Abstract:

Efforts in life-long learning and competence development on Artificial Intelligence (AI) have been on the rise for several years. These initiatives have mostly been applied to STEM disciplines. Even though there has been significant development in Digital Humanities to incorporate AI methods and tools in higher education, the potential for such competences in arts, humanities and social sciences is far from being realised. Furthermore, there is an increasing awareness that the STEM disciplines need to include competences relating to AI in humanity and society.

The aim of the presented work is to provide the foundations for educational activities in AI in Sweden and beyond by identifying and agreeing on the subject matter content and based on this developing a curriculum for AI. An important aspect of the curriculum is the ethical, legal, societal aspects of AI and to not limit the curriculum to the STEM subjects, instead extending to a holistic, human-centered AI (CHAI) perspective.

An investigation was made of categorisations of AI topics in the field of AI and related applied areas. The most influential international research conferences and the ACM Computer Science Curriculum were reviewed. Further, course curriculum of selected courses spanning across topics were reviewed, selected based on an inventory of courses on AI.

The data was analysed along four dimensions:
(1) subject matters and their classifications;
(2) target groups and advancement levels;
(3) fundamental vs. applied perspectives;
(4) interdisciplinary vs. disciplinary education.

A theoretical framework for analysis was developed in order to place AI technology and related knowledge and skills in a human-centered framework, which provides the fundamental aspects of the human, the society and the environment. Through thematic analyses of AI terminologies and of reviews of existing courses, workshops with teachers and researchers in AI and other topics, a preliminary curriculum was developed and evaluated.

The proposed AI curriculum is organised along the theoretical framework into the following four complementary layers:
Layer 1) includes the core AI theory and technologies, which are already well covered in existing AI education.
Layer 2) includes topics relating to Humans and AI from the individual’s perspective, in particular Human-AI Interaction and Collaboration. This includes also an individual’s participation in the development of new AI systems.
Layer 3a) includes topics relating to societal and humane aspects, and critical perspectives relating to these on the abstract level, which we summarise as Trustworthy AI.
Layer 3b) includes AI technologies that are in today’s society mainly applied to contribute to technical infrastructures (Distributed and Edge AI, and Robotics, Control and Autonomous Systems), and which are also well covered in existing AI and CS education.
Layer 3c) includes AI technologies applied as method for i) research in Medicine, Law, Social Science, etc; ii) addressing societal needs and challenges: improving health, decision making, warfare/defence, etc.; iii) managing societal infrastructures: business, transactions, public transportation, etc.
Layer 4) addressing the history of AI, and the futures of AI.

Existing education programs were analysed using this curriculum, to identify needs for further development of education programs and courses. Initial blueprints of new master programs are being developed based on the curriculum.

Keywords:

Artificial Intelligence, Human-Centred Artificial Intelligence, Artificial Intelligence Education, Curriculum.