In an article that is part of a number of items ahead of a research report, this news service talks to Swiss private bank Piguet Galland about how AI features in its business.

This article continues WealthBriefing’s interview series ahead of its forthcoming AI in Wealth Management research report, produced in partnership with Arta Finance.

For Swiss private bank Piguet Galland, AI’s value lies in helping employees work more efficiently while preserving the personal relationships on which private banking depends.

Jordi Domenech (main picture), who leads the bank’s technology function, describes this as the “augmented banker”: a human advisor equipped with better tools, whose expertise and judgment remain central to the service.

Clients may welcome quicker processes, but important conversations about their finances still call for human contact. The bank is still cautious about putting AI directly in front of clients.

Practical gains from everyday tools
Domenech said earlier investment in cloud infrastructure, document classification and employee training helped the bank introduce Microsoft Copilot quickly. He personally delivered voluntary training, with usage initially building gradually before gaining wider acceptance.

His priority was measurable improvement in everyday work. He reported efficiency gains of around 10 per cent, explaining that across a workforce of 180, even modest individual improvements could have a substantial collective impact.

The approach also gives employees room to experiment. Domenech said about 10 in people in business teams had built prototypes that changed the way in which they worked, without relying on IT to develop the initial solution.

Those experiments are treated as prototypes. Before wider deployment, the technology team reviews them, applies appropriate guidelines and integrates them into the bank’s systems.

For a smaller institution, the challenge is to combine the speed of experimentation with controls that make new applications dependable. Employees need both permission to explore and a clear route towards supported operational use.

More recently, Piguet Galland deployed internal large language models on its own infrastructure, opening up experimentation with internal data. Potential operational applications include ticketing and help-desk processes. Domenech described a differentiated approach, matching tools and models to technical specialists, operational teams and everyday users.

Avoiding a divide between employees
For Domenech, adoption is as much a workforce question as a technology project. He warned against creating two groups within the same bank: employees advancing rapidly with AI and colleagues who feel increasingly left behind.

“We should not leave people behind,” he said.

He challenged the assumption that younger employees necessarily bring the strongest understanding of how to apply AI. Familiarity with a tool does not automatically confer knowledge of the business processes it might improve.

Experienced colleagues hold expertise that can help shape useful applications. Pairing them with younger employees offers a way to develop tools while transferring knowledge and teaching people how to assess the results critically.

That transfer also matters when experienced staff retire. Domenech sees scope to capture elements of their knowledge in reusable agents and skills, while ensuring that successors understand how to use them wisely.

Expertise must remain visible
AI is also changing the questions clients bring to their advisors. Domenech described the incidence of clients arriving with conclusions generated by public AI tools and expecting their banker to respond.

Simply dismissing those answers risks weakening the relationship. Instead, he argued, advisors should be equipped to examine them against reliable information, the client’s circumstances and the bank’s expertise.

The banker’s value lies in explaining what is relevant, what may be misleading and which risks have been overlooked. Human judgment remains essential to maintaining confidence in the advice.

“The machine is just there to assist,” Domenech said.

The forthcoming WealthBriefing report, produced in partnership with Arta Finance, will explore how firms turn AI experimentation into practical value, including the training, governance and organisational changes needed to support adoption.