Corporate AI training, on your processes and in your own tools

In the catalogues of the large providers, expect anything from a few hundred euros excl. VAT per participant for a few hours of awareness to more than €2,000 excl. VAT for a two-day course. Custom training, organised for a group within the company, is generally priced on a quote. That is the case at Kreante, and measuring your teams’ level is free.

To be funded by an OPCO (the French skills operators), or by other public or pooled funders, a training course must be delivered by a Qualiopi-certified provider^[5]. The CPF (France’s personal training account) only funds certain courses, notably those leading to a certification listed in the RNCP or the Répertoire spécifique^[6]. Custom training built for one company therefore falls rather under the skills development plan, funded by the company and, where applicable, its OPCO.

Yes, and they are often a good starting point. The French Ministry of the Economy lists several^[4]:

• Bpifrance Université offers an AI self-assessment and a path of about ten hours,
• France Num recommends several MOOCs, such as Objectif IA, about six hours long, or a 30-hour course designed by the University of Helsinki,
• Inria offers “ChatGPT expliqué en 5 minutes”, to quickly understand how the tool works.

These courses explain the basics well. They do not replace work on your own tools and your own tasks, which is often where everything is decided.

No. The training is for every function, and the initial measurement makes it possible to adapt the content to each person’s level. An employee who has never used AI and another who uses it every day will not follow exactly the same path.

Both. The training sessions are at its heart, but they rest on a level measurement, a study of each department’s real work and the appointment of an internal lead. We work alongside your teams throughout the engagement, then step back once the new practices are in place.

By asking exactly the same questions, to the same people, a few months later. Comparing with the starting point shows what has improved, role by role, and what still needs work. It says far more than a satisfaction survey filled in on the way out of the room, while everyone still has the taste of coffee in their mouth.

The ones they already use, or that you have decided to roll out: Microsoft 365 Copilot, ChatGPT, Gemini or an assistant developed for your company. The exercises are done directly in these tools, with your own documents, in line with the confidentiality rules set at the start.

Training is only a beginning. For new practices to take hold, three things make the difference:

• clear rules on the tools allowed and the data that can be entrusted to them,
• an internal lead, available to answer questions and share good practices,
• a new level measurement a few months later, to see what has improved and what still needs work.

And if it is management that is not ready yet, that is where to start: without rules or an example set from the top, teams rarely get very far.

When the training reveals tasks that would benefit from automation, we can also help you design those solutions.

For the vast majority of jobs, AI transforms work more than it eliminates it. According to a study by the International Labour Organization published in 2025, one worker in four worldwide holds a job partly exposed to generative AI, a share that reaches about a third of jobs in high-income countries. Its authors conclude that the transformation of jobs is the most likely effect, as most roles rely on tasks that require human input^[7]. Training your teams is precisely about preparing them for this change in their roles.