Data Gaps Hindering AI Scale in Banking
A new report from Info-Tech Research Group finds that while banks are accelerating AI initiatives, many are being held back by legacy data environments. The primary obstacle is a structural inability to access and process real-time data, which is critical for functions like fraud detection and risk analytics.
The reliance on legacy systems, often built on decades-old programming languages like COBOL, costs the financial industry over $57 billion annually in maintenance and missed opportunities. Many of these core systems run on mainframes, which still handle over 95% of all non-cash transactions and are used by over 90% of financial services companies. This infrastructure, while reliable, was designed for batch processing, not the real-time data access required for modern AI applications. Maintaining these outdated systems consumes a staggering 60-80% of a bank's entire IT budget, leaving little for innovation. This spending is often underestimated; one report found that the true total cost of ownership for legacy systems can be 3.4 times higher than budgeted when factoring in staff inefficiency and compliance workarounds. The talent pool with the necessary skills to manage these older systems is also shrinking, driving up costs for specialized engineers. AI-driven fraud detection, a critical area for banks, relies on analyzing vast datasets in real-time to spot anomalies and patterns that signal illicit activity. Companies like PayPal and JPMorgan Chase use these systems to identify everything from fraudulent transactions to money laundering schemes by establishing baseline behaviors and flagging deviations instantly. Without the ability to process data streams in real-time, banks are left in a reactive, rather than proactive, security posture. To bridge this gap, many firms are turning to unified APIs, which create a single, modern interface to access and integrate data from disparate legacy systems without requiring a complete overhaul. This approach simplifies data management, reduces maintenance costs, and allows for the scalable deployment of new tools and services. A phased migration to modern platforms has proven effective, with one European bank saving 38% in costs within 18 months and accelerating its time-to-market for new products by 62%. This challenge mirrors trends in the HR technology space in India, where the market is projected to grow from over $1.2 billion in 2025 to nearly $2.33 billion by 2034. Indian HR tech is rapidly adopting AI for recruitment, predictive analytics, and personalized employee experiences. Much like in banking, the goal is to create a unified, data-driven framework to move from administrative tasks to strategic workforce planning. The Indian startup ecosystem, particularly in Bangalore, reflects this broader tech movement. As India's fintech capital, Bangalore is home to over 2,300 fintech companies and has attracted significant investment, with the sector raising $981 million in 2025. The city is a hub for startups in both fintech and AI, with a growing number of accelerators and venture capital firms focusing on these areas.
