Header

School of Applied Linguistics

Automatic text production

With the ongoing digital transformation, new technologies like ChatGPT can offer valuable support for writing. The question is how.

How can AI and ChatGPT be integrated effectively into teaching – and what challenges come with it? How is ChatGPT changing management and administrative work at universities? These are the questions that the in-depth project launched in spring 2023 addressed empirically. 

Our data was based on 20 focus group discussions conducted with a total of 70 individuals in autumn 2023. These individuals also completed questionnaires. The participants intentionally reflected a broad spectrum, including lecturers as well as employees from the respective president's offices and administrative teams. 

We analysed the verbal data using content analysis and evaluated the results with respect to two major thematic areas. Firstly, the encouraging best practices that emerge from the data. Secondly, the unanswered questions and uncertainties that remain. Both these thematic areas served as the framework for our detailed overall report.

Overall evaluation

The overall evaluation of the questionnaires and focus group discussion provides an insight into the current state of generative AI usage in university teaching. The results show that a majority of respondents have already integrated generative AI into their teaching methods or are planning to do so in the near future. In addition, the data reveals the most common areas of application of generative AI as reported by the respondents as well as the most important aspects in connection with the use of AI from their point of view. The findings offer a window into current practices and future challenges in dealing with generative AI in education. The evaluation and the results are presented in detail in section 3.1 of the overall report.

«Also zusammenfassend: auf Augenhöhe mit ChatGPT (.) sein, in die Lehre integrieren und sinnvolle Aufträge für die Studierenden zur Verfügung stellen.»
(Dozent:in für wissenschaftliches Schreiben)

Best Practices

The respondents reported on the specific use of generative AI in teaching as well as the use of the technology for preparing and developing teaching content. They also discussed ways to strengthen their own methodological skills and those of their students, and reflected on suitable methods for assessing writing skills. We have extracted these statements and categorised them as best-practice examples for different areas of application. We also address any areas of uncertainty in dealing with generative AI. A detailed overview of these results can be found in section 3.2 of the overall report.

«And I thought, okay, there's obviously there's got to be work done and we structured and it was very experimental (.) experimental and the feedback I think was very positive. What I saw in the classroom, what they were doing for half an hour and I just went around saying, that works, that doesn't kind of thing, but it works. But you have to find a way to do it.»
(Dozent:in für Englisch)

Open questions

Across the board, the respondents raised a wide range of questions at various levels that arise as a result of the widespread use of AI-based applications such as ChatGPT. Such questions concerned

A systematic overview of these questions can be found in section 3.3 of the overall report.

«Par contre, on a un devoir et ça je sais pas comment on va le faire, mais un devoir de (.) réflexion critique sur le sujet, de leur apprendre à être critique avec cet outil, de leur apprendre à l'utiliser judicieusement, (.) de les sensibiliser à ça (.) plutôt que de devoir punir ou de ou de faire l'autruche comme si voilà, ça, ça n'existait pas. Cette réalité existe, faut en tenir compte. Je [ne] sais pas comment, dans les différents niveaux de responsabilité qui sont les miens je vais mettre ça en place ?»
(Institutsleiter:in Bereich Sprachen)

Conclusion and outlook

Generative AI holds great potential for university education – both for innovative teaching methods and the development of new skills. It is important, however, to take ethical, legal and social concerns seriously. Clear guidelines and critical engagement with the technology are necessary in order to harness the opportunities presented by AI without jeopardising the quality of academic work and the critical thinking of students. A detailed conclusion can be found in section 4 of the overall report.

«Ist (.) es vielleicht einfach so den Aspekt, dass wir eben die Hochschule auch nicht (.) isoliert betrachten, dass sich eben wirklich auch die Welt ausserhalb, auf die wir ja ausbilden wollen, verändert durch diese Tools und dass wir da wirklich am Ball bleiben müssen, auch wie wir diese Kompetenzen vermitteln können […] Sodass, dass wir nicht nur sagen, wie ändert sich unsere Lehre im Sinn von Didaktik, sondern auch wie ändern sich halt (.) die Fachinhalte, die Skills, die die Studierenden und später haben (.) müssen.»
(Studiengangleiter:in Bereich Wirtschaft)

Overall report

Download PDF(PDF 639,0 KB)

The overall project

As part of the Digital Literacy in University Contexts project, we worked on AI-based technologies for the reception and production of texts in academic contexts. We offer students, teachers and writing consultants tutorials, services and good practices that support the reflective use of AI for text reception and production in academic contexts.

Projektteam Automatische Textproduktion

Contact

We look forward to receiving your feedback!

­