Why do most banks struggle to adopt and scale AI successfully?

Many banks struggle to scale AI because legacy systems block digital progress. These outdated systems can consume up to 80% of tech budgets, limiting access to the high-quality data AI needs. As a result, as many as three-quarters of GenAI pilots fail to move beyond proof-of-concept. Other major challenges, such as a shortage of AI-ready talent and the inability to demonstrate tangible ROI to secure continued investment, also slow adoption.

How can finance leaders achieve measurable ROI from AI in banking and corporate finance?

To achieve measurable ROI, leaders should align GenAI initiatives with clear strategic goals like cost reduction, revenue growth or risk management. The most effective approach is incremental modernization, which delivers value early and continuously to sustain funding. Measuring ROI requires tracking metrics such as revenue per customer, cost-to-serve, or modernization speed, turning AI from a pilot expense into a proven growth driver.

What AI governance framework helps banks balance innovation with compliance and risk management?

A strong AI governance framework in banking should focus on three key pillars: Data quality: Establish rigorous cleansing, normalization and lineage tracking to ensure data accuracy and reliability. Model development: Define use cases and train models on reliable data, ensuring the ability to detect and mitigate bias. Cybersecurity: Protect AI systems and customer data through encryption, access controls, and continuous threat detection. Together, these elements help banks innovate confidently while staying compliant and resilient.

How can explainable AI help banks and financial institutions ensure transparency, fairness and compliance?

Lack of transparency in AI models is one of the biggest barriers to accountability. Explainable AI helps banks trace decisions back to the data and logic that shaped them, making outcomes easier to understand and justify. This transparency enables early detection of bias, supports fairer decisions in areas like lending and credit scoring, and strengthens regulatory compliance and stakeholder trust.

Start your bank’s GenAI transformation

Read the e-book

The GenAI guide to financial industry excellence

Learn how to scale AI in banking and gain the market edge

Up to 75% of GenAI pilots in banking fail.
Make yours a success story.

Banks aren't asking if GenAI is valuable anymore. They're asking how to get from pilot to impactful results, fast.

While many have AI strategies in place, the #1 barrier to success remains legacy debt. Over half cite outdated systems as the biggest obstacle to achieving digital goals, often locking up up to 80% of IT budgets.

This e-book offers a way forward. Discover how to overcome legacy, data and strategy barriers to turn GenAI pilots into real business success.

Read the e-book

The ROI of doing modernization differently

One global bank earned $800 million in new revenue in just two years by modernizing incrementally, with zero downtime and a 30% cost reduction.

How GenAI helps demystify legacy

GenAI achieved 80% accuracy in understanding complex legacy code, helping one major firm analyze and modernize 4.5 million lines in just 14 weeks.

Why hyperscalers are changing the economics of AI

Cloud hyperscalers are removing the entry barriers to GenAI, giving banks on-demand computing power without the massive upfront cost.

GenAI’s ripple effect across the enterprise

From 30% faster delivery to greater digital resilience, banks using GenAI tools are gaining new speed, sharper insights and a stronger competitive edge.

The advances that GenAI is making possible are such that institutions that fail to integrate the technology into their growth strategies — or that lag the adoption curve — will quickly find themselves outperformed by more nimble competitors.

Keep reading

The advances that GenAI is making possible are such that institutions that fail to integrate the technology into their growth strategies — or that lag the adoption curve — will quickly find themselves outperformed by more nimble competitors.

Keep reading

Meet the authors

Sannesh Prabhu

Global Head of Pre-Sales Engineering and Solutions at Thoughtworks

See bio

Sandeep Reddy

Global Head of Financial Services Strategic Initiatives and Delivery at Thoughtworks

See bio

Wayne Te Paa

Managing Director of Banking, Financial Services and Insurance Services (APAC) at Thoughtworks

See bio

Omar Bashir

Technical Director of Banking and Financial Services (APAC) at Thoughtworks

See bio

FAQ

Many banks struggle to scale AI because legacy systems block digital progress. These outdated systems can consume up to 80% of tech budgets, limiting access to the high-quality data AI needs. As a result, as many as three-quarters of GenAI pilots fail to move beyond proof-of-concept.

Other major challenges, such as a shortage of AI-ready talent and the inability to demonstrate tangible ROI to secure continued investment, also slow adoption.
To achieve measurable ROI, leaders should align GenAI initiatives with clear strategic goals like cost reduction, revenue growth or risk management. The most effective approach is incremental modernization, which delivers value early and continuously to sustain funding.

Measuring ROI requires tracking metrics such as revenue per customer, cost-to-serve, or modernization speed, turning AI from a pilot expense into a proven growth driver.
A strong AI governance framework in banking should focus on three key pillars:
- Data quality: Establish rigorous cleansing, normalization and lineage tracking to ensure data accuracy and reliability.
- Model development: Define use cases and train models on reliable data, ensuring the ability to detect and mitigate bias.
- Cybersecurity: Protect AI systems and customer data through encryption, access controls, and continuous threat detection.
Together, these elements help banks innovate confidently while staying compliant and resilient.
Lack of transparency in AI models is one of the biggest barriers to accountability. Explainable AI helps banks trace decisions back to the data and logic that shaped them, making outcomes easier to understand and justify.

This transparency enables early detection of bias, supports fairer decisions in areas like lending and credit scoring, and strengthens regulatory compliance and stakeholder trust.