Executive summary

Artificial intelligence (AI) has immense potential across every sector, but the opportunity in corporate, commercial and small business banking is on another level. Whether it’s credit, payments or foreign exchange, these banks’ services are complex, document-heavy and tightly regulated — the perfect conditions for AI to generate efficiencies, insight and competitive advantage.

Yet adoption is patchy. Nearly every bank has run multiple small-scale projects designed to test AI’s feasibility and benefits. But very few have scaled AI across the organization. A recent EY sponsored report with MIT Technology Review Insights on agentic AI found that 52% of banks had piloted the technology but just 16% had deployed use cases.

Where they had used it, the impact didn’t meet their expectations.

What’s holding them back? Focused on corporate and commercial banking, we interviewed banking leaders from seven institutions to discuss the barriers to widespread implementation and explore potential solutions. Additionally, we examined where these banks initially focused their AI efforts and identified their future priorities.

Alongside these banking leaders, members of EY’s corporate, commercial and small business banking and technology teams provided valuable insights into how banks can fully leverage the AI opportunity:

AI has immense potential in corporate, commercial and small business banking, but adoption is inconsistent.

of banks have piloted agentic AI

have fully deployed use cases

Source: MIT Technology Review Insights, Imagining the future of banking with agentic AI

1

Pivot from internal to external use cases.

Today, AI is primarily used for lower-value, internal applications.

ACTION: Increase consumer-facing deployments and explore how technology can transform operating models, and the services provided.

2

Empower the business.

At most banks, technology teams guide AI investment.

ACTION: Require business leaders to own the AI agenda and prioritize use cases that improve customer experience and drive revenue.

3

Dont ignore return on investment (ROI).

Assessing the ROI of AI is complex, and in the short term the ROI is often lower than anticipated. As a result, some banks are disregarding this calculation completely.

ACTION: Always measure ROI for individual use cases — even if the evaluation is preliminary and done crudely.

4

Create an AI platform for sustainable success.

Many banks struggle to scale AI because they build use cases from scratch.

ACTION: Pivot to a platform approach that's based on reusable key capabilities.

5

Explore modern ways of fixing data issues.

Incomplete or poor-quality data is the top barrier to scaling AI.

ACTION: Explore how new AI-powered tools can address data challenges and reduce the need for manual remediation.

6

Re-evaluate the balance of cloud and on-premises.

There are a variety of approaches to technology infrastructure within banks.

ACTION: Evaluate how deployment of generative AI (Gen AI) and agentic AI changes the relative merits of cloud vs. on-premises, and build-vs.-buy decisions.

7

Assess future skills requirements.

Insufficient technology skills could wreck AI ambitions.

ACTION: Map out which new skills are needed as AI is scaled and devise a strategy to attract and retain top talent.

8

Renew attention to risk.

Scaling AI, especially in customer-facing applications, brings risk and reward.

ACTION: Consider new risks arising from Gen AI, and update model risk management processes accordingly. Involve risk teams at the beginning.