The wholesale sector rarely takes over the customer relationship or the production process itself, but is right in the middle of it. Orders come in through customers' systems, inventory is aligned with suppliers, and the margin lies in the speed and precision of the link in between. That makes AI use in this sector less visible than in sectors with their own front office. There is no shop floor where someone sees a chatbot talking to a customer. There is a planner who uses a tool to forecast purchasing volumes, an account manager who has quotes drawn up by a language model, a warehouse system with an AI component that no one recognizes as such anymore because it has been running under a different name for years.
The IT department of a wholesale company generally manages the core systems: the ERP, the WMS, the links with customers and suppliers. What grows on top of that — spreadsheet tools with a built-in forecasting function, a separate subscription for text generation, an AI plugin in the CRM that was activated by a single team — often falls outside that administration. That is not because anyone is hiding anything, but because no one has kept track of it. Anyone in this sector who wants to know what is actually running has to ask the people who use it. And that only works if the question doesn't feel like an accusation. A buyer who uses a tool to predict supplier reliability will only mention it once they are sure the answer won't be used against them.
Not every use of AI in the wholesale sector carries the same weight. A tool that checks invoices for discrepancies has a different impact than a system that automatically adjusts prices based on supply and demand, or a model that rates suppliers and thereby influences who gets a contract. The classification depends on what the system decides, for whom, and what happens if it goes wrong. A pricing algorithm that structurally miscalculates hits the margin directly. A tool that only summarizes internal reports mainly affects the readability of a report. These differences determine what level of oversight, documentation and review fits a given application. The exact regulatory content that applies is not covered on this page; that changes and is tracked elsewhere. What matters here is the mechanism: knowing what is running, knowing what risk it carries, and recording that in a way that a board or supervisory authority can inquire into.
What distinguishes the wholesale sector from many other sectors is the extent to which AI use does not start within the company itself. Customers submit automated orders, suppliers send forecasts about availability, and between those two flows sits a wholesale company that also uses algorithms itself to coordinate between them. A governance approach that only looks at its own organization misses part of the picture. The question is not only which AI runs internally, but also which automated decisions come in from outside and how these are checked before they affect inventory, price or delivery time.
A scan that starts from what is actually being used, rather than from what is on the IT list, typically brings a layered picture to the surface in the wholesale sector: a core of approved systems, a middle layer of tools that have become functional without formal review, and a fringe of individual use that was never reported. These three layers do not call for the same approach. Some applications can simply be documented and monitored; others call for a conversation about why they started being used and what the owner expects from them. The result is not a verdict on who did something wrong, but a basis on which a board can explain what is going on and why it is under control — or where that is not yet the case.
The way AI spreads within an organization differs by sector, but the underlying movement is comparable. In the manufacturing industry, AI is often interwoven into production planning and quality control, in the transport sector it mainly plays a role in route planning and delay prediction, and in professional services it mainly turns up in reporting and customer communication. Anyone working in the wholesale sector will recognize elements from all of these sectors, because the chain runs through all of them.
An inventory of what AI is running shows where the risk lies and where oversight is missing. It does not answer another question that is just as pressing in many wholesale companies: which part of the daily work can actually be taken over by AI, and which part cannot. That is a different calculation, focused on tasks rather than on systems. FTE TO AI offers a work scan for this purpose that calculates, per task, which part of the work can be taken over by AI, as a follow-up step once it is clear what is already running and under what oversight it should fall.
The Responsible AI Scan for the wholesale sector is currently being built. Anyone who would like to be contacted as soon as the scan becomes available can sign up for the waiting list.
Vraag maar. Governance begint bij weten wat er draait — ook wat niemand heeft goedgekeurd.
Answers come from this site’s knowledge base. Not tailored advice, and not a scan of your company.