Hos SmartLearning benytter vi cookies for at gøre din oplevelse af hjemmesiden bedre. Ved fortsat brug af hjemmesiden samtykker du til vores brug af cookies.   Læs mere     OK
smartlearning logo

Pædagogisk anvendelse af kunstig intelligens 

Af José Bidarra fra Universidade Aberta i Portugal og Henrik Køhler Simonsen fra SmartLearning.


Artiklen her er en omskrevet og forkortet version af en artikel indsendt til Empower Technical Report 2021. Artiklen er skrevet af José Bidarra fra Universidade Aberta i Portugal og Henrik Køhler Simonsen fra SmartLearning.



 The purpose of this small article is to discuss the pedagogical applications of artificial intelligence in higher education and to answer the overall research question: how AI applications may support specific online learning activities in higher education?

The study draws on empirical insights from a structured analysis of nine different cases, describing nine AI applications in higher education in Portugal, the United Kingdom and Denmark, respectively. The analysis of the nine cases focused on particular parameters that may be connected to pedagogical and/or didactical factors related to the actual use of AI in Education.

In a recent article offering a systematic review of research on artificial intelligence applications in higher education, Zawacki-Richter et al. (2019) ask a crucial question. They ask: “where are the educators?” This question has been raised before by Holmes et al. (2019), Rienties et al. (2020) and Simonsen (2020), and it is about time that we discuss potential pedagogical approaches to artificial intelligence applications in higher education. To answer the research question elements from the ABC Learning Design Approach (Young & Perovic (2016) are combined with an emerging framework outlining potential applications of different AI technologies, based on the three-tier classification developed by Holmes et al. (2019).


AI Applications in Higher Education

Existing literature does not fully discuss the potential associations between AI applications and learning activities in higher education. Holmes et al. (2019) do discuss how AI works in education and how different AI applications work in education, but they do not propose an overall pedagogical framework indicating which type of AI application can be used to support a specific learning activity.

The overall classification of learning and artificial intelligence that we propose Holmes et al. (2019) uses three main descriptors of AI in education. The classification is very relevant to this discussion as it divides the overall uses of AI in education into three main areas: Learning with AI, Learning about AI and Learning for AI.

However, we still miss a recognized pedagogical approach that allows us to describe systematically which AI applications to use to support specific learning activities and to reach specific learning objectives. So, we need a third theoretical building block. For this purpose, we selected elements from the ABC Learning Design method Young & Perovic (2016), which builds on Laurillard’s six learning types Laurillard (2012). The ABC Learning Design approach recommends that educators work together in teams to design a visual storyboard outlining the structures and sequences of learning activities, which are required to meet specified learning outcomes.

The six learning types or activities outlined in the ABC Learning Design method include acquisition, collaboration, discussion, investigation, practice and production. The definitions of each learning type can be found in Young & Perovic (2016) and are written on the actual cards used in the ABC Learning Design method.


AI Pedagogy Planner

The idea of using an instructional design approach in the building of HE programmes has had a huge impact on educators and researchers all over the world, so we developed the AI Pedagogy Planner based on Bower (2008), Laurillard (2012), Fung (2015), and Salmon (2013). Let’s first outline how the ABC Learning Design method works.

The AI Pedagogy Planner is a so-called decision tool, where you start with the inner white wheel. First, you select the type of learning that you want to work with, at a particular stage in your curriculum design, by turning the inner wheel either left or right. The abbreviations ACQ, COL, DIS, INV, PRA, and PRO are the six learning types listed in the left-hand side of Figure 1.

Having selected the learning type, then it is time to select the AI learning types Bidarra et al. (2020). The educator now selects the AI learning type in question, that is, whether it is learning WITH AI, learning ABOUT AI or learning FOR AI, by turning the second light-grey wheel. The abbreviations LWAI, LAAI and LFAI are the three AI learning types listed in the left-hand side of Figure 1.

Next it is time to select the actual AI application, which the educator can use to realize the didactical learning activities required. The educator now turns the outer dark grey wheel and selects the AI application(s), which support(s) the selected learning type in question.

Once the learning type, the AI learning type and the equivalent AI applications have been selected, the educator flips the decision tool and gets concrete pointers to actual exercises, practical learning activities and AI tools that may be used to realize the learning outcomes of the learning module in question. The abbreviations AWE, CB, DBTS, ELE, ITS, LL, LA and AR/VR are abbreviations of different AI applications listed in the left-hand side of Figure 1.

AI Pedagogy Model

Figure 1. AI Pedagogy Planner



This article analysed and discussed pedagogical applications of artificial intelligence in higher education and examined how AI applications may support specific online learning activities in higher education.

We contend that the decision support tool based on a picker wheel approach to AIED could be used in practice by instructional designers to facilitate pedagogically based decisions in the process of building the curriculum in higher education. Educators should lead AI. Educators should not be led by AI.

This research has received funding through ERASMUS+ (KA203-2019-002).



Bidarra, J., Holmes, W., & Simonsen, H. K. (2020). Artificial Intelligence in Teaching (AIT): A road map for future developments. In Empower EADTU. https://empower.eadtu.eu/events/repository (Accessed 5 September 2020)

Bower, M. (2008). Affordance analysis – matching learning tasks with learning technologies, Educational Media International, 45:1, 3-15, DOI: 10.1080/09523980701847115.

Fung, D. (2015). A Connected Curriculum for Higher Education. In: http://discovery.ucl.ac.uk/1558776/1/A-Connected-Curriculum-for-Higher-Education.pdf

Holmes, W., Bialik, M., & Fadel, C. (2019). Artificial Intelligence in Education: Promises and implications for teaching and learning. Boston, MA: Center for Curriculum Redesign.

Laurillard, D. (2012). Teaching as a Design Science: Building Pedagogical Patterns for Learning and Technology. New York and London: Routledge.

Rienties, B., Simonsen, H. K., & Herodotou, C. (2020) Defining the Boundaries Between Artificial Intelligence in Education, Computer-Supported Collaborative Learning, Educational Data Mining, and Learning Analytics: A Need for Coherence. Frontiers in Education. (5)128. doi: 10.3389/feduc.2020.00128.

Salmon, Gilly (2013). E-tivities: The Key to Active Online Learning. New York: Routledge.

Simonsen, H. K. (2020): AI i uddannelsessektoren – hvor langt er vi? In: Forsøg med uddannelsesdigitalisering og uddannelsesformater. Edited by Henrik Køhler Simonsen. Pages 51-58. Samfundslitteratur ISBN: 978-87-7209-369-7.

Young, C., & Perovic, N. (2016). Rapid and Creative Course Design: As Easy as ABC? Procedia - Social and Behavioral Sciences. Volume 228, pages 390 – 395.

Zawacki-Richter, O., Marín, V. I., Bond, M., &Gouverneur, F. (2019). Systematic review of research on artificial intelligence applications in higher education – where are the educators? International Journal of Educational Technology in Higher Education, Vol. 16, No. 1, pp. 1–27.


Artiklen her er en af flere artikler og præsentationer om kunstig intelligens i læring og undervisning på videregående uddannelser, som AIT-gruppen har lavet. AIT-gruppen har de sidste små to år arbejdet på ERASMUS+-projektet Artificial Intelligence in Teaching and Learning. Gruppen består af:

Wayne Holmes, UCL London,

Bart Rienties, Open University,

José Bidarra, Universidade Aberta,

Henrique Mamede, Universidade Aberta,

Vitor Rocio, Universidade Aberta,

Tue Bjerl Nielsen, SmartLearning og

Henrik Køhler Simonsen, SmartLearning.

Projektet er finansieret af ERASMUS+ Project 2019-1-DK01-KA203-060293.

Du kan læse mere om projektet her.