The United States healthcare system is currently grappling with a significant paradigm shift as the government pilots a program leveraging artificial intelligence (AI) for insurance-coverage decisions, specifically within the contentious realm of prior authorization. This initiative, spearheaded by the Trump administration, aims to streamline processes and reduce what it identifies as wasteful spending in Original Medicare, but it has ignited a firestorm of debate among physicians, patient advocates, and lawmakers who fear it could exacerbate existing challenges and lead to an increase in wrongful denials of medically necessary care.
Understanding Prior Authorization: A Double-Edged Sword
Prior authorization, often referred to as "pre-approval," is a requirement from health insurance companies that patients or their healthcare providers obtain approval before a service, prescription, or procedure is covered. Originally conceived as a mechanism to control healthcare costs by preventing overuse of services and ensuring patients receive the most appropriate, cost-effective treatments, its implementation has increasingly become a source of frustration and delay for millions of Americans.
For many patients, the process of securing pre-approval for physician-recommended medical care is fraught with bureaucratic hurdles. Personal narratives frequently highlight the arduous journey patients undertake, navigating complex paperwork, endless phone calls, and lengthy waiting periods to secure coverage for essential medications, specialized procedures, and even hospital admissions. This administrative burden not only stresses patients and their families but also diverts valuable time and resources from healthcare providers.
While prior authorization can serve a legitimate purpose in curbing unnecessary expenditures and promoting evidence-based care, its current application often leads to significant care delays. A substantial majority of physicians, as evidenced by a 2025 American Medical Association (AMA) survey, express profound concerns that these delays compel patients to abandon recommended treatments altogether. The waiting game for insurers to verify eligibility and confirm medical necessity can be critical, especially for conditions requiring timely intervention. When care is denied, patients have the option to appeal, but this process adds another layer of complexity and time, often with uncertain outcomes.
The Allure and Alarm of AI in Healthcare Approvals
The advent of artificial intelligence, with its unparalleled capacity to process and analyze vast datasets rapidly, presents a tempting solution to the inefficiencies of prior authorization. Proponents suggest that AI could significantly expedite the approval of clear-cut, unambiguously allowable claims, thereby reducing administrative backlogs and mitigating care delays. The promise is a system where routine approvals are instantaneous, allowing human reviewers to focus on more complex cases.
However, the integration of AI into prior authorization is far from universally embraced. Significant resistance stems from fears that AI algorithms, if improperly designed or deployed, could lead to an increase in wrongful denials. The 2025 AMA survey revealed that a striking 61 percent of doctors harbor concerns that AI tools will worsen the problem of denying treatments they deem medically necessary. This apprehension is rooted in the black-box nature of some AI systems and the potential for algorithms to prioritize cost-cutting over patient well-being.
Health policy analysts like Camm Epstein articulate a clear principle: "AI should be used to make appropriate care easier to approve, not necessary care easier to deny." This sentiment underscores a broader call for ethical AI deployment that augments, rather than replaces, sound clinical judgment and patient advocacy. The AMA has been a leading voice in this regard, advocating for insurers to provide transparent, detailed clinical reasoning for any denial of coverage and demanding greater clarity regarding the underlying algorithms used by AI systems. This push for transparency aims to ensure accountability and allow for effective oversight of AI-driven decisions.
The WISeR Model: A Government Pilot Program
In response to ongoing challenges and as part of broader efforts to manage healthcare spending, the Trump administration launched a demonstration project this year called the Wasteful and Inappropriate Service Reduction (WISeR) Model. This program, slated to run through December 2031 across six states, explicitly integrates AI and machine learning with human clinical review to identify and reduce waste and fraud within Original Medicare. The model specifically targets services deemed vulnerable to overuse, fraud, and abuse, including skin and tissue substitutes, electrical nerve stimulator implants, and knee arthroscopy for knee osteoarthritis.
The introduction of prior authorization into Original Medicare, traditionally less reliant on such mechanisms compared to Medicare Advantage, marks a significant policy shift. While CMS asserts that the WISeR model will "ensure timely and appropriate Medicare payment for select items and services," critics remain deeply skeptical. Wendell Potter, a health insurance reform advocate and former Cigna executive, has highlighted significant political pushback against the model. Similarly, Zena Wolf, a researcher with the Center for Health & Democracy, has pointed to investigations by prominent news outlets—including the Washington Post, KFF Health News, and the Seattle Times—suggesting that within its initial months, the WISeR model has already contributed to care delays and denials in the pilot states.
A particular area of concern revolves around the compensation structure for vendors participating in the WISeR model. These contractors are hired to implement the AI-driven prior authorization and earn a share of what CMS terms "averted expenditures." This direct financial incentive for rejecting care requests raises serious ethical questions about potential conflicts of interest and the prioritization of profit over patient health. Critics argue that such a model could inadvertently encourage denials, leading to medically appropriate care being withheld. This concern resonates with long-standing worries about profit-making within the healthcare system based on discouraging patients from accessing necessary treatments.
In response to these anxieties, several lawmakers have introduced resolutions and amendments aimed at blocking funding for the WISeR model, citing grave threats to patient access to care. Their actions underscore the bipartisan apprehension surrounding the unchecked deployment of AI in critical healthcare decision-making processes, especially when tied to financial incentives for denials.
Mounting Concerns: Physician and Patient Perspectives
The public widely perceives prior authorization as a significant burden, a sentiment consistently echoed in surveys. A KFF Health Tracking Poll identified prior authorizations as one of the public’s biggest burdens when seeking healthcare. This burden is particularly pronounced in Medicare Advantage (MA), the privately run alternative to Original Medicare, which now covers approximately 55 percent of Medicare-eligible seniors and disabled individuals.
In Medicare Advantage plans, insurers issue millions of full or partial claim denials annually based on prior authorization requirements. Recent government reports issued in June shed further light on these practices. For instance, an HHS Office of Inspector General (OIG) memorandum published in 2022 indicated that more than one in ten instances saw Medicare Advantage plans denying beneficiaries access to services despite those services apparently meeting coverage rules. While a significant percentage of these denials (81 percent in 2024) are overturned upon appeal, the initial denial creates substantial stress, delay, and potential harm. More recent OIG reports in June highlighted instances where plans even rejected requests for skilled nursing and rehabilitation admissions, raising serious concerns about access to medically appropriate care.
The human cost of these denials and delays is substantial. A newly released Commonwealth Fund survey in June 2025 revealed that roughly one in five American working-age adults with private insurance reported themselves or a family member being denied coverage for physician-recommended medical care in the past year. Of those who experienced a prior authorization denial, 41 percent reported delayed care, and over a quarter stated that their health problem worsened as a direct result. Patients often find themselves in a "prior authorization purgatory," where they run out of time or viable treatment options while awaiting insurer decisions, as reported by NBC News.
A Divergent Regulatory Path and Industry Responses
Paradoxically, the Trump administration appears to hold a nuanced, if not contradictory, stance on prior authorization. While expanding its use in Original Medicare through the AI-driven WISeR model, the administration has simultaneously pressured private insurers, including Medicare Advantage plans, to reduce and streamline prior authorization requirements. CMS Administrator Mehmet Oz publicly warned insurance company executives that failure to ease the burden of prior authorization would lead to federal regulation, stating, "If you don’t do it yourselves, then we’re going to do it for you."
This pressure from the executive branch, alongside legislative initiatives, has spurred some industry action. The Biden administration, for example, issued a rule in 2024 designed to reduce delays for patients with government-run plans and streamline the prior authorization process for physicians. This rule, which went into effect on January 1 of this year for most public sector health plans, mandates that insurers make urgent prior authorization decisions within 72 hours and non-urgent decisions within seven calendar days.
In an apparent effort to preempt further government intervention, private insurance companies have also made pledges. Last year, in conjunction with the Trump administration, they vowed to standardize electronic requests by 2027 and to "reduce the volume of medical services subject to prior authorization" by 2026, including for common procedures like colonoscopies and cataract surgeries. Recent industry-based survey data suggest some movement, with requests for prior authorization reportedly declining by 11 percent between June 2025 and April 2026. However, it remains unclear whether this reduction in requests has translated into a decrease in denial rates, a critical metric for patient access.
Furthermore, an industry group survey conducted last year indicated that all responding health plans affirmed they do not use "AI or algorithms without clinician or practitioner review… to deny prior authorization requests that involve medical necessity or clinical considerations." Insurers also promised greater transparency regarding the clinical reasoning behind prior authorization decisions. While these assurances might alleviate some concerns about purely automated denials, they do little to assuage the deeper skepticism regarding the fundamental role and impact of prior authorization itself.
The Road Ahead: Navigating the Future of AI and Prior Authorization
The debate surrounding AI’s role in prior authorization encapsulates the broader tension within healthcare: the imperative to control costs versus the commitment to patient access and quality of care. While AI offers tantalizing possibilities for efficiency, its application in a system already criticized for its bureaucratic hurdles and denial rates raises profound questions about fairness, ethics, and human oversight.
Experts like Jared Dashevsky, a physician and founder of Healthcare Huddle, articulate a vision where AI could genuinely improve healthcare by "eliminate barriers, reduce administrative waste, [and] give us more time with patients." However, he cautions that the current trajectory seems to be fostering an "arms race to deny faster and appeal faster," leading to "more automation of a broken system that shouldn’t exist in its current form."
The journey to integrate AI responsibly into healthcare will require careful calibration, robust regulatory frameworks, and unwavering vigilance. The outcomes of the WISeR model and the industry’s self-regulatory efforts will be closely watched. Ultimately, the success of AI in prior authorization will not be measured solely by cost savings or processing speed, but by its ability to genuinely enhance patient care, reduce administrative burdens for providers, and ensure that medically necessary treatments are approved without undue delay or arbitrary denial. The challenge lies in harnessing AI’s power to serve patients and clinicians, rather than allowing it to become another barrier in an already complex and often tortuous healthcare landscape.
This article was originally published on Undark.
