Banks Struggle to Scale AI as Legacy Tech Devours IT Budgets

Why It Matters

The imbalance between legacy upkeep and AI investment erodes margins and hampers banks’ competitiveness against fintechs, risking market share loss and regulatory pressure.

Key Takeaways

• 43% IT budget tied to legacy systems.
• Only 29% allocated to transformative technologies.
• Under 20% customers feel banks meet AI expectations.
• 51% AI products fail to deliver cost savings.
• Reskilling receives 23% focus versus 40% external hiring.

Pulse Analysis

Banks’ AI ambitions are colliding with entrenched legacy architectures that soak up nearly half of IT budgets. The Capgemini report highlights a stark allocation gap: while 43% of spending sustains aging mainframes and custom code, only 29% is earmarked for next‑generation platforms such as large‑language models and cloud‑native services. This fiscal mismatch forces many institutions to linger in pilot phases, delaying the transition from proof‑of‑concept to production‑grade AI that can drive measurable revenue or cost efficiencies.

Customer expectations are evolving faster than banks can adapt. Almost half of corporate and investment‑bank clients now demand real‑time responsiveness and personalized engagement, yet fewer than one‑fifth report that banks meet these standards. Fintech challengers and non‑bank players are leveraging agile tech stacks to roll out innovative products at speed, intensifying competitive pressure. Coupled with a projected slowdown in banking revenue growth to 5.4% over the next five years, the urgency to modernize becomes a strategic imperative rather than a nice‑to‑have.

To break the legacy‑AI deadlock, banks must redesign operating models and embed AI across core processes. Building enterprise‑grade platforms, fostering trusted ecosystems, and investing in internal AI talent—rather than relying predominantly on external hires—are critical steps. Robust governance frameworks will also mitigate regulatory costs that currently deter AI scaling. By aligning budget priorities, accelerating data foundation upgrades, and reskilling staff, banks can transform AI from a siloed experiment into a profit‑center that meets client expectations and safeguards future growth.