Abstract:

Artificial intelligence (AI) is a new instrument with great applicability in multiple areas, including education. It is possible to apply AI as an intelligent tutoring system that helps students in their learning process [1-2]. It is also applicable in evaluation processes (including auto-assessment). Currently, first-year university students are users of artificial intelligence, however, in a few years, students will be native to AI.

The research team surveyed first-year university students in the field of engineering to explore their perception and use of AI. The responses, collected in 2024, allowed us to verify the artificial intelligence software the students used (or knew). As expected, they mostly use ChatGPT and/or browser artificial intelligence. On the other hand, the information obtained allowed us to check their interest in the application in the learning or assessment processes, the interest being greater in the first case. Students generally prefer to use AI as a self-assessment tool (hence formative assessment) and not so much as a professor assessment tool (summative assessment).

For the analysis of the perception of the risks and potentialities that students encounter in the application of AI, several questions have been asked using a Likert scale (from 0 to 5) in a survey. Regarding the positive aspects of the use of artificial intelligence in the educational field, the most valued aspects are the management of time (and the reduction of the time needed to perform certain tasks) and the easy accessibility of AI instruments (both aspects with ratings ≥4.0 over 5). With regard to the risks or dangers of AI, students are not highly concerned about this issue, with the highest value is associated with the risk of AI providing erroneous information being 2.95 out of 5. One of the aspects to note is that a risk identified by some students as very important is that AI affects jobs, replacing them and not being able to get a good job upon leaving university.

A correlation matrix study of the correlations between all questions and answers was carried out. The best positive correlation is associated with the responses on two risks of AI: the decrease in critical capacity and the generation of a false sense of learning. The most negative correlation is found between responses to the positive perception that AI helps improve evaluation results and the perception of the risk of obtaining misinformation. The results of this study should be complemented with the analysis of students' perceptions after the completion of learning and assessment tasks designed using AI.

References:
[1] W. Bagunaid, N. Chilamkurti, P. Veeraraghavan, “AISAR: Artificial intelligence-based student assessment and recommendation system for e-learning in big data,” Sustainability, vol. 14, art. 10551, 2022.
[2] Z. Swiecki, H. Khosravi, G. Chen, R. Martínez-Maldonado, J.M. lodge, S. Milligan, N. Selwyn, D. Gasevic, “Assessment in the age of artificial intelligence,” Computers and Education: Artificial Intelligence, vol. 3, art. 100075, 2023.

Keywords:

Artificial intelligence, perception, university, survey.