ING Accelerates AI Adoption Despite Generational Divide

Speaking to FinTech Magazine, ING's Chief Analytics Officer Bahadir Yilmaz discusses the Dutch bank's digital transformation journey and its approach to artificial intelligence adoption across its European operations.

The banking giant has positioned itself at the forefront of digital innovation by embracing a cloud-first approach that prioritises partnerships with major hyperscalers over traditional on-premises infrastructure. 

This strategic pivot reflects a broader industry trend towards scalable, flexible technology solutions that can adapt to rapidly evolving customer demands.

Cloud partnerships drive innovation

ING has forged particularly strong relationships with leading cloud providers, with Google emerging as a key strategic partner. 

"We always opted for a strategy that prioritises cloud services over on-prem services, and we are really moving in that direction," Bahadir explains. 

"We are working with the biggest names in the hyperscaler space, and we are also doubling down in our relationships with them, especially with Google."

This partnership extends beyond traditional vendor relationships, with ING gaining access to early-stage features whilst contributing anonymised data to help improve Google's services. 

Such collaborative arrangements highlight how major financial institutions are increasingly viewing technology partnerships as strategic assets rather than mere procurement exercises.

The bank's approach to AI implementation demonstrates a mature understanding of both the opportunities and challenges facing established financial institutions. 

Bahadir emphasises that whilst the technical barriers to AI adoption continue to diminish, the real challenge lies in responsible deployment at scale.

Risk management remains paramount

Despite the apparent simplicity of implementing AI solutions, ING maintains a cautious approach rooted in comprehensive risk management. "In the first year of AI transformation, we really focused on the safe, responsible element of AI," Bahadir notes. 

"As a bank, what we have to do is really do our homework in terms of risk management, and this risk management is evaluating 20 different risks by doing a variety of evaluations."

The bank's risk framework encompasses multiple evaluation criteria, ensuring that AI implementations meet both regulatory requirements and internal governance standards.

The technical accessibility of AI tools continues to improve dramatically. "I don't think there's anything that is hard in building a chatbot," Bahadir observes. 

"If you go to any of the hyperscalers, you get a chatbot functionality and you can build that chatbot within an hour. Right now it's getting easier and easier."

Navigating generational expectations

Perhaps the most significant challenge facing ING involves managing divergent customer expectations across different demographic segments. 

The bank serves tens of millions of clients across Europe, each with varying levels of comfort and enthusiasm for AI-powered services.

"First generation AI adoption and expectation from the clients is a bit of a generational thing," Bahadir explains. "There are certain client segments that really believe in it and they want it, and there are certain client segments that are really sceptical."

This generational divide requires careful navigation, particularly for an established institution with extensive mortgage portfolios and long-standing customer relationships. 

The challenge extends beyond simple technology deployment to encompass broader questions of customer experience design and service personalisation.

"Our job is balancing these expectations and providing experience that is appealing for both sides and providing the optionality," Bahadir concludes. "I think this is also part of the personalisation challenge that everybody's facing."