Empowering ethical innovation: designing a responsible computing workshop for emerging economies

Abstract

This paper reports on a multi-method collaborative autoethnographic (CAE) research project aimed at improving the understanding of the status quo of artificial intelligence (AI) and AI ethics in a university situated in an emerging African country. The rapid and disruptive technological developments of the Fourth Industrial Revolution (4IR), which include recent AI developments, have resulted in an ethics and skills gap among African researchers and tertiary students. In addressing this problem, the University of Pretoria embarked on a collaborative research project that includes reaching out to other African universities to plan and facilitate training workshops for academics and postgraduate students to enhance the effective use of AI. The multi-phased study primarily collected qualitative data through an initial scoping literature review to explore extant research on AI ethics globally and, specifically, in African Universities. Based on identified gaps, further confirmed by the scoping review results, a training workshop series was designed to offer practical and theoretical support towards fostering the responsible use of AI in research. Furthermore, the workshop was designed to equip and encourage researchers and students to research and test AI-powered tools and platforms in teaching, learning, and research. The workshop's design was informed by the scaffolded nature of the Zone of Proximal Development frame (ZPD). Pre– and post-workshop observations and surveys were used to add to the knowledge gained from the scoping review. The post-workshop survey was used to gauge the success and usefulness of the workshops, as reflected in the expectations of workshop delegates from the pre-workshop surveys. Data from the scoping review, the reflexive data from the CAE process, and workshop surveys were triangulated and thematically analyzed. Findings are that the workshop was informative and timely, and more interventions of this nature are needed to further entrench the real-life experience of ethical AI use in an academic environment. This paper reports on a multi-method collaborative autoethnographic (CAE) research project aimed at improving the understanding of the status quo of artificial intelligence (AI) and AI ethics in a university situated in an emerging African country. The rapid and disruptive technological developments of the Fourth Industrial Revolution (4IR), which include recent AI developments, have resulted in an ethics and skills gap among African researchers and tertiary students. In addressing this problem, the University of Pretoria embarked on a collaborative research project that includes reaching out to other African universities to plan and facilitate training workshops for academics and postgraduate students to enhance the effective use of AI. The multi-phased study mainly collected qualitative data through an initial scoping literature review to explore extant reported research on AI ethics globally and specifically in African Universities. Based on identified gaps, further confirmed by the scoping review results, a training workshop series was designed to offer practical and theoretical support towards fostering the responsible use of AI in research. Furthermore, the workshop was designed to equip and encourage researchers and students to research and test AI-powered tools and platforms in teaching, learning, and research. The workshop's design was informed by the scaffolded nature of the Zone of Proximal Development frame (ZPD). Pre– and post-workshop observations and surveys were used to add to the knowledge gained from the scoping review. The post-workshop survey was used to gauge the success and usefulness of the workshops, as mirrored by the expectations of workshop delegates from pre-workshop surveys. Data from the scoping review, the reflexive data from the CAE process, and workshop surveys were triangulated and thematically analysed. Findings are that the workshop was informative and timely, and more interventions of this nature are needed to further entrench the real-life experience of ethical AI use in an academic environment.