KKR and Neuberger Invest in Datavant — Healthcare Data Infrastructure Is Becoming an AI Control Point

The significant minority investment leaves New Mountain Capital in control, but the bigger RCM signal is where sophisticated capital is concentrating: the infrastructure that moves healthcare data between providers, payers and the applications increasingly making decisions on top of it.

The transaction

Funds managed by Neuberger Capital Solutions and Neuberger Private Markets, together with funds and accounts managed by KKR, have agreed to acquire a significant minority stake in Datavant. New Mountain Capital, an investor since 2014, will retain control. The transaction is expected to close in the fourth quarter of 2026, subject to customary conditions.

The scale matters. Datavant says its ecosystem reaches more than 80,000 providers, 75 of the 100 largest U.S. health systems and more than 350 ecosystem partners. Its data touches roughly 90% of the U.S. population and 80% of Medicare Advantage participants.

The RCM implication: data movement is becoming strategic infrastructure

Revenue cycle automation depends on far more than a good model. Eligibility, prior authorization, coding, claim submission, payment integrity, denials and appeals all require information to move between systems and counterparties. The more autonomous RCM becomes, the more valuable reliable connectivity and normalized healthcare data become.

That creates an important distinction in the AI market. Applications can be replaced quickly. Infrastructure embedded across thousands of providers and payers is harder to replicate. Datavant's position across both sides of the healthcare transaction gives it a potential role as connective tissue for the next generation of AI-enabled workflows.

Why this matters for providers

Providers are already buying point solutions for documentation, coding, denials, patient access and collections. But those tools are only as effective as the data they can access. Over time, the competitive advantage may shift from which RCM vendor has the best isolated agent to which platform can securely assemble the most complete context around a patient, claim, payer and payment event.

That could make interoperability and data-access architecture part of the RCM technology buying decision. Health systems evaluating AI should increasingly ask not only what a product automates, but what data it can reliably access, how quickly it can access it and whether the underlying connections can support automation at enterprise scale.

The payer-provider boundary is especially important

Datavant operates across providers, payers, analytics companies and other healthcare organizations. That matters because some of the hardest RCM problems exist precisely at the boundary between payer and provider systems: medical records requests, prior authorization evidence, claim status, payment integrity reviews, denials and appeals.

If AI agents begin handling more of those interactions, the infrastructure that securely exchanges the underlying clinical and administrative data becomes more valuable—not less.

Follow the capital

KKR is investing primarily through its Strategic Investments Group, while New Mountain retains control. The structure is notable alongside reporting that KKR and Neuberger were also anchoring a roughly $2.1 billion Datavant debt recapitalization. Taken together, the activity points to substantial institutional conviction around Datavant's position in healthcare data infrastructure.

For RCM leaders, the takeaway is broader than Datavant itself. Capital is moving toward businesses that own durable layers of healthcare workflow: claims infrastructure, credentialing networks, benefit administration and health-data exchange. Those layers may ultimately determine which AI applications can scale.

What RevCycleAI is watching

The next question is whether Datavant increasingly becomes infrastructure for AI-native administrative workflows rather than primarily a data-exchange platform. If that happens, the strategic value of its network could extend further into revenue cycle, payment integrity and payer-provider automation.

In an AI-first revenue cycle, the model may become commoditized faster than the trusted data rails underneath it.