Thursday, 19 December 2024

Lessons learned in scaling up generative AI for financial services

by BD Banks

Generative AI (Gen AI) has emerged as a transformative force in the financial services sector.

Scaling generative AI for financial services

Within just a year, financial institutions have shifted from experimentation to scaling Gen AI programs, with the technology now viewed as a strategic necessity.

However, the challenge lies in implementing Gen AI effectively, avoiding common pitfalls, and realising its full potential.

Here are six key lessons learned from a year of rapid adoption.

1. AI Must Be a Strategic Priority

For Generative AI to succeed, it must be integrated into a company’s core strategy.

Institutions that have made the most progress treat AI as a CEO-level priority, ensuring organisational alignment and sufficient resources.

By positioning AI as a strategic initiative, businesses can eliminate bottlenecks and secure the funding needed for sustained momentum.

Recent surveys indicate that 60% of data and analytics leaders in banking now consider Gen AI a top strategic priority, signalling a shift in how institutions approach the technology.

2. Centralised Leadership Drives Success

The financial institutions seeing the greatest returns from Gen AI adopt a centralised approach to governance and deployment.

A review of Gen AI usage across 16 major banks by McKinsey & Company, revealed that more than half rely on a centralised model to manage initiatives, streamline decision-making and allocate resources effectively.

Centralisation also simplifies risk management and regulatory compliance, ensuring consistency across the organisation.

However, centralisation must be balanced with business unit autonomy to maintain alignment with core operational goals.

Centralised leadership should focus on creating scalable infrastructure while allowing localised innovation where it adds value.

3. Prioritise and Sequence Use Cases

Many financial institutions have struggled to achieve meaningful results by launching multiple pilots simultaneously.

Instead, focusing on a few “lighthouse” domains – such as customer service or software development – can generate early wins and build momentum.

Successful applications include agent co-pilots for customer service, coding assistance tools for software development and back-office automation for compliance tasks.

By sequencing use cases and integrating complementary analytics tools, institutions can scale Gen AI efficiently while delivering tangible benefits.

4. Build Reusable AI Infrastructure

Scaling Gen AI requires robust and reusable infrastructure.

Institutions need multi-layered tech stacks that include machine-learning operations, data governance frameworks, and access to large language models.

This “scaffolding” enables organisations to deploy AI solutions across multiple functions, from credit risk assessments to customer servicing.

Reusable infrastructure ensures that investments in AI support a wide range of applications, reducing duplication and driving long-term value.

5. Treat Data as a Corporate Asset

Data quality and governance remain critical barriers to scaling Gen AI.

Institutions that treat data as a corporate asset are better positioned to leverage AI effectively.

This involves addressing challenges like unstructured data, security classifications and data permissioning.

Elevating data management to a strategic level allows organisations to extract maximum value from their AI initiatives.

6. Focus on People and Change Management

Gen AI is as much about people as it is about technology.

Institutions that invest in reskilling, robust change management and end-user adoption see higher success rates.

For example, long-tenured employees may resist AI-driven tools, viewing them as disruptive.

Addressing these cultural challenges requires clear communication about AI’s benefits and training programs that demonstrate its value.

Looking Ahead

As Gen AI capabilities evolve, financial institutions must continue experimenting, refining and scaling their efforts.

By prioritising strategy, centralising governance, sequencing use cases and addressing cultural challenges, organisations can unlock the transformative potential of Gen AI.

The journey is still unfolding, but the lessons learned in this first year provide a strong foundation for future success.

Gen AI is not just a tool; it’s a powerful enabler that, when managed effectively, can redefine the financial services landscape.

 

The post Lessons learned in scaling up generative AI for financial services appeared first on Payments Cards & Mobile.

signup-banner

Loading