Article appeared in Insurance Innovation Reporter, March 2026
Years of system simplification and data strategy have positioned the carrier to apply AI as a productivity tool across the enterprise.
As insurers move to incorporate artificial intelligence into their operations, many are doing so atop fragmented systems and inconsistent data. New York Life Group Benefit Solutions (New York) has taken a different path. After more than a decade of platform consolidation and data strategy, the carrier has largely completed that foundational work—positioning it to apply AI across integrated workflows rather than isolated use cases.
Matt Marze, CIO, New York Life Group Benefit Solutions, describes the organization as being in the “final innings” of a long-running modernization effort—one that has eliminated hundreds of legacy systems and reduced the remaining footprint to a small number of applications slated for retirement. The strategy, he indicates, has been less about pursuing technology for its own sake than about enabling business outcomes: consolidating core capabilities onto a handful of commercial platforms, establishing a unified data layer, and creating the conditions for more consistent customer experiences. With that foundation largely in place, Marze and his team are now turning their attention to how AI can be applied across the business—not as a standalone initiative, but as a logical extension of the architecture already in place.
A Modernization Strategy Rooted in Business Outcomes
Marze’s perspective is shaped by a long tenure in group and voluntary benefits, including more than a decade leading technology at Cigna before joining New York Life as a result of its 2019 acquisition of Cigna’s group life, accident, and disability insurance businesses in a deal valued at $6.3 billion. Over that period, he has seen the business shift from a model centered on underwriting discipline and risk management to one increasingly defined by digital engagement, regulatory complexity, and the need for integration across a fragmented ecosystem of employers, brokers, and benefits technology providers.
At New York Life GBS, that evolution translated into a clear imperative: modernize core systems in a way that supports product agility and consistent client experience. Rather than pursuing large-scale technology rationalization as a standalone objective, Marze and his team tied modernization efforts to business initiatives—new product development, enhanced reporting capabilities, and improved service delivery.
“We didn’t create projects to shut down legacy systems,” Marze explains. “We created projects around enhancing products or improving capabilities—and within those efforts, we addressed the legacy environment.”
That approach has yielded a markedly simplified architecture. Core capabilities—policy administration, claims, and billing—have been consolidated onto the FINEOS (Dublin) AdminSuite platform, while front-end engagement is supported through Salesforce (San Francisco) and financial operations through SAP (Walldorf, Germany). Additional capabilities, including underwriting and digital engagement, are handled through a small number of complementary platforms and internally developed systems. The result is a tightly integrated environment built around a limited set of strategic systems rather than a proliferation of disconnected applications.
The business impact is evident in areas such as product development. Marze points to the launch of a suite of voluntary benefits products—hospital indemnity, accident, and critical illness—delivered on the same core platform architecture. The products were brought to market within roughly 18 months, the fastest such rollout in the organization’s history, and exceeded initial sales expectations.
Just as important, the simplified architecture has improved responsiveness to regulatory change, particularly in the rapidly evolving absence management space. With core capabilities centralized and configurable, new state leave programs can be incorporated with significantly less disruption than would have been required in a more fragmented environment.
Across these efforts, a consistent theme emerges: modernization was not pursued as an abstract technical goal, but as a means of enabling faster product innovation, more consistent service delivery, and greater operational flexibility. Those outcomes, in turn, have created the conditions that now make broader application of AI both feasible and practical.
Data Foundation Enables AI at Scale
If platform consolidation provides the structural backbone of New York Life GBS’s modernization effort, its data strategy supplies the connective tissue. Central to that strategy is the GBS Data Hub, a unified layer designed to bring together operational and analytical data across the organization’s core systems while maintaining traceability back to source systems.
“The key there is the data—having the data strategically available,” Marze says.
The Data Hub supports a range of use cases that predate the current wave of generative AI, including reporting, analytics, and predictive modeling. But its architecture—integrated, governed, and extensible—also positions it to support newer AI-driven capabilities. Because data from systems such as Salesforce (San Francisco), FINEOS, and SAP is already normalized and accessible within a common framework, AI can be embedded into workflows without the need to reconcile disparate data sources in real time.
This distinction is significant. In many organizations, AI initiatives are constrained by fragmented data environments, limiting their impact to narrowly defined use cases. At New York Life GBS, by contrast, the data layer has been designed as a shared enterprise asset, allowing AI capabilities to be developed once and applied across multiple functions.
Early use cases reflect this approach. One example, deployed in revenue management, addresses the long-standing challenge of reconciling client payments with accounts receivable in a self-administered billing environment. By applying AI to interpret remittance advice and match payments to outstanding balances, the organization has improved accuracy in cash application from approximately 25 percent to 65 percent.
The importance of that result lies not only in the efficiency gain, but in how it is achieved. The AI capability is not operating in isolation; it is embedded within an integrated workflow, drawing on structured data and feeding results directly into financial systems. In this sense, the use case illustrates a broader principle: AI delivers the greatest value when it is applied within a well-defined operational and data context.
For Marze and his team, this reinforces a pragmatic view of AI adoption. Rather than pursuing isolated experiments, the focus is on building reusable capabilities that can be applied horizontally across the enterprise—an approach made possible by the consistency of both the application and data layers.
AI Applied with Discipline, Not Urgency
For Marze, the current phase of New York Life GBS’s technology strategy represents a continuation of the same disciplined approach that has defined its modernization effort to date. Having established a simplified application environment and a unified data foundation, the organization is now focused on how to apply AI in ways that are both scalable and governed—avoiding the fragmentation that has characterized earlier waves of enterprise technology adoption.
“Our strategy is always evolving,” Marze notes. “One thing we’ve learned is not to get too married to something.”
That approach begins with architecture. Rather than relying on AI capabilities embedded within individual commercial platforms, New York Life GBS is developing its own AI layer, primarily within its existing AWS environment. This includes use of services such as Bedrock for large language model access, along with supporting tools for observability, evaluation, and governance. The goal is to create a set of reusable capabilities that can be applied across functions, rather than tied to specific systems.
“We’re trying to build reusable agents that we can leverage horizontally as opposed to point solutions,” Marze explains.
This emphasis on reuse reflects a broader concern with maintaining architectural flexibility in a rapidly evolving landscape. While AWS serves as the primary platform, Marze indicates that the organization is also exploring capabilities in other cloud environments, recognizing that the competitive dynamics among hyperscalers—and the capabilities they offer—are still in flux.
“I think you have to be good at multi-cloud going forward,” he says. “It would be hard to pick winners right now.”
Equally important is the decision to keep AI decoupled from core systems of record. By avoiding tight integration with vendor-specific AI features, New York Life GBS preserves the ability to evolve its AI capabilities independently of its core application stack.
“We want our AI to be agnostic of the commercial platforms,” Marze says. “That keeps our options open.”
That architectural stance is paired with a measured approach to deployment. AI initiatives are evaluated through the same lens as other technology investments, with defined business cases and expected outcomes.
“We look at AI investments no different than any of our other investments,” he explains. “We have to have a business case and see the outcomes.”
At the same time, Marze and his team are beginning to explore how AI can reshape the way work itself is organized. Central to this effort is the concept of an intent-based operating model, supported by what the organization calls its Nexus Command Center—a role-based workspace designed to unify context, guide decision-making, and recommend next-best actions across systems.
“In today’s world, work is very system-centric,” Marze observes. “In an AI world, you can move to a more intent-based model—driven by outcomes.”
In this model, users are no longer required to navigate multiple applications and assemble information manually. Instead, work is presented in a more coordinated and dynamic way, with AI helping to prioritize tasks and align actions with desired outcomes. While still in development, the concept reflects a broader shift from system-centric workflows to more integrated, outcome-driven processes.
For now, however, the emphasis remains on progression rather than disruption. AI is being introduced incrementally, embedded within the architecture and processes already in place.
“It’s about moving from specific use cases to elevating how work gets done,” Marze says.
Reducing Friction in a Complex Business
With much of its core modernization complete and AI capabilities beginning to take shape, New York Life GBS is now turning its attention to a more practical objective: reducing friction in how clients and partners interact with the business.
“We’re probably in the eighth inning of our tech modernization,” Marze says. “Now we’re focused on making it easier to do business with us.”
That focus reflects the inherent complexity of the group benefits market, where carriers must coordinate with employers, brokers, payroll systems, and a growing ecosystem of benefits technology platforms. Even with modern core systems in place, onboarding new clients, integrating data, and maintaining ongoing service relationships can involve significant manual effort and extended timelines.
“We want to take onboarding from a multi-month process to something measured in weeks,” Marze explains.
Achieving that goal depends in part on the integration capabilities the organization has built alongside its core platforms. The GBS data exchange layer, designed to connect with a wide range of external systems—including enrollment platforms, HR systems, and payroll providers—plays a central role in enabling more seamless interactions across this ecosystem.
At the same time, the organization is looking to apply AI to long-standing process challenges that affect both internal operations and client experience. One area of focus is the RFP process, which can be time-consuming and resource-intensive, particularly for large and complex accounts.
“We get RFPs from brokers and clients that can take months to respond to,” Marze notes. “We’re looking at ways to streamline that process.”
The broader objective is not simply to accelerate individual processes, but to create more continuous, less fragmented experiences across the lifecycle of a client relationship—from initial engagement through onboarding and ongoing service.
“Insurance is complex, and it’s not always easy to do business,” Marze says. “We’re looking at how we can take friction out of the system and make those experiences more seamless and intuitive.”
In this context, the value of the organization’s modernization effort becomes more tangible. Simplified systems, integrated data, and emerging AI capabilities are not ends in themselves, but enablers of a more responsive and accessible operating model—one that aligns more closely with the expectations of employers and employees navigating an increasingly complex benefits landscape.
In an industry still working to reconcile the promise of AI with the realities of legacy complexity, New York Life GBS’s approach suggests a different path—one in which advantage accrues less from early experimentation than from sustained preparation. By simplifying its core systems, establishing a unified data foundation, and maintaining architectural flexibility, the organization has created the conditions to apply AI as a practical tool for improving productivity, guiding decisions, and enhancing client and employee experiences rather than as a speculative overlay. The result is not a sudden transformation, but a position of readiness—one that reflects the same steady, business-driven discipline that has defined its modernization journey to date.
