Assoc Prof Jonathan Shock

Jonathan trained as a theoretical physicist, with postdoctoral positions in Beijing, Santiago de Compostela and Munich before joining the University of Cape Town in 2013, he now works at the intersection of machine learning, neuroscience and the frontiers of Artificial Intelligence. He is Associate Professor in the Department of Mathematics and Applied Mathematics and Interim Director of the UCT AI Initiative, chairs the International Alliance of Research Universities working group on AI, and is Principal Investigator of the African Hub on AI Safety, Peace and Security and of the African Compute Initiative. Much of his current work asks how we can understand the internal workings of AI systems, and what the particular implications are in an African context. 

Some of his publications and preprints include:

• Segun, S. et al. 2025. Toward an African Agenda for AI Safety. arXiv:2508.13179.
• Ireri, G. et al. 2026. Assessing the Case for Africa-Centric AI Safety Evaluations. arXiv:2602.13757.
• Ngcobo, S., Madela-Mntla, E., Shock, J., Louw, M., Mbonambi, L., Serite, T. & Rossouw, T. 2025. Artificial Intelligence for HIV Care: A Global Systematic Review of Current Studies and Emerging Trends. Journal of the International AIDS Society. 28(10).
• Solms, M., Grimbly, S.J., Bassett, B., Boonstra, E., Hodson, R., Kuske, N., Mahadew, K., Rosman, B., van Hoof, C. & Shock, J. 2026. Inferring Affective Consciousness in an Artificial Agent: A Case Study. Journal of Consciousness Studies. 33(7-8): 14-34.
• Formanek, C., Tilbury, C. & Shock, J.P. 2024. Opportunities of Reinforcement Learning in South Africa's Just Transition. Southern African Conference for Artificial Intelligence Research (SACAIR).