How Treasury can lead AI adoption across the enterprise
As AI moves from isolated experiments into day-to-day finance work, treasury has an opportunity to take ownership of its adoption across the enterprise, prove value, build support, and turn pilots into approaches the wider business can use. That's the focus of the latest episode of Invention Mode, where Tanya Kohen, Head of Finance Practice, speaks with Bojan Belejkovski, treasurer, board member and AI enthusiast.
In the latest Invention Mode episode, Bojan Belejkovski and Tanya Kohen discuss how treasury can shape how AI is adopted across the business, drawing on its view across cash, risk, forecasting, banking relationships and working capital.
The conversation follows the path from choosing a first use case and proving its value to building executive support without owning the technology budget, putting the right governance in place, and becoming a reference point that other functions can learn from.
Why treasury is well placed to lead AI adoption
Treasury already works across some of the areas where AI can have a direct effect on business performance, including payments, forecasting, risk, banking relationships and working capital. This gives treasury leaders a close view of the manual work, fragmented data and process gaps that can slow down finance teams and affect decisions elsewhere in the business.
For years, treasury could identify those problems while the route from an idea to a working solution still depended heavily on translating requirements for IT and waiting for the development work to follow. Bojan sees AI as narrowing that gap by giving finance professionals a more direct role in defining, testing and improving solutions around the problems they understand best.
How treasury can build credibility around AI
Start with a problem that can show value quickly
Bojan uses three practical filters when evaluating a potential AI pilot. If an established vendor already solves the problem well at scale, there may be little reason to build another solution. Stronger candidates tend to involve fragmented data that people are manually bringing together across systems and a use case where a visible proof of concept can be produced within weeks.
That approach has taken him into areas such as foreign exchange, insurance risk, credit agreement compliance and accounts receivable, where the underlying business problem is clear and the effect of an improvement can be demonstrated.
Make the internal case with measurable results
Treasury may sponsor an AI initiative without controlling the technology budget, so gaining support depends heavily on how the business case is presented. Bojan frames proposals around measurable outcomes such as time saved, cost reduction or lower risk, giving executives something more concrete to evaluate than broad claims about efficiency or transformation.
He also stresses the importance of working within approved systems, governance and policies from the beginning. Once that framework has been accepted and proven, additional use cases can extend an approach that the organization already understands instead of requiring a completely new case to be made each time.
Become a reference point for other functions
In Bojan's experience, broader AI adoption develops as successful work becomes visible and other teams begin asking how it was done. Treasury can support that process by sharing results, documenting both successes and failures, and making the guardrails clear, particularly when other functions may reuse the same patterns in their own work.
As treasury becomes a source of working examples and reusable practices, its influence can extend beyond individual finance projects into the way the wider organization approaches AI adoption.
Build the foundation for more than one use case
The conversation also covers what needs to sit underneath that growth. Bojan points to data hygiene and governance as essential parts of the foundation, particularly because solutions designed around an immediate quarterly need can leave additional complexity behind when the organization later tries to expand them.
A durable approach gives teams a clearer basis for adding new use cases while keeping data, controls and technology decisions consistent as adoption spreads.
What AI adoption could mean for the treasury role
Bojan also connects AI adoption with the longer-term development of the treasury function. In his view, treasury professionals who learn how to apply AI to finance problems can gain greater influence in areas such as capital allocation and risk strategy because they combine domain expertise with the ability to improve the way decisions and processes work.
For treasurers, CFOs and finance transformation leaders, the episode offers a practical look at how treasury can move from an initial AI use case toward a repeatable approach that earns support across the organization.
