware that some services now require advance approval that previously did not.
AI in Medicare Advantage: Insurer Controversies
While CMS is piloting AI in Original Medicare with built-in safeguards, the use of AI by private Medicare Advantage (MA) insurers has been far more controversial.
The nH Predict Algorithm
The most prominent case involves UnitedHealthcare and its subsidiary naviHealth, which developed an AI tool called nH Predict. This algorithm was designed to predict how long patients would need post-acute care services such as skilled nursing facility stays and rehabilitation.
According to a class action lawsuit filed against UnitedHealth Group, the company used nH Predict to override the clinical judgment of treating physicians and deny coverage for post-acute care. The lawsuit alleges that the algorithm carried a 90% error rate, meaning that nine out of ten denials based on the tool were ultimately reversed when patients appealed.
In February 2025, a U.S. District Court in Minnesota ruled that the case could proceed, finding that the claims primarily turned on whether UnitedHealth broke its own contract provisions stating that coverage decisions would be made by clinical staff, not by an algorithm. In March 2026, the court ordered UnitedHealth to disclose internal documents about whether the AI tool was designed to override doctors' clinical judgment.
Humana and CVS
UnitedHealthcare is not the only insurer facing scrutiny. Humana has been sued for allegedly using the same nH Predict tool to cut payments prematurely for rehabilitative care, favoring AI-generated predictions over physicians' recommendations.
CVS Health, which operates Aetna's Medicare Advantage plans, has also come under scrutiny. According to a 2024 U.S. Senate investigation, CVS initially projected that its algorithmic tools would save $10 million to $15 million over three years, but later revised that estimate to $77.3 million, raising questions about whether the tools were being used to maximize cost savings at the expense of patient care.
The Senate Investigation
In October 2024, the Senate Permanent Subcommittee on Investigations released a report examining the three largest Medicare Advantage insurers. The findings were significant:
- UnitedHealthcare's denial rate for post-acute services increased from 8.7% to 22.7% between 2019 and 2022
- UnitedHealthcare's skilled nursing facility denial rate increased ninefold during the same period
- Humana's denial rate for long-term acute-care hospitals increased 54% between 2020 and 2022
- The report concluded that MA insurers were "intentionally using prior authorization to boost profits by targeting costly yet critical stays in post-acute care facilities"
These findings have added urgency to calls for stronger regulation of how insurers use AI in coverage decisions.
CMS Regulations: What Guardrails Exist?
The regulatory landscape around AI in Medicare is still evolving. Here is where things stand as of early 2026.
What CMS Proposed
In the proposed rule for Contract Year 2026, CMS put forward new guardrails for Medicare Advantage plans using AI, citing the growing use of artificial intelligence in healthcare and reports that AI may lead to "algorithmic discrimination." The proposed requirements would have required MA plans to demonstrate that their AI tools do not produce inequitable treatment or bias in healthcare decisions.
What Was Finalized
In the CY 2026 final rule, published in April 2025, CMS did not finalize those AI-specific guardrails. The agency acknowledged "broad interest in regulation of AI" and stated it would "continue to consider the extent to which it may be appropriate to engage in future rulemaking in this area."
Existing Protections
Even without new AI-specific rules, several existing CMS requirements apply:
- Every MA coverage decision must be based on the individual member's circumstances, including their medical history, physician recommendations, and clinical notes
- Algorithms may be used to help predict outcomes (such as expected length of stay in post-acute care), but they cannot be the sole basis for terminating coverage
- A qualified healthcare professional must review any denial before it is issued to the patient
- Plans must provide written notice explaining why a service was denied, along with instructions on how to appeal
The Fax Machine Phase-Out
In a related modernization effort, CMS finalized a rule in 2026 to phase out fax machines and paper mail from large portions of the healthcare claims process. The rule establishes the first HIPAA-adopted standards for electronic health care claims attachments, enabling secure digital exchange of medical records, imaging, clinical notes, and lab results.
CMS estimates this will save the healthcare industry approximately $782 million per year. The rule takes effect on May 26, 2026, with a two-year transition period for full compliance. While this is not directly about AI, the shift to electronic records creates the digital infrastructure that AI systems rely on to process claims and prior authorization requests.
What This Means for Beneficiaries
If you are enrolled in Medicare, whether through Original Medicare or a Medicare Advantage plan, here is what you should know about AI and your coverage.
Your Rights Have Not Changed
Regardless of whether AI is involved in reviewing your claim, your fundamental rights as a Medicare beneficiary remain the same:
- You have the right to receive all medically necessary care covered by your plan
- You have the right to a clear explanation if a service is denied
- You have the right to appeal any denial, and the appeals process includes multiple levels of review
- A human clinician must be involved in any decision to deny coverage
The Appeals Process Matters More Than Ever
Research consistently shows that Medicare beneficiaries who appeal coverage denials have a strong chance of success. According to available data, approximately 80% of appealed Medicare Advantage prior authorization denials are decided in the patient's favor. This suggests that many initial denials, including those influenced by AI tools, do not hold up under closer review.
If you receive a denial:
- Read the denial notice carefully. It must explain why the service was denied and cite specific coverage criteria.
- Talk to your doctor. Ask whether the denial aligns with their clinical recommendation and whether additional documentation could support your case.
- File an appeal promptly. You typically have 60 days to file an appeal for a Medicare Advantage denial. For Original Medicare, the timeframe is 120 days.
- Keep records. Save copies of all correspondence, denial letters, and medical documentation.
- Get help if needed. Your State Health Insurance Assistance Program (SHIP) offers free counseling. You can also contact 1-800-MEDICARE (1-800-633-4227) or visit Medicare.gov for guidance on the appeals process.
What to Watch For
As AI becomes more embedded in Medicare operations, there are several things beneficiaries should be aware of:
- If you live in Arizona, New Jersey, Ohio, Oklahoma, Texas, or Washington, your Original Medicare benefits may now require prior authorization for certain procedures under the WISeR pilot. Ask your provider whether any planned services are affected.
- If you have a Medicare Advantage plan, be aware that your plan may use AI tools as part of its coverage review process. This does not change your rights, but it means you should pay close attention to any denial notices and not hesitate to appeal.
- Ask questions. If a service is denied, ask your plan whether an algorithm or AI tool was used in the review process. While plans are not currently required to disclose this in all cases, asking can help you understand the basis for the decision.
- Review your Medicare Summary Notice (MSN) or Explanation of Benefits (EOB) regularly to ensure that services you received were processed correctly.
The Balance Between Efficiency and Accuracy
AI has genuine potential to improve Medicare administration. Processing prior authorization requests faster, identifying billing errors, reducing paperwork, and catching fraudulent claims are all areas where technology can benefit both the program and its beneficiaries.
However, the controversies around insurer AI tools highlight real risks. When algorithms are used to deny care based on statistical predictions rather than individual clinical assessments, patients can lose access to treatments their doctors believe are necessary. The high reversal rate on appeal suggests that some AI-driven denials are not reaching the right conclusion.
The challenge for CMS, insurers, and the healthcare system is to harness AI's efficiency without sacrificing the accuracy and individualized review that beneficiaries deserve. The WISeR model's requirement for human clinical review of all denials represents one approach to this balance. Whether future rulemaking will impose similar or stronger requirements on Medicare Advantage plans remains an open question.
How to Stay Informed
Medicare policies around AI are likely to continue evolving throughout 2026 and beyond. To stay up to date:
- Visit Medicare.gov for official information about your coverage and rights
- Call 1-800-MEDICARE (1-800-633-4227) with questions about claims, denials, or the appeals process, available 24 hours a day, 7 days a week
- Contact your State Health Insurance Assistance Program (SHIP) for free, personalized counseling
- Review your plan's Evidence of Coverage document, which outlines what services require prior authorization and how coverage decisions are made
Understanding how AI fits into Medicare does not require technical expertise. What matters most is knowing your rights, reviewing your notices, and being prepared to advocate for the care you need.