AI’s Double-Edged Sword: Government Pilots Artificial Intelligence for Healthcare Prior Authorization Amidst Deep Concerns Over Patient Access

The United States government has embarked on a contentious journey, piloting a program that integrates artificial intelligence (AI) into insurance-coverage decisions, specifically targeting the notoriously complex process of prior authorization. This initiative, spearheaded by the Trump administration through the Centers for Medicare and Medicaid Services (CMS), aims to harness AI’s analytical power to streamline approvals and curb wasteful spending. However, the move has ignited a fierce debate among medical professionals, patient advocates, and lawmakers, who fear that instead of expediting care, AI could exacerbate the already significant challenges patients face in accessing medically necessary treatments.

The Enduring Burden of Prior Authorization

For many Americans, the phrase "prior authorization" conjures images of bureaucratic nightmares, endless phone calls, and agonizing waits for critical medical care. This administrative hurdle requires healthcare providers to obtain approval from an insurance company before a prescribed treatment, medication, or procedure can be covered. While initially conceived as a mechanism to control costs, prevent overuse of services, and ensure appropriate care by checking for less costly alternatives, prior authorization has evolved into a formidable barrier to timely and effective healthcare.

Personal stories frequently surface, detailing patients caught in a "purgatory" of delays as they navigate complex requirements to secure coverage for physician-recommended care. Surveys consistently highlight the profound impact on patient well-being. A 2025 American Medical Association (AMA) survey of physicians revealed that a staggering majority voice concerns about care delays directly attributable to prior authorization. These delays can lead to patients abandoning recommended treatments, with potentially severe consequences for their health. The Commonwealth Fund’s 2025 affordability survey further underscored this crisis, finding that approximately one in five working-age adults with private insurance reported a denial of coverage for physician-recommended care for themselves or a family member. Among those denied, 41 percent experienced care delays, and more than a quarter reported a worsening of their health condition as a direct result.

The scale of this burden is particularly evident in Medicare Advantage (MA), the privately run alternative to original Medicare that now serves roughly 55 percent of Medicare-eligible seniors and disabled individuals. Insurers operating MA plans issue millions of full or partial claim denials annually based on prior authorization requirements. Federal government reports, including those from the HHS Office of Inspector General (OIG), have repeatedly documented instances where MA plans denied beneficiaries access to services even when those services met established coverage rules. Crucially, in 2024, MA plans overturned 81 percent of denials upon appeal, a statistic that profoundly highlights the often-erroneous nature of initial denials and the immense, unnecessary stress and delay they impose on patients and providers alike. Such high reversal rates suggest that many initial denials are not based on genuine medical inappropriateness but rather on administrative hurdles or potentially, an incentive to deny.

A Timeline of Reforms and Resistance

The struggle to reform prior authorization is not new, reflecting a long-standing tension between cost control and patient access.

• Pre-2020s: Prior authorization remains a significant and growing administrative burden, widely criticized by physicians and patient advocacy groups.
• 2022: An HHS OIG memorandum points to systemic issues within Medicare Advantage, noting that over one in ten prior authorization denials by MA plans were for services that appeared to meet Medicare coverage rules. This report serves as a critical backdrop to the current debate, illustrating the existing flaws in manual prior authorization processes.
• 2024 (Biden Administration Rule): Former President Joe Biden’s administration issues a significant rule aimed at reforming prior authorization for government-run plans (Medicare Advantage, Medicaid, CHIP, and Qualified Health Plans on federal exchanges). This rule mandates stricter timelines for decisions: 72 hours for urgent requests and seven calendar days for non-urgent requests. It also requires greater transparency and interoperability of health information.
• January 1, 2025: The timeline requirements of the Biden administration’s 2024 rule officially go into effect for most public sector health plans, signaling a concerted effort to reduce delays.
• 2025 (AMA Survey): The American Medical Association releases a survey revealing that 61 percent of physicians are concerned that AI tools will exacerbate denials of necessary treatments, setting a cautious tone for AI integration.
• 2025 (Commonwealth Fund Survey): This survey highlights the severe impact of prior authorization denials on working-age adults with private insurance, detailing care delays and worsened health outcomes.
• 2025 (Trump Administration & Insurers’ Pledge): Alongside insurers, the Trump administration pledges to further streamline and accelerate prior authorization processes. Private insurance companies vow to standardize electronic requests by 2027 and reduce the volume of services subject to prior authorization by 2026, including common procedures like colonoscopies and cataract surgeries.
• 2026 (WISeR Model Launch): The Centers for Medicare and Medicaid Services (CMS) launches the Wasteful and Inappropriate Service Reduction (WISeR) Model, a demonstration project using AI for prior authorization in original Medicare across six states.
• June 2026 (Federal Reports): OIG reports further document instances where MA plans rejected requests for skilled nursing and rehabilitation admissions, raising ongoing concerns about obstacles to medically appropriate care.
• June 2026 (Industry Data): Health plans release data suggesting a reduction in prior authorization requests by 11 percent between June 2025 and April 2026, aiming to demonstrate compliance with administration demands.
• December 2031: The WISeR demonstration project is scheduled to conclude, after which its efficacy and impact will be evaluated.

WISeR: A Deep Dive into the Government’s AI Pilot

The WISeR model, initiated by CMS in 2026, represents a significant expansion of prior authorization into original Medicare, where it has historically been used sparingly compared to Medicare Advantage. The model is currently being piloted in six states and is slated to run through December 2031. Its primary objective, according to CMS, is to reduce waste and fraud in original Medicare by decreasing unnecessary procedures and ensuring "timely and appropriate Medicare payment for select items and services."

The WISeR model employs a hybrid approach, combining advanced technologies such as machine learning with human clinical review. It focuses on evaluating services that CMS identifies as potentially vulnerable to overuse, fraud, and abuse. Examples of targeted services include skin and tissue substitutes, electrical nerve stimulator implants, and knee arthroscopy for knee osteoarthritis. The underlying assumption is that AI can efficiently sift through vast datasets, identify patterns indicative of inappropriate use, and flag claims that warrant closer scrutiny or denial, thereby theoretically expediting the approval of unambiguously allowable claims.

However, the very premise of WISeR has drawn sharp criticism. The integration of AI into a process already fraught with issues raises alarms about potential unintended consequences. While CMS touts the model’s ability to ensure timely payments, critics argue that the actual outcome could be further delays and wrongful denials, particularly given the checkered history of prior authorization in Medicare Advantage.

Mounting Opposition and Ethical Quandaries

The introduction of AI into prior authorization has been met with significant resistance from a diverse coalition of stakeholders, each voicing distinct yet interconnected concerns.

Medical Professionals: The AMA has been vocal in its opposition to unchecked AI use in this context. Their 2025 survey highlighted physician apprehension, with 61 percent expressing worry that AI would lead to more denials of medically necessary treatments. The AMA advocates strongly for insurers to provide detailed clinical reasoning to justify any denial of coverage, demanding greater transparency regarding the AI algorithms used. As health policy analyst Camm Epstein succinctly put it, "AI should be used to make appropriate care easier to approve, not necessary care easier to deny." Physicians fear that without adequate human oversight and transparent criteria, AI could become an opaque "black box" that prioritizes cost savings over patient well-being, eroding the physician-patient relationship and clinical autonomy.

Patient Advocates: Public opinion polls, such as those conducted by KFF, consistently show that prior authorization is viewed as a "major burden" by the public. Advocacy groups warn that the complexities of appealing AI-driven denials could disproportionately affect vulnerable populations, who may lack the resources or health literacy to navigate a convoluted appeals process. Stories of patients "stuck in prior authorization purgatory" losing valuable time or treatment options resonate deeply, fueling fears that AI could exacerbate these life-threatening delays. The promise of efficiency, they argue, must not come at the cost of equitable access to care.

Government Watchdogs: The HHS Office of Inspector General’s previous findings regarding Medicare Advantage plans serve as a critical precedent. The OIG’s 2022 memorandum pointed to numerous instances where MA plans denied services that met coverage rules, demonstrating existing systemic flaws. The fact that 81 percent of these denials were overturned on appeal in 2024 strongly suggests that initial decisions are often flawed, imposing an unnecessary administrative burden and emotional toll on patients and providers. Critics argue that introducing AI into such a system without addressing these foundational issues risks automating and amplifying existing problems rather than solving them.

Critics of WISeR Specifics: Wendell Potter, a prominent advocate for health insurance reform and former Cigna executive, has covered the political pushback against WISeR extensively. Zena Wolf, a researcher with the Center for Health & Democracy, cited investigations suggesting that even in its early months, the WISeR model has already caused delays and denials in some instances across its six pilot states. This contradicts the stated goal of expedited approvals. Furthermore, concerns have been raised about the potential for an increased administrative burden on healthcare providers, who may find themselves dedicating more resources to appealing AI-generated denials, effectively shifting the administrative load rather than reducing it.

The Profit Incentive: Perhaps one of the most contentious aspects of the WISeR model is its financial structure. Vendors participating in the program, hired to carry out the AI-driven prior authorization, earn a share of what CMS terms "averted expenditures." This means their revenue is directly tied to the cost savings achieved by denying care requests. This model has sparked intense debate, drawing parallels to long-standing concerns about profit-making incentives in the healthcare industry that discourage patients from receiving medically necessary care. Several lawmakers have responded by introducing resolutions and amendments to block funding for the WISeR model, citing these inherent conflicts of interest and the potential threat to patient access. They argue that tying vendor compensation to denials creates a perverse incentive that could prioritize financial gain over patient health.

A Tale of Two Policies: Contradictions in Government Approach

The current administration appears to be navigating a complex and seemingly contradictory path when it comes to prior authorization. On one hand, CMS is actively expanding the use of AI-driven prior authorization within original Medicare through the WISeR model, indicating a belief in its potential to control costs and reduce fraud. On the other hand, the administration has simultaneously exerted significant pressure on private insurers, including Medicare Advantage plans, to lessen and streamline their prior authorization processes.

CMS Administrator Mehmet Oz has publicly warned insurance company executives that failure to ease the burden of prior authorization voluntarily would lead to federal regulation. His stark message, "If you don’t do it yourselves, then we’re going to do it for you," underscores a clear intent to rein in the administrative excesses of private plans.

In response to this pressure, health plans recently released industry-based data, suggesting they are complying with administration demands. A survey indicated that between June 2025 and April 2026, requests for prior authorization declined by 11 percent. Insurers also pledged to standardize electronic requests by 2027 and reduce the volume of services subject to prior authorization by 2026. Furthermore, they committed to increased transparency around clinical reasoning for denials and affirmed that "AI or algorithms without clinician or practitioner review are not used to deny prior authorization requests that involve medical necessity or clinical considerations."

However, critics quickly point out that while a reduction in requests is positive, it remains unknown whether the denial rate has decreased. Without this crucial piece of data, the true impact of these industry pledges on patient access and care remains opaque. The lack of transparency around denial rates continues to fuel skepticism among patient advocates and medical professionals who fear that the fundamental issues of access and inappropriate denials persist, merely obscured by a focus on reducing initial request volumes.

The Future of Healthcare Gatekeeping: Automation vs. Patient Welfare

The deployment of AI in healthcare prior authorization is more than just a technological upgrade; it represents a pivotal moment in the ongoing struggle to balance cost containment with patient access and quality of care. The WISeR model serves as a critical test case, with its outcomes likely shaping future policies regarding AI integration across the healthcare landscape.

The central tension remains stark: AI offers undeniable potential for efficiency, capable of processing vast amounts of information far quicker than human reviewers. This could, in an ideal scenario, lead to faster approvals for straightforward cases, freeing up human staff to focus on more complex medical decisions. However, the current implementation raises profound ethical and practical concerns. The "arms race to deny faster and appeal faster," as physician Jared Dashevsky, founder of Healthcare Huddle, describes it, risks automating a broken system rather than fundamentally fixing it. He argues that while AI could eliminate barriers and reduce administrative waste, the current trajectory seems to be toward an "automation of a broken system that shouldn’t exist in its current form."

For AI to truly benefit healthcare, robust oversight, transparency in algorithms, and clear accountability mechanisms are paramount. The "black box" nature of some AI systems, combined with potential profit incentives for denials, erodes trust and could lead to algorithmic biases that disproportionately affect certain patient populations. Ensuring meaningful human review for all medical necessity decisions, as promised by insurers, is crucial, but the effectiveness of this promise remains to be seen.

The implications extend beyond mere administrative efficiency. The integration of AI into such sensitive decisions directly impacts patient health outcomes, the integrity of the patient-physician relationship, and public trust in the healthcare system. As the WISeR model continues its pilot phase through 2031, its successes and failures will provide invaluable lessons. The challenge lies in harnessing AI’s potential to genuinely improve healthcare delivery and patient access, rather than allowing it to become another barrier in an already tortuous system. Without careful design, transparent operation, and unwavering commitment to patient welfare, AI in prior authorization risks making a bad situation worse, ultimately failing to deliver on its promise of a more efficient and equitable healthcare future.