How FinTech Customer Experience Leaders Are Driving Enterprise Loyalty in Age of AI

While recent surveys show AI adoption is broadening, many companies are still struggling to move past pilot programs and fully integrate solutions. One reason for the gap is enterprise buyers still expect best-in-class human interactions, backed by smart internal systems. Building that requires equipping teams with the skills and confidence to use new technology responsibly.

James Jackson, a senior executive specializing in customer experience within the fintech industry, shares his insights. As the current SVP of Customer Experience at APEXX Global, his career at firms like Worldpay and CACI gives him a sharp take on a tricky problem: how to foster an AI-first mindset without sacrificing the human element that builds trust and loyalty.

For Jackson, the case for augmenting rather than replacing human agents comes down to the nature of B2B relationships. Enterprise clients often demand context, judgment, and personal connection—qualities that AI struggles to replicate. “Enterprise businesses are not looking for a bot to replace a human interaction," Jackson says. "They are looking for a best-in-class customer experience, however that is brought to them.”

• Scenario, not data: His view is backed by data showing customers remain loyal to companies that prioritize human interaction. The real challenge, then, is empowering teams to surface internal knowledge without exposing sensitive information. It's a delicate balance in a regulated space like fintech. "We're not using the customer's data. We're using the scenario and the situation. It's about how we can surface the knowledge that we have internally but not expose anything that is outside of what would be acceptable use," he explains.

Jackson has seen this firsthand during an early training session on generative AI tools, delivered by the award-winning AI education firm Taught By Humans. One non-technical team member, previously hesitant to experiment, was inspired to tackle a recurring manual task. Using an LLM, he created a Python script that reduced a 45-minute daily process to just a minute or two.

For Jackson, the measure of success isn't found in typical KPIs, but in the visible growth of his team’s confidence and initiative. That confidence, Jackson explains, comes from a new approach to problem-solving, as the forgiving, conversational style of LLMs stands in contrast to the unforgiving 'get it right the first time' nature of traditional search.

• The real ROI: By communicating a clear vision for transformation, leaders can lower the psychological barrier to experimentation, fostering an AI-ready culture and encouraging the very confidence needed for real improvements. "I saw a noticeable lift in confidence, and in the team attempting something that may not have been attempted previously. That was all I was really looking for, that uplift in confidence and skill," he outlines.

• Permission to play: This shift in interaction fundamentally alters the risk-reward ratio for employees, transforming AI tools into collaborative partners that actively guide users toward solutions, even when initial queries are imprecise. "It's about having the confidence to try and fail. And you'll realize that the likelihood of failure is now much lower because of what you have available to you to help, assist, and iterate," Jackson advises.

Looking forward, Jackson describes this empowerment scaling into a future of advanced self-service, a vision he finds both thrilling and a bit scary. He is most excited by the rise of agentic AI. "Everything that excites me about what could come down the road is something that could make me redundant. That level of self-service capability could easily be a thing of the next three to five years, if not less. It's super exciting, and a bit scary as well," he observes.

• Architecting experience: But Jackson doesn’t see this future as a threat. Instead, he believes it elevates the human role. In his view, as AI agents take on more mechanical tasks, people can become the designers of the process. The employee who built the time-saving script is a perfect example of this evolution, transitioning from performing a manual task to becoming the architect of an automated one, strengthening the human connection by freeing up time for higher-value work. "Someone's got to come up with the ideas of what the next agent is, so you're thinking more about the customer experience than being part of the customer experience," he says.

But not everyone gets this balance right. Jackson points to the high-profile example of Klarna, which announced it had replaced a large number of support staff with AI, only to backtrack months later. For Jackson, the story illustrates a failure to distinguish between mass-market consumer needs and high-touch enterprise relationships. The lesson is that you can't simply remove the personality from the partnership. As Jackson puts it, "You will not replace those relationships."