Readers Write: AI in Revenue Cycle Demands More Than Innovation

AI in Revenue Cycle Demands More Than Innovation
By Patrice Wolfe

Patrice Wolfe, MBA is CEO of AGS Health.

​It’s hard not to conclude that the US healthcare system is at an inflection point. After more than 40 years in this industry, I feel that few other moments (perhaps COVID?) have carried the same weight of urgency, disruption, and potential.

Our complex healthcare ecosystem has always operated under pressure. Financial constraints, reimbursement changes, and a shifting regulatory environment are constants in the revenue cycle and across the broader system. What is different now is the pace and scale of technological change, particularly with artificial intelligence (AI).​

Healthcare has never been known for leading in technological innovation. Our industry is deeply tied to regulatory requirements and complex data structures and infrastructures that tend to slow adoption. Even so, we are seeing rapid movement in several pockets of our industry. AI is no longer a future consideration. It is becoming central to how revenue cycle operations and care delivery evolve.​

At the same time, the conversation has shifted from possibility to practicality. The question is no longer what AI can do in theory, but what works in real-world environments that are constrained by margin pressure, operational complexity, limited data liquidity, and uncertainty.​

The One Big Beautiful Bill Act (OBBBA) and other recent legislative and policy changes are beginning to translate into real financial impact. Analysis from Premier Inc. suggests that as much as $68 billion in hospital revenue could be at risk, with some provider organizations facing net patient revenue declines of up to 10%. For many health systems, revenue cycle optimization has already been a key strategic priority. It is increasingly becoming a necessity across the board.​

At the same time, insurance coverage continues to shift. Federal Marketplace enrollment declined 5% in 2026. That is better than expected. But signup numbers are a poor proxy for coverage. Enrollees have until March 31 to pay their premium bill, and after that, coverage will be retroactively terminated, driving higher uninsurance rates. We won’t have a clear picture until July 2026 of the impact that this will have on the insurance mix.

Pressure is also coming from the payer side, where AI adoption has progressed more quickly. Roughly 20% of claims are now being denied, and more than 60% of those denials are never appealed. That represents both a growing challenge and an opportunity for providers to recover revenue more effectively.​

Against this backdrop, health systems are taking a more disciplined approach to AI investment.

Interest in denials management, prior authorization, automation, and clinical documentation integrity remains high. The use cases are compelling. However, the standard for adoption has changed. Organizations are demanding clear, measurable return on investment before committing to solutions that often require high upfront cost and operational change.​

This shift is reflected in conversations across the industry. One health system CIO recently described being approached by a steady stream of AI vendors, each pledging transformation. His response was direct. Show proven results in comparable environments or the conversation does not move forward.

That perspective is increasingly common. Emphasis is shifting to pragmatism over experimentation. Even with that focus, implementation is not simple.​

AI adoption requires more than selecting the right use case. It depends on underlying capabilities that many organizations are still developing. Cybersecurity architecture and governance must be strong enough to support more advanced technologies. Oversight, both operational and regulatory, remains in flux. Federal-level AI regulation has shown some movement, but clarity is limited on what that framework will ultimately look like. In the meantime, organizations are moving forward in an environment that is defined by uncertainty.​

Given these conditions, the way forward is not about broad, rapid adoption. It is about targeted, disciplined execution. There is real opportunity. Modeling from McKinsey & Company suggests that AI could reduce provider collection costs by 30% to 60% over time. Realizing that potential will require a measured approach that balances automation with skilled human expertise.​

Innovation on its own is not enough. Solutions must function within existing workflows, not outside of them. Healthcare revenue cycle workflows are complex, and successful transformation depends on adopting technology that reduces friction rather than adds to it. When done effectively, this can streamline manual work, boost financial performance, and improve both patient and provider experience. The common thread is execution. ​

Healthcare does not lack ideas or innovation. What it requires now is the ability to apply both in ways that are practical, scalable, and measurable. AI will play a central role in that transformation, but only if it is deployed with discipline and a clear understanding of what success looks like in actual conditions.