Since its release in November 2022, ChatGPT has spread rapidly among the public. Although the adoption of technical innovations is normally rather slow in the education sector, ChatGPT has also very quickly arrived in the classroom, followed by other AI tools (Trautmann 2024). Thus, the use of AI tools in educational contexts has recently become a much-discussed topic. While some see it as a danger – for example due to the possibility of faking performance – others see it as a great opportunity – especially regarding the individualisation of learning processes (Helm et al. 2024).

Findings from various recent surveys in German-speaking countries suggest that many students are already using ChatGPT for school related tasks, in particular to complete homework and written assignments, to prepare for exams or even to cheat on exams. They also show that currently AI tools are being used significantly more by teachers in vocational secondary schools than by teachers in other types of school (Helm et al. 2024).

Against this background, the project presented takes a closer look at business academies (as one important form of vocational secondary schools in Austria) for which no specific findings had been available so far. The aim was to investigate to what extent business academy students currently use AI tools and which factors influence their future intention to use AI tools for business subjects as well as to find out more about their attitude towards AI tools. Following on from these empirical research questions, a further aim of the project is to discuss, from a didactic perspective, what implications can be derived from the findings for the future design of business education.

Prior to developing a questionnaire, a comprehensive literature review was carried out in order to ascertain the current state of research (for an overview see e.g. Helm et al. 2024, Huber et al. 2024, Ipek et al. 2023, Pishtari et al. 2024). At the same time, a pool of items for the descriptive part of the questionnaire was compiled from existing surveys (e.g. Bhattacharya 2023, Hassler/Wegmüller 2024, Schiel et al. 2023, Schlude et al. 2024, Stojanov et al. 2024, von Garrel et al. 2023).

In order to analyse which factors influence students' future intention to use AI tools, the Unified Theory of Acceptance and Use of Technology was used. The original UTAUT model (Venkatesh 2003) consists of six variables: use behaviour is predicted by behavioural intention and facilitating conditions; in turn behavioural intention is predicted by performance expectancy, effort expectancy and social influence. Facilitating conditions refers to the perceived level of usage support provided by organisational and technical infrastructure. Performance expectancy refers to the extent to which an increase in performance is expected as a result of use; effort expectancy refers to the ease of use; social influence refers to the extent to which the individual thinks that relevant people consider the use to be important. The UTAUT model has been empirically supported in numerous studies in various contexts. It has also proven its worth in a variety of studies in education (Or 2023). In this study the model was used in a slightly modified form as proposed e.g. by Schopf et al. (in print) for the analysis of the factors influencing the acceptance of explanatory videos in the context of business education. Since behavioural intention and (actual future) use behaviour cannot be meaningfully surveyed in a cross-sectional study, only behavioural intention was taken into account. Because, contrary to theory, some direct effects of facilitating conditions on behavioural intention have been observed in several studies, this variable was nevertheless included.

In May/June 2024 an online survey was conducted at eight Viennese business academies. In total, 555 students took part, mainly from years II and IV (age range 15 to 19; 65 % female). The questionnaire consisted of two parts. The first, descriptive part contained a variety of questions to find out which AI tools the students know and use, for which subjects and how often they use them, for which tasks and in which way they use them, whether they have had positive or negative experiences with AI tools so far as well as what effects the students themselves believe AI use has on their motivation and learning success. Many of these questions were formulated in a semi-open-ended manner so that the students could add other known/used AI tools, tasks, types of use, etc. in addition to those mentioned. Furthermore, some open questions were asked about what the students associate with the term AI and what specific advantages and disadvantages they see in the use of AI. Some of these items were adopted and adapted from existing studies, while others were constructed from scratch. The second, explanatory part aimed at analysing the factors influencing the future intention to use AI tools for business subjects. Therefore, the five variables of the UTAUT model were operationalised with four items each. The basis was the questionnaire used by Schopf et al. (in print), which in turn was based on the original questionnaire by Venkatesh et al. (2003) and the questionnaire by Bardakci (2019) tailored to the use of YouTube. In addition to descriptive analyses of the closed questions and content analysis of the open questions in the first part of the questionnaire, regression and dominance analyses (Budescu 1993) were carried out on the basis of the data from the second part of the questionnaire. Complementary to the quantitative survey, six focus groups were conducted in the same period with a total of 38 students in year IV (age range 17-19; 38 % female) from four different business academies in Vienna and Lower Austria. The aim of these discussions was to gain a deeper understanding of students’ AI use, particularly with regard to use behaviour, perceived advantages and disadvantages as well as wishes for the future. The focus group discussions were fully transcribed and interpreted by use of a content-structuring qualitative content analysis strategy (Kuckartz/Rädiker 2024).

The findings provide an initial comprehensive picture of business academy students’ current AI use. ChatGPT and Google Translate are clearly the number one AI tools, used by 93 % of the students. 86 % also use them for school related tasks, mainly in languages but also in business subjects. The most frequent tasks are researching information, answering comprehension questions, finding ideas, translating, summarizing, correcting or writing texts. Students predominantly report positive experiences. Although they see many advantages, especially in terms of saving time, they assess the use of AI tools for school related tasks quite critically. What is more, the focus group discussions reveal considerable fears in relation to the rapid development of AI. The UTAUT based analyses confirm that the most important predictor of students’ future intention to use AI tools for business subjects is performance expectancy. Implications for the design of business education include that AI tools should be implemented in business teaching in order to give individual learning support and to enable students to use those tools in a helpful and transparent way and to reflect on the results. In this context, it is important to communicate or negotiate clear guidelines on the use of AI tools for school related tasks. At the same time homework and examination/performance assessment practices need to be reconsidered.

Bardakci, S. (2019): Exploring High School students’ educational use of YouTube. International Review of Research in Open and Distributed Learning, 20(2), 260-278. Bhattacharya, K. (2023): Lernen mit KI. Einsatz von ChatGPT & Co. beim Lernen. IU Studie. https://static.iu.de/studies/Kurzstudie-Lernen-mit-KI.pdf (accessed: 13.05.2024). Budescu, D. V. (1993): Dominance analysis: A new approach to the problem of relative importance of predictors in multiple regression. Psychological Bulletin, 114(3), 542–551. Garrel, J. von/Mayer, J./Mühlfeld, M. (2023): Künstliche Intelligenz im Studium. Eine quantitative Befragung von Studierenden zur Nutzung von ChatGPT & Co. Hochschule Darmstadt, https://doi.org/10.48444/h_docs-pub-395 (accessed: 11.05.2024). Hassler, D./Wegmüller, R. (2024): ChatGPT, DeepL und Co. im Unterricht. Herausforderung und Anwendung in der beruflichen Grundbildung am Beispiel der Kaufleute in der Schweiz. schule verantworten, -(1), 25-34. Helm, C./Große, C.S./öbv (2024): Einsatz künstlicher Intelligenz im Schulalltag - eine empirische Bestandsaufnahme. Erziehung und Unterricht, -(3-4), 370-381. Huber, S.G./Klein, U./Lussi, I./Schneider, N./Hoffmann, J./Wathiyage Don, A. (2024): Bildung im Kontext der digitalen Transformation in Deutschland, Österreich und der Schweiz. Überblick über zentrale Studien und Ergebnisse. schule verantworten, -(1), 51-65. Ipek, Z.H./ Gözüm, A.I.C./Papadakis, S./Kallogiannakis, M. (2023): Educational Applications of the ChatGPT AI System: A Systematic Review Research. Educational Process International Journal, 12(3), 26-55. Kuckartz, U./Rädiker, S. (2024): Qualitative Inhaltsanalyse. Methoden, Praxis, Umsetzung mit Software und künstlicher Intelligenz. 6th edition, Weinheim/Basel: Beltz. Or, C. (2023): Revisiting Unified Theory of Technology and Use of Technology using meta-analytic structural equation modelling. International Journal of Technology in Education and Science, 7(1), 83-103. Pishtari, G./Wagner, M./Ley, T. (2024): Ein Forschungsüberblick über den Einsatz von Künstlicher Intelligenz für das Lehren und Lernen in der Hochschulbildung. Bericht für Arbeitspaket 3 (Preprint). Schiel, J./Bobek, B. L./Schnieders, J. Z. (2023): High School Students' Use and Impressions of AI Tools. ACT, https://www.act.org/content/dam/act/secured/documents/High-School-Students-Use-and-Impressions-of-AI-Tools-Accessible.pdf (accessed: 12.05.2024). Schlude, A./Mendel, U./Stürz, R. A./Fischer, M. (2024): Verbreitung und Akzeptanz generativer KI an Schulen und Hochschulen. https://www.bidt.digital/publikation/verbreitung-und-akzeptanz-generativer-ki-an-schulen-und-hochschulen/ (accessed: 23.05.2024). Schopf, C./Schiffinger, M./Raffer, P. (in print): YouTube & Co – Zur Nutzung und Akzeptanz von Erklärvideos als Ergänzung des Wirtschaftsunterrichts. In: Loerwald, D. / Goldschmidt, N. (Eds.): Digitalisierung in der Ökonomischen Bildung. Springer Gabler. Stojanov, A./ Liu, Q./Koh, J.H.L. (2024): University students’ self-reported reliance on ChatGPT for learning: A latent profile analysis. Computers and Education: Artificial Intelligence, -(6), 1-8. Trautmann, M. (2024): KI ist in der Schule angekommen. Pädagogik, -(3), 6-10. Venkatesh, V./Morris, M.G./Davis, G.B./Davis, F.D. (2003): User Acceptance of Information Technology: Toward a Unified View. MIS Quarterly, 27(3), 425-478.