Generative AI: a means of developing a skills-based approach
By helping to personalize learning paths, artificial intelligence enriches the interactive dimension of learning systems, enabling the skills-based approach to be fully deployed.
Chrysta Pelissier, University of Montpellier
If the changes that have begun in education with and through digital technology have been an interesting phenomenon to study since their emergence, the release of ChatGPT 5 (GPT-5), on August 7, 2025, by OpenAI, gives these research questions particular acuity. The pace of change in educational practices has accelerated considerably in recent months, and technological tools such as generative artificial intelligence (AI) have become increasingly important.
Teachers, pupils, students and training managers have embraced these new ways of transmitting information and sharing ideas.
Can this range of tools help to develop new ways of thinking about learning, with a focus on a skills-based approach?
The competency-based approach: a fast-changing process
The competency-based approach took shape with the national framework text for courses leading to the award of national bachelor's, professional bachelor's and master's degrees, in 2014. It was supplemented in 2018 by the decree relating to the bachelor's degree, and in 2019 by the decree relating to the professional bachelor's degree.
This approach is driving profound new changes in universities. It is defined as an approach to teaching that attempts to overcome some of the limitations of objective-based teaching.
For teachers, the changes associated with the competency-based approach are mainly to be found in the ways in which knowledge is transmitted and the help given to individual students. In terms of transmission methods, the competency-based approach appears to be a movement in which teaching objectives are no longer defined as content to be transmitted, but rather as action capabilities that learners must develop in order to carry out their activities in a professional situation.
Confronted with these new professional situations, students are encouraged to identify their own needs in terms of knowledge, know-how and interpersonal skills, and then set about selecting personal knowledge (already acquired), researching documents and asking questions of professionals.
Examples of how generative AI can be used by learners
Firstly, in activities that stimulate discovery-based learning, learners can explore and ask questions. For example, they can ask a generative AI to suggest situations specific to a job they are considering, to give them a glimpse of their diversity, but also to help them formulate a hypothesis for solving each one.
Then, the learner can ask the AI to give him examples of jobs in line with his own skills, which he will have identified himself. Please note that the aim is not to impose a particular profession, but rather to refer the learner to possible professions. It will then be up to them to verify the missions associated with these jobs, and the skills that go with them.
Finally, ChatGPT can help you " learn to learn " by enabling you to plan your revisions ("What are your priorities?"), identify a time allocation ("What are your breaks?"), consider a working method ("What did you find easy today? "Why?", or "Could you have done it differently?"), to benefit from encouragement in the formulation of personal objectives ("What do you want to understand today?", "What was your objective for the day?", "Are you proud of yourself?").
Examples of how generative AI can be used by teachers
Firstly, original data can be used to devise teaching scenarios, leading teachers to propose escape games, quizzes, board games or even immersive contexts, specific to a profession targeted by the training.
Secondly, generative AI usage practices can propose activities aimed at developing skills that the learner has selected himself or herself, those targeted by the teacher, but also answering questions related to a particular concept that may not have been understood in class.
By analyzing multiple media, AI tools can help identify collective needs (on the same class) or create training materials, explanations, feedback in different formats (texts, podcasts, videos and image montages), adapted to different learners or groups of learners.
Finally, the teacher can propose activities aimed at critical thinking, using different methodologies for creating prompts, comparing the results of several AIs and developing the ability to argue.
This article is published as part of the Fête de la science (which took place from October 3 to 13, 2025), of which The Conversation France is a partner. This year's theme is "Intelligence(s)". Find all the events in your region on the Fetedelascience.fr website.
Chrysta Pelissier, Senior Lecturer in Language and Educational Sciences, University of Montpellier
This article is republished from The Conversation under a Creative Commons license. Read theoriginal article.