An ambition that is partly carried out by AI leans on a chain. Data comes from somewhere, software runs at someone's site, integrations go through a third party. The question is not whether your suppliers use AI. The question is whether their way of working allows part of the work to shift to AI execution with human oversight, without delivery grinding to a halt on your end.
That is a different question than before. In the past you tested a supplier on price, delivery time, and quality of the end result. Now you also test something under the hood: can this party demonstrate which part of their own work is already done by AI, which part is under oversight, and which part remains human work. A supplier who cannot explain that also cannot demonstrate that in two years they will still match the pace and price of competitors who can.
Some suppliers have already sorted their own work into the three categories: what AI takes over, what AI does under the control of a human who approves or rejects, and what remains human work. Those suppliers can tell you precisely why their turnaround time is getting shorter or their price is staying stable. Other suppliers have not made that breakdown. For them, AI sits somewhere in the process, but no one can say which part of the capacity it affects or how reliable that part is.
The difference is not in the sector and not in the size of the supplier. It lies in whether someone within that organization has broken down its own tasks and looked, task by task, at what is transferable. A supplier who did this can sometimes look the same on paper as a supplier who did not. The difference becomes visible as soon as you probe for evidence instead of promises.
A promise is not evidence. "We use AI" says nothing about which part of the work it affects and how stable that is. Evidence that does count:
If this evidence is missing, that does not automatically mean the supplier is unsuitable. It means the readiness is unknown, and that is a different fact than unreadiness.
Three outcomes are possible, and each calls for something different.
The supplier can provide the evidence and the breakdown holds up under probing. Then the question is answered and the test moves on to the next link in the chain.
The supplier cannot provide the evidence, but is transparent about where the breakdown is still missing. That is a starting point for a conversation about timeline, not for a decision today. Some suppliers are on their way and need time to make that breakdown; that pace belongs with the phasing of your own ambition, not separate from it. Here the question touches on how you phase an ambition that is too big for one year, because a supplier's readiness rarely runs on the same timeline as your own.
The supplier cannot provide evidence and also cannot explain why not. That is a risk to your own operational feasibility, and that question is separate from what you do with this supplier going forward. What an employer does with a supply contract or with own personnel affected by this outcome falls under its own legal and contractual requirements; this test provides no basis for that.
Supplier readiness does not stand apart from the rest of the ambition. It is connected to the operational feasibility of the ambition itself, to the margin the ambition needs to deliver, and to how a vision is made measurable before you go on to test suppliers against it. A supplier test that stands apart from these other questions measures something, but not the right something.
The underlying question — which work in your own company, and at your suppliers, can genuinely be taken over by AI — is answered task by task with the work scan from FTE TO AI.
This page describes a gate within the ambition test: eight dimensions, five confidence gates, a translation back into roles and decision rights. The full ambition test is under construction. What is already available now: a free readiness check of eight short questions, one per dimension, giving you a picture of where your organization is furthest along and where it lags behind. That picture is a starting point for the conversation with suppliers, not the end point.
Vertel wat u wilt bereiken, dan kijken we samen wat daarvoor moet staan.
Answers come from this site’s knowledge base. Not tailored advice, and not a scan of your company.