How Agentic AI Can Power Core Insurance IT Modernization

For insurers, core system modernization is no longer about cost. It has become an existential act, the difference between remaining competitive in a digital-first market and sliding into potential irrelevance. In an industry where 74% of transformations fail—and where competitors are scaling fast as AI advances at breakneck speed—the choice is clear: modernize now or lose ground.

In a previous article, we mapped three strategic paths to modernizing core IT, identifying several pivotal best practices. (See Exhibit 1.)

In this article, we focus on the execution stage of modernization—traditionally every insurer’s nightmare. Specifically, we examine two make-or-break best practices: exploiting innovation by deploying agentic AI through every phase and employing zero-based design. We also look at the critical role of people and process in a modernization effort.

Agentic AI’s ability to accelerate and simplify onerous modernization processes is remarkable. But a successful modernization effort also requires resetting the conceptual design approach by using zero-based methods to adopt the new software’s features. When combined, these practices have a profound impact. They can deliver a program that is financially more feasible, much shorter in duration, and considerably less risky.

Applying Agentic AI Across the Modernization Journey

As the industry that has taken the lead in early AI adoption, insurers understand the potential of using agentic AI for upgrading and improving core tech. Agentic AI is not merely a tool to achieve incremental improvement; it is a transformation engine that can accelerate and simplify every phase of a large IT program, including those focused on core modernization.

The real opportunity for insurers lies in applying AI across the modernization journey—not just for decision making but in technology, processes, and operations. This is particularly true for agentic AI, which can impact every phase from legacy discovery and design to testing, roll-out, and migration. It streamlines the modernization activities that have traditionally required substantial time, cost, and expert effort.

A typical core modernization program unfolds in eight phases. (See Exhibit 2.) The Strategy phase defines the scope and priorities for modernization, setting the foundation for the subsequent executional phases, where agentic AI and generative AI (GenAI) play a more direct and tangible role.

Across the executional phases, the business value and maturity of agentic AI and GenAI differ. The value reflects a number of criteria: How critical is the step being powered by AI? How significant is the productivity improvement it delivers? How fully does it solve a problem? How scalable is the solution? The more dimensions AI affects, the greater its value. These dimensions translate directly into program outcomes like speed of delivery, quality of execution, and scalability across markets.

Maturity refers to agentic AI and GenAI’s current capabilities, scalability, and commercialization in insurance IT modernization. It’s worth noting that while some use cases are already being scaled, others are still in the experimental stages—although with AI, the trajectory from experimentation to commercial availability is surprisingly rapid.

Exhibit 3 provides a snapshot of the current value and maturity of agentic AI and GenAI tools by phase. Let’s look at each phase in detail.

Adopt and Converge Through Zero-Based Design

Zero-based design focuses on simplifying requirements and redesigning. It means rethinking processes without being constrained to legacy setups. The idea is to align processes to the logic of the new standard software rather than modifying it, and customizing only where doing so truly adds value. This is how a company eliminates the legacy complexity that traditional adaptation only perpetuates.

Our discussion centers around three concepts. “Adopt” refers to adopting standard software; “adapt” focuses on customizing where necessary; and “converge” refers to the common core elements of the software design, those that apply to all company units in multi-entity or multi-country programs.

In our version of zero-based design, “adoption and convergence by design,” a company deliberately challenges the existing design to align as closely as possible with the logic and best practices of the target software. By adopting standard solutions, insurers can maximize convergence—the reuse of components to minimize complexity, reduce duplicative efforts, and accelerate issue resolution.

The concept is particularly relevant in multi-country core modernization programs, where one core platform must fit the needs of different business entities. Then at a local level, adaptation can take place where necessary. In multi-country programs, it is important that any necessary adaptation occur within a design framework common to the whole enterprise (convergence) rather than market by market. Adoption and convergence by design reinforces consistency rather than fragmentation.

Companies should start from the big picture: devising a multi-layered technical architecture to manage the different types of business requirements. They should identify and allocate to the proper cluster local elements (for example, regulations, channels) and common (convergence) elements, such as the core products, core policy management processes, and data models, as well as integrations with the company’s other enterprise platforms (for example, CRM, analytics, other AI engines).

Where adaptation is necessary, insurers should justify every deviation according to a strict “comply-or-explain” approach. In this way, they maintain convergence while respecting local considerations in areas critical to market differentiation and business impact.

The “People Factor” Is the Key to Successful IT Modernization

AI technology and zero-based methods are critical tools for core IT modernization, but the impact of people and process management, as in any transformation, is massive. While many insurance executives understand this, they nonetheless struggle with managing these aspects of modernization.

Indeed, leading insurers already know the 10-20-70 rule: that 10% of a program’s efforts should focus on algorithms, 20% on technology and data, and the remaining 70% on people and processes.

Lack of clarity about roles and responsibilities, inattention to culture issues, and weak change management are the sorts of issues that have often derailed complex transformations. Similarly, many companies give short shrift to process management. One of the biggest culprits in an insurance modernization program is out-of-control demand management. Without strict prioritization based on business impact, the flood of requirements quickly leads to chaos, undermining the entire modernization effort.

In fact, all such challenges are magnified in existing modernization programs, which typically span many countries. Each market, product line, or function tends to pursue modernization on its own and considers itself unique, often insisting on preserving legacy business processes and products that have evolved over decades. The result is inconsistent architectures, duplication, and fragmented governance.

Whether re-engineering the existing system or rolling out brand new standard software, zero-based design guides both the IT solution architecture and the organizational mechanisms—notably, governance and the people-related aspects of transformation.

Despite the formidable—and transformative—capabilities of AI technologies, nearly three-quarters of all legacy modernization efforts still fail. With an implementation as complex as core modernization, it is easy for companies to become preoccupied with the technology and lose sight of the broader organizational aspects. This largely explains why siloed decision making, unclear accountabilities, and traditional waterfall delivery models still prevail.

Good governance enables convergence. The goal is not control, but to promote alignment and reuse—and thus speed—across entities through a federated model, based on joint prioritization, central design, and local adoption.

Thus, the central team delineates central versus local activities, establishes standards and KPIs, funds and prioritizes the common backlog, and ensures reuse across markets. It also defines the productivity and impact framework with which to measure and monitor improvements that agentic AI and GenAI will produce across the modernization phases. Local entities are empowered to move quickly due to their ability to reuse the shared components. This approach also allows insurers to gain control over demand management. (See “Reining in Demand Management.”)

Part of what makes a modernization effort so complex is the sheer number of interrelationships and dependencies. As the entity with full visibility across the program, the central team is naturally positioned to govern people and processes. This team not only coordinates between internal and external stakeholders—tech partners, software providers, and system integrators—but it also defines the operating model, capacity plan, and technical best practices needed to scale delivery.

Putting the Pieces Together for Competitive Advantage

Core IT modernization isn’t just complex and time-consuming: it can cost between 2% and 4% of non-life GWP in capex. For insurers about to embark on agentic AI-based modernization, the effort will be significant. But the return can be transformational. By our estimations, insurers can achieve a reduction of up to 0.5% and 1% of GWP of traditional capex expenditure. And tapping into the speed of advancements in agentic AI and GenAI can rapidly confer a competitive edge that can be difficult to beat.

The key to success lies in combining the power of agentic AI with adoption and convergence by design. By also addressing the distinct national and local design and governance needs, this approach maximizes efficiency, flexibility, and speed of execution for large cross-country modernization programs.

Now is the time for CEOs, COOs, and CIOs to take decisive action. Insurers that follow this two-pronged approach will lead the next wave of core modernization. In a digital-first, increasingly disruptive market, this is every insurer’s best policy for securing continued competitive advantage.