How AI is Accelerating Legacy Modernization: A Bupa Case Study

Overcoming the Risks of End-of-Life Technology

For years, companies hesitated to replace critical legacy systems due to high costs, technical complexity, and operational risks. However, waiting for aging systems to fail creates compounding security and compliance vulnerabilities. According to Bupa CIO Asifa Sherazi, end-of-life technology presents hidden dangers that accumulate quietly before manifesting all at once. In healthcare environments handling sensitive personal information, losing vendor support severely restricts an organization’s control over its technology roadmap. Furthermore, relying on older frameworks introduces severe talent constraints as specialized engineering pools shrink over time. For Bupa, Microsoft’s conclusion of support for the Xamarin framework in 2024 prompted a proactive migration. Rather than waiting for a crisis, the health insurance provider transitioned its primary self-service application to native Swift and Kotlin. This rewrite yielded measurable customer benefits, including a user-perceived crash rate drop of nearly 24 percentage points on Android and eight points on iOS, alongside a doubled Android login success rate of 77%.

How Artificial Intelligence Transforms Modernization Economics

The historical friction of software migration is shifting due to modern AI tools. According to Infosys Senior Vice President Sanjeev Tripathi, artificial intelligence alters the fundamental economics of modernization by lowering effort, risk, and timeline constraints. Infosys classifies legacy modernization as one of six strategic value pools within its public AI-first framework. Infosys approached Bupa’s application transformation through a two-phase process combining reverse engineering and forward engineering. First, AI-assisted reverse engineering extracted business logic, rules, and processes from the legacy Xamarin codebase. This archeological phase preserved vital institutional knowledge and generated hundreds of native-ready user stories and regression scenarios. Second, forward engineering re-architected the application to guarantee long-term maintainability, platform stability, and feature parity in a single release. By integrating AI automation into discovery and documentation, the project team saved substantial manual business analyst effort and completed the work in a fraction of the time otherwise required.

Managing the Human Element and Cultural Challenges

Technology replacement demands careful attention to the human dimension, including preserving institutional knowledge and protecting teams from burnout. According to Asifa Sherazi, modernization requires giving employees the confidence, capability, and clarity to embrace change without dropping ongoing business-as-usual commitments. Automated code harvesting removed heavy documentation burdens, stopping institutional memory from remaining locked within a handful of individual specialists. Sherazi emphasizes that leadership’s primary duty during intense technical transformations is to absorb ambiguity and foster an environment where employees feel safe reporting challenges early.

Building Modern Foundations for Predictive AI Ecosystems

Modernized platforms provide the operational flexibility required to support future technological advancements. According to Sanjeev Tripathi, modern systems will form the base for intelligent AI-driven ecosystems where automation is embedded directly into design and operations rather than functioning as an afterthought. This architectural cleanup ultimately alters internal organizational questions. As Asifa Sherazi observes, platform conversations have shifted from asking whether aging infrastructure can support a feature to determining whether a capability aligns with customer needs.