Assessment of three strategies for teaching an AI literacy program, based on a neutrosophic 2-tuple linguistic model hybridized with the ARAS method

This study analyzes some general Artificial Intelligence competency frameworks, aimed at teachers, and it proposes a comprehensive Artificial Intelligence literacy training program, designed to develop in university teachers the necessary skills to integrate Artificial Intelligence into their curricula and research. Through a mixed theoretical-practical approach, the program addresses the gaps identified in the existing literature. The program covers the foundations of Artificial Intelligence to its specific application in teaching, highlighting ethical aspects, critical thinking, and its integration into pedagogy. Based on frameworks such as DigiComEdu, the program is presented as a strategic response to empower teachers, ensuring their preparation to lead the responsible implementation of AI in the classroom and research. One of the issues to consider is how to apply the program. To this end, there were three options: (1) Teaching classes that are 80% theoretical and 20% practical (2) Teaching classes that are 50% theoretical and 50% practical (3) Teaching classes that are 20% theoretical and 80% practical. For the general evaluation and ranking of each of the program's teaching alternatives, a hybrid method between the neutrosophic 2-tuple linguistic model and the Additive Ratio Assessment System (ARAS) was used. Among the advantages of using this method are the possibility for experts to use natural language in their evaluations, the accuracy of the evaluations as more possible states of knowledge are incorporated, and finally the simplicity in the application of the method. Keywords: Neutrosophic 2-tuple linguistic model, ARAS method, AI literacy, AI competencies, Artificial Intelligence, AI literacy program, Teacher training.