How can AI skills be identified and developed at a time when jobs are changing at such a rapid pace? The rise of artificial intelligence tools is already transforming many businesses, forcing companies to rethink their approaches to skills management and training. But with uncertainties about how these tools will be used, internal policies that are still unclear and difficulties in expressing needs, how can we take concrete action? Philippe Argouges, an expert in training engineering and needs analysis, provides you with operational levers and best practices to support these changes.
According to the’OECD, In the United Kingdom and the United States, a third of the skills required for the average job have changed. In the age of AI, we might even wonder whether this figure is still underestimated.
The rapid evolution of AI skills
Two years ago, AI tools existed, but apart from a few geeks, very few professionals were using them. Mainstream newspapers were writing about it as if it were a revolution. And the question they all raised was: Will software soon be more intelligent than human beings? A philosophical question, of course, but not very interesting from a practical point of view.
The right question was: how to use it effectively? In other words, What skills will be needed?
The fact remains many professions have undergone profound changes with the emergence of these tools.
Those linked to IT development, for example. Today's artificial intelligences, even generalist ones, code well. Very well indeed. For a developer, mastering these tools and using them wisely has become an absolute necessity. key competence. This can save most business projects weeks. Of course, this was not the case two years ago.
The same applies to communication professions. From now on, anyone can retouch or create images. It's no longer the preserve of graphics tablet or Photoshop specialists. For a graphic designer, Expertise in the use of AI tools is now a key skill.
And all that in just two years. Even less, if you think about how these tools were used by the uninitiated at the time. For many of them, it was an overnight process. Adapt or disappear, in short.
You might think that this only concerns a few professions and that, in the majority of cases, nothing will change. AI is a powerful tool, but we can do without it...
Is it?
In reality, just ask any professional. They all have a possible use for AI in their business, and they all need skills.
[Also read]
What is the real impact of artificial intelligence on our professions? What new roles and skills are emerging from this revolution? Find out in the ORSYS white paper The new AI professions - The impact of AI on professions.
Skills management in the age of AI
It is therefore vital to plan for the development of your teams' skills in this area.
Tools...
Fortunately, we have a powerful tool for this: GPEC, forward-looking management of jobs and skills. In simple terms, this enables us to map skills and plan requirements in order to build an appropriate training plan.
[Training]
The problem is that it takes time to draw up these maps. Hence the risk of sclerotising the analysis process when, on the contrary, we should be more responsive.
... and obstacles
And when it comes to AI-related needs, employees often come up against these two obstacles.
1/ Companies have no clear policy
Generalist AI tools for the general public offer few guarantees as to the confidentiality of the information shared. Today, we don't know how ChatGPT, Gemini and all the others protect the information entrusted to them. Isn't there a risk that confidential data will be made accessible to everyone? For many companies, that's a risk they can't afford to take.
Several options are available.
They can develop their own tools, This is an internal, closed system for employees only. For example, they could invest in an engine like Mistral. The learning phase can take place with defined information only: private data remains within the company's ecosystem.
They can also invest in a paying tool and contract with a supplier to guarantee the confidentiality of their information.
Of course, employees must refrain from using the tools available to the general public.
2/ Users find it difficult to identify their needs in terms of AI skills
This is a recurring pitfall in training needs analyses. Users find it difficult to list the skills they lack, precisely because they lack them. This is even truer when it comes to AI.
In a traditional project, users compare themselves to their colleagues or family members, and identify their shortcomings. They can project themselves into their future activity and deduce what they need to know.
This no longer works with AI projects. Users have empirical knowledge, which varies widely. However, many cannot imagine what will be useful to them in their profession, which is changing rapidly.
Sometimes they don't even see the benefits that these tools bring. In fact, they have no idea (or a misconception) about what they can do with them.
Best practices for developing AI skills in practice
Despite this, training managers have a number of levers at their disposal.
1/ Acculturate employees One of the first pitfalls is the varying levels of knowledge about AI products and their possibilities. Training managers can therefore offer seminars that explain and demonstrate what employees can do.
At this printer based in the Paris region, the management set up three seminars differentiated by profession. Assistants, sales staff and IT specialists were able to discover the technology and put it to practical use. For each audience, the demonstrations and practical exercises helped them to choose the tools they would prefer to use in-house. This has helped to boost usage by all.
2/ Bringing AI to life in training courses
As professions evolve, so must training. All subjects can now include AI components, not just dedicated training. It is vital to update existing training courses to take account of these tools and the benefits they bring.
At this trade union training centre, training in letter writing and leafleting includes the use of AI.
These courses describe best practice, particularly for prompting.
They also show bad examples of use and ask participants how they perceive these uses.
[AI skills: the expert's opinion]
In all the training courses I run, I now talk about AI. I show the tools and share how I use them. This technology has consequences in all areas. That's why I've added at least one page on the subject to all my training materials. On some subjects, I have completely rebuilt the entire training course.
3/ Adapting the offering
Each company has its own practices. The training on offer must take this into account, drawing on all the systems available depending on the situation. Sharing experience between peers is undoubtedly an appropriate solution for this type of project. But the training manager can also imagine training courses created in-house, or even dedicated asynchronous training courses.
A transport company used Genially to create short training modules dedicated to AI.
The aim was to introduce fun tools to measure employees' knowledge of the subject.
Participants carried out these activities in sub-groups to facilitate discussion.
And how are you doing in your business?
- Is your company's AI policy clear?
- Do all your employees know what they can do and what IT management won't let them do?
- Has your IT department already chosen tools that it recommends using or giving priority to?
- Have any seminars on the subject already been held?
- Do you regularly provide training in the use of AI?
- Have you set up peer-to-peer experience sharing?
- Do the training courses you organise include an AI component?
- Do your in-house trainers use AI to create and deploy their training courses?
In conclusion, AI is no longer just a tool for creating memes or amusing images. There are now a whole host of products available under this term, both free and paid for. Your employees may be familiar with them, but for maximum efficiency, they need to be properly trained and choose the recommended tools. The role of the training manager is to help them. To do this, it is essential to know where they stand. Then, and only then, can he or she propose an appropriate training offer.
