1
The great AI opportunity: optimize, then transform

For banks, the real AI opportunity lies not in efficiency but in enhancing customer experience and transforming business models.

The great AI opportunity: optimize, then transform

Where does AI have the greatest potential in corporate and commercial banking? Not where it’s being used today.

Every bank executive we spoke with believes the technology is most advantageous when used to benefit the customer: by halving the time it takes to complete a loan application form by pre-populating answers; by providing immediate answers to account-related queries through chatbots; or by conveying more personalized advice by empowering relationship managers with AI-generated insights.

But banks’ early AI experimentations have been designed to create efficiencies in back-office processes, such as knowledge management, know your customer (KYC) information refreshing and data management. These use cases might significantly cut costs, but they don’t boost revenue.

“If AI is only about efficiency, you’re missing the point,” says Ulku Rowe, CIO, Commercial at Lloyds Banking Group. “Its true power is to reinvent how we decide and how we serve.”

If AI is only about efficiency, you’re missing the point. Its true power is to reinvent how we decide and how we serve.

Ulku Rowe, CIO, Commercial at Lloyds Banking Group

Customer experience is the next frontier of AI

Yet the report with MIT Technology Review Insights suggests that client-facing use cases are not a priority.

Asked about the most important outcomes they expect from deploying agentic AI, banks listed fraud detection first and security second. Enhancing customer experience came third.

There are two reasons why AI is not being used in the most profitable areas. First, there’s a worry that GenAI, which is relatively new technology, will hallucinate — give factually incorrect answers that are based on insufficient or overgeneralized training data. Without a human involved, this could lead to customers receiving incorrect information or non-compliant advice.

Used internally, AI that misfires in this way is frustrating and costly; used externally, it could be reputationally damaging. Until banks gain confidence in how the technology works, or master how humans and AI can work together to mitigate errors, it’s unlikely they will use the technology in customer-facing applications.

“Our intention during the next two years is to take internal AI use cases such as corporate chatbots and make them external,” explains Matthew Parker-Jones, Global Head of Product for Global Transaction Banking at Scotiabank. “But we can’t do that yet because we need to ensure there’s no risk of it giving the client bad advice.”

The second factor that’s slowing down banks’ adoption of AI for client-facing purposes is ownership. Technology leaders who don’t work directly with clients are behind AI adoption in most banks, and they naturally overlook its use for this purpose.

Our intention during the next two years is to take internal AI use cases such as corporate chat bots and make them external. But we can’t do that yet because we need to ensure there's no risk of it giving the client bad advice.

Matthew Parker-Jones, Global Head of Product for Global Transaction Banking, Scotiabank

Transformation is the ultimate AI use case

Banks are also overlooking an opportunity that could be even more rewarding than customer experience: a total rethink of their operating models and the services they provide.

Take transaction banking. Many banks are already using AI to improve how accurately they can identify fraudulent payments. Reducing the number of legitimate payments that are unnecessarily flagged for human review shortens the average execution timeframe, improving customer satisfaction. This is important, but it’s not revolutionary.

Now imagine a bank also using AI to evaluate all of a particular client’s outstanding outgoing payments. If the AI model were given basic contractual information such as the payment date, it could start to explore and suggest the potential liquidity benefits of holding on to that cash. It could also assess the trade-offs between different forms of payments.

For example, a faster but more costly real-time payments transaction might be preferable to a slower but cheaper traditional automated clearing house (ACH) payment. The bank is no longer a payments processor — it has become a valuable extension of the client’s corporate treasury team.

The EY Voice of the Treasurer research suggests customers would be open to these services. For example, 90% are interested in an AI financial advisor that would make recommendations in response to financial issues, and 86% would value an AI treasury assistant that provided bespoke insights.

In another example, AI could be deployed in credit not just to ease the application process and speed up time-to-fund, but also to proactively offer loans to clients when they need them. By evaluating all the internal and external financial and operational data the bank has about a client, AI could identify when, for instance, there’s a cash-flow shortfall.

Treasurers are open to new AI-powered services provided by banks:

are interested in an AI financial advisor that would make recommendations in response to financial issues

would value an AI treasury assistant that provided bespoke insights

Source: EY Voice of the Treasurer research.

I want to leverage AI to grow our business and in the long run even change how we do things like lending. That’s the real magic.

Osamu Abe, Chief of Staff for Asia Pacific, MUFG Bank

This is the real value of AI in corporate and commercial banking. There’s a first-mover advantage in using the technology in this way and changing how the bank operates to do it. But very few banks are considering what’s possible, and even fewer have started using AI in any genuinely transformative ways.

MUFG Bank is an exception. “I want to leverage AI to grow our business and in the long run even change how we do things like lending,” says Osamu Abe, the bank’s Chief of Staff for Asia Pacific. “That’s the real magic.”

Understandably, many banks are focusing on internal deployments that drive efficiencies while they augment their understanding of the technology and build foundational capabilities. But this is not the time to be complacent. The ones that move the fastest stand to gain considerable market share.

“Today, banks are in a race to build their AI capabilities and deploy impactful use cases,” says Matt Cox, EY Global Corporate, Commercial and SME Banking Consulting Leader. “They need to enter the race to use AI to transform their business. That’s the first-mover advantage.”

Today, banks are in a race to build their AI capabilities and deploy impactful use cases. They need to enter the race to use AI to transform their business. That’s the first-mover advantage.

Matt Cox, EY Global Corporate, Commercial and SME Banking Consulting Leader

Key actions

Make sure every AI use case is documented, amalgamated and visible to senior executives.
Pivot toward client-facing use cases.
Encourage everyone who works with AI to imagine how it could change the operating model and increase the value provided to clients.