Oliver Wyman just published their 2026 Healthcare RCM Survey — 200+ decision-makers and 90 end users across health systems, medical groups, and outpatient facilities. The findings are worth reading carefully, not because they're surprising, but because they put hard numbers on something the industry has been arguing about for three years: is AI in RCM actually working?
The answer, at least for a meaningful slice of the market, is yes.
The Pilot Era Is Over
Roughly 20% to 40% of organizations surveyed report broad or enterprise-wide AI deployment — meaning it's running across a majority of sites or departments, not confined to a single test unit. That's not fringe adoption. That's a market that has made a decision and is building on it.
The investment trajectory reinforces this. Between 70% and 90% of decision-makers expect to increase AI spending over the next three years, with many planning moderate to significant annual growth. When you see that kind of forward commitment from a survey of this size, it's not hype — it's capital allocation.
"AI in RCM is moving from isolated use cases to being integrated into day-to-day operations." — Oliver Wyman
The No-Regret Stack Is Becoming Clear
One of the more useful findings: 92% of respondents agree there are no-regret AI investments to pursue in RCM, even if they disagree on which ones. But four use cases floated to the top consistently:
- Ambient documentation — reduces physician documentation burden at the point of care
- Clinical documentation improvement (CDI) — ensures coded complexity matches true patient acuity
- Coding automation — up to 46% reduction in coding time on complex cases
- Electronic prior authorization (ePA) — removes the most friction-heavy manual step in the revenue cycle
These aren't random picks. They share a common profile: they plug into existing workflows without requiring full-stack transformation, they attack documented friction points, and they produce measurable financial output. CDI alone, done well, can recover millions annually by ensuring the case mix index reflects what's actually being treated.
The Gap Between Leaders and Laggards Is Widening
This is the part that should concern smaller and community-based providers the most. Adoption is not uniform. Leading systems are scaling AI across workflows and capturing compounding gains in both efficiency and revenue optimization. Organizations that are still in early deployment stages are falling further behind with each quarter.
Oliver Wyman is direct about this: the divergence "may reshape competitive dynamics across the healthcare ecosystem." That's consultant language for: the organizations moving now are building a structural advantage that gets harder to close over time.
For smaller providers, the path forward isn't to build — it's to buy, deliberately. The survey frames it as a vendor consolidation question: best-of-breed point solutions, or a smaller number of strategic partners? Given the AI governance and risk management overhead that comes with broad deployment, the case for consolidation is getting stronger.
What This Means for Payers
Better documentation and coding accuracy on the provider side isn't just a provider win — it directly hits payer financials. As providers capture more complexity in their CDI and coding, reimbursement levels increase. Oliver Wyman notes this is already contributing to measurable shifts in cost-of-care trends.
The payer response, per the report, needs to be accelerated investment in payment integrity, analytics, and audit capabilities. And longer term, a move toward payment models that better align incentives with clinical value rather than documentation completeness.
This is the AI arms race that's been building for two years. Providers get better at capturing revenue. Payers get better at scrutinizing claims. The net result is more administrative complexity for everyone — unless the underlying payment structure changes.
The Bottom Line
If you're still treating AI in RCM as something to evaluate, you're behind. The Oliver Wyman data confirms what the leading health systems already know: the tools work, the ROI is measurable, and the organizations that move now are locking in an operational advantage. The question is no longer whether to deploy AI in your revenue cycle. It's which workflows to prioritize first — and how fast you can scale.