AI Is Changing How Healthcare Gets Paid: Here's What You Need to Know

IMC Board
Feb 6
6 min read

Key Takeaways:

Health systems increased AI adoption for denial management from 23% to 60% between 2024 and 2025
AI reduces appeal letter writing time from 45 minutes to 10 minutes
41% of providers now report denial rates exceeding 10%, up from 30% in 2022
Less than 1% of denied claims are appealed by patients
71% of insurers report using AI for utilization management
Effective AI implementation requires integration with existing systems and change management strategies

New Reality: When Both Sides Deploy the Same Weapon

What happens when the referee and the players both use the same playbook?

Healthcare is experiencing an unprecedented AI arms race. Providers are deploying artificial intelligence to protect revenue streams. Payers are using similar technology to manage costs. And caught in the middle is a system that's becoming more efficient at everything except the thing that matters most: getting patients the care they need without financial devastation.

The tension is palpable. Hospital executives openly acknowledge they're using AI defensively. Insurers maintain they're simply improving efficiency. Both sides are investing heavily in automation.

Patient Experience: Caught in the Crossfire

Perhaps the most troubling aspect of this AI arms race is how it affects patients. In 2023, around 73 million Americans on Affordable Care Act (ACA) plans had claims for in-network services denied. Less than 1% attempted an appeal.

Why so few appeals? The process is lengthy, confusing, and overwhelming. Many patients lack the medical expertise to challenge denials effectively. Others can't afford to wait months for resolution.

Some patients now use AI tools to fight back using services that analyze denial letters, review insurance policies, and draft customized appeals. While this democratizes access to appeals assistance, it also highlights a deeper problem: the system has become so complex that artificial intelligence is required on both sides just to navigate it.

Escalation Pattern

The numbers tell a clear story. Between 2024 and 2025, health system adoption of AI for denial management jumped from 23% to 60%. For prior authorization, the increase was equally dramatic—from 44% to 60%.

This isn't gradual adoption. This is rapid deployment driven by necessity.

Meanwhile, denial rates continue climbing. In 2022, 30% of providers reported denial rates above 10%. By 2025, that figure reached 41%. The trend shows no signs of reversing.

Here's what's particularly striking: insurers report that 71% now use AI for utilization management. Only 12% acknowledge using it specifically to deny prior authorization requests. The gap between those numbers raises questions about transparency and intent.

Where AI Creates Genuine Value

The technology itself isn't the issue. When deployed thoughtfully, AI solves real problems that have plagued healthcare finance for decades.

Nurses previously spent 45 minutes crafting appeal letters for denied claims. AI reduces that to 10 minutes. That's 35 minutes returned to patient care per appeal. Multiply that across thousands of denials monthly, and the impact becomes substantial.

Cleveland Clinic reduced days in accounts receivable from 53 to 47 between 2023 and 2025. Their aging receivables above 90 days dropped from 34.5% to 30.3%. These improvements didn't happen by accident; they resulted from integrated AI applications working across patient payments, documentation, coding, and billing.

The key difference? Integration. Solving individual issues separately can create unnecessary duplication. Integrated systems that work through electronic health records help drive lasting improvement.

Provider Perspective: Defense and Efficiency

Health systems face a fundamental challenge. They need to accelerate reimbursement while controlling costs. Manual processes can't keep pace with the volume and complexity of modern claims management.

AI addresses multiple pressure points simultaneously. It flags missing documentation before submission. It identifies claims likely to be denied based on historical patterns. It automates routine tasks that consume staff time without adding clinical value.

Some organizations report reducing annual coding costs by $500,000 while simultaneously decreasing discharged, not final billed accounts (DNFB) by 50% and increasing charge capture by 10%.

The workforce impact matters. Healthcare already faces severe staffing shortages. Ninety percent of healthcare leaders report that revenue cycle labor challenges exacerbate operations. AI doesn't replace staff; it redirects their energy toward higher-value activities that require human judgment.

Payer Perspective: Managing Risk and Cost

From the insurance perspective, the pressures are equally intense:

Medical loss ratios are climbing.
Utilization rates are increasing.
Regulatory scrutiny on risk adjustment is intensifying.

AI helps insurers manage these dynamics through improved care coordination and utilization management. In one Epic-Humana pilot, 75% of coverage checks processed automatically. Registration time per patient dropped by 90 seconds. These gains reduce administrative costs for both providers and payers.

Value-based care enablement requires sophisticated data analysis. AI makes it possible to assess provider performance, identify quality opportunities, and steer members toward high-quality care options.

The challenge for payers is demonstrating that AI improves accuracy rather than simply increasing denial volume. Trust erodes when patients receive denial letters stating their claim was reviewed by an AI program, especially when 90% of those denials are overturned on appeal.

What Smart Organizations Are Doing Differently

The organizations seeing real results share common characteristics. They're not just implementing technology, they're rethinking processes.

Prioritize integration over point solutions—Technology that works in isolation creates more problems than it solves, but systems that communicate across the revenue cycle generate sustainable improvements.
Invest in change management alongside technology deployment—AI adoption requires cultural shifts, training, and psychological safety. Organizations that neglect these human elements see technology fail despite sound implementation.
Maintain transparency about how AI makes decisions—Black box algorithms erode trust. Explainable AI that shows its reasoning builds confidence among staff and stakeholders.
Measure what matters—Clean claims, days in accounts receivable, denial overturn rates, and staff satisfaction are all metrics that reveal whether AI is actually improving operations or just automating existing problems.

Regulatory Response and Its Limitations

Regulators are beginning to catch up. California's Physicians Make Decisions Act, effective January 2025, requires that denials based on medical necessity be reviewed by qualified physicians with relevant expertise. It prohibits purely algorithmic denials.

The Centers for Medicare and Medicaid Services issued rules requiring that coverage determinations account for individual patient circumstances. Medicare Advantage plans must have qualified healthcare professionals review denials before they're issued.

These regulations help, but they don't address the fundamental misalignment. Insurance companies profit from denials. That business model creates incentive structures that AI amplifies rather than corrects.

Path Forward: Collaboration Over Competition

The current trajectory is unsustainable. Both sides deploying increasingly sophisticated AI to outmaneuver each other wastes resources and delays care.

The alternative requires difficult conversations about shared goals. What if insurers and providers used AI collaboratively to reduce unnecessary utilization while ensuring appropriate care reaches patients promptly?

Some promising examples exist. Automated prior authorization pilots show that when both sides commit to transparent processes, administrative burden drops dramatically. Clean claims increase, denials decrease, and patients receive faster decisions.

The infrastructure exists and the technology works. What's missing is alignment around purpose.

Strategic Considerations for Insurance Carriers and Agents

For carriers and agents navigating this landscape, several priorities emerge:

Invest in AI that improves accuracy as well as efficiency—Speed without precision creates more problems downstream. Technology that catches errors before they become denials generates better outcomes for everyone.
Build trust through transparency—When AI makes a coverage determination, explain the reasoning. When denials increase after implementing new technology, acknowledge the pattern and address it proactively.
Prioritize interoperability with provider systems—The more seamlessly information flows between insurers and providers, the fewer claims get denied for administrative reasons.
Consider the long-term brand implications—Patients remember when AI denied them coverage. They remember when appeals took months. They remember when the process felt like fighting a machine.

Most importantly, recognize that AI is a tool rather than a strategy. The question is how to deploy it in ways that align with organizational values and member needs.

Sources:

American Medical Association: How AI is leading to more prior authorization denials
Experian Health: State of Claims 2025: The denial problem (and is AI the answer?)
Healthcare Brew: The AI arms race between insurers, providers has begun
Modern Healthcare: Providers lean on AI startups to limit, challenge insurance denials
PBS NewsHour: How patients are using AI to fight back against denied insurance claims
Stateline: AI vs. AI: Patients deploy bots to battle health insurers that deny care

Further Thoughts

AI in healthcare revenue cycle management has moved from experimental to essential. By 2026, 72% of healthcare executives identify technology like automation and AI as their highest revenue cycle investment priority.

The arms race metaphor captures current dynamics, but it doesn't have to define the future. Technology deployed defensively creates adversarial relationships. Technology deployed collaboratively creates system-level improvements.

For digital marketers, insurance agents, and carriers, the opportunity lies in positioning AI as an enabler of better healthcare experiences rather than a cost-reduction mechanism that sacrifices care quality.

The organizations that thrive will be those that use AI to remove friction from the system instead of those that use it to shift costs between participants. That's the real competitive advantage in a market in which trust, transparency, and patient outcomes increasingly differentiate winners from everyone else.