The AI Initiative, UCT School of IT and Cybersecurity Capacity Centre for South Africa (C3SA) we are excited to host two leading researchers from CISPA Helmholtz Center for Information Security on Friday, 13 February 2026 for thought-provoking public talks that challenge common assumptions in cybersecurity and AI.

Talk 1: Security Myths and the Spread of Fear, Uncertainty and Doubt

Time: 13:00 - 13:45 SAST

Users of digital systems are bombarded with security and privacy advice. Not all this advice, given by security experts and authorities, is helpful, usable, or even based on evidence. In this talk, Dr Jonas Hielscher will look at famous security myths, explore the underlying causes for their spread and discuss why the spread of myths lead to fear, uncertainty and doubt among users and subsequently to misconceptions and dangerous behaviour. He will also discuss a recent study with 12,000 participants (1,000 from South Africa) exploring security misconceptions around the world.

Talk 2: AI for Social Good: When Good Intentions Fail

Time: 13:45 - 14:30 SAST

In this talk, Dr Krikamol Muandet explores the gap between the aspirational promise of AI for social good and the realities of deploying AI systems in complex, high-stakes settings. It examines how well-intentioned interventions can produce unintended harms when confronted with epistemic uncertainty, data limitations, misaligned incentives, and strategic behavior. By unpacking these recurring failure modes, the talk argues for a more epistemically humble, mechanism-aware, and responsibility-driven approach to the design and governance of AI systems intended for social benefit.

Event Details: 
Friday, 13 February 2026 
13:00 - 14:30 SAST
CS 203, Computer Sciences Building, University of Cape Town Upper Campus

Both talks are open to all. Whether you're in tech, policy, academia, or simply curious about how we can build more trustworthy digital systems, these insights are essential.

About our Speakers
Dr Jonas Hielscher

Jonas Hielscher is a Tenure-track Faculty with a focus on human-centered and organization security. He leads the CHESS (Cybersecurity, Human Factors and Enterprise Security Studies) Lab@CISPA. In his works he combines methods from cybersecurity, human computer interaction, psychology, and sociology to empirically assess cybersecurity in the practice of organizations and individuals. He has published several papers on security leadership, security misconceptions and analysis of large data sets about users and security policies. Currently he is looking into the security of critical infrastructure organizations, usable security aspects of industrial control systems and the security economics.

Five Selected Publications 

  1. Hielscher, J., Menges, U., Parkin, S., Kluge, A., & Sasse, M. A. (2023). {“Employees} Who Don’t Accept the Time Security Takes Are Not Aware Enough”: The CISO View of Human-Centred Security. In 32nd USENIX Security Symposium (USENIX Security 23) (pp. 2311-2328).
  2. Hielscher, J., & Parkin, S. (2024). “What Keeps People Secure Is That They Met the Security Team”: Deconstructing Drivers and Goals of Organizational Security Awareness. In 33rd USENIX Security Symposium (USENIX Security 24) (pp. 3295-3312).
  3. Hielscher, J., & Golla, M. (2025, November). Quantifying Security Training in Organizations Through the Analysis of US SEC 10-K Filings. In Proceedings of the 2025 ACM SIGSAC Conference on Computer and Communications Security (pp. 51-65).
  4. Hielscher, J., Schöps, M., Opdenbusch, J., Reichmann, F., Gutfleisch, M., Marky, K., & Parkin, S. (2024, December). Selling Satisfaction: A Qualitative Analysis of Cybersecurity Awareness Vendors' Promises. In Proceedings of the 2024 on ACM SIGSAC Conference on Computer and Communications Security (pp. 2666-2680).
  5. Herbert, F., Becker, S., Schaewitz, L., Hielscher, J., Kowalewski, M., Sasse, A., & Dürmuth, M. (2023, April). A World Full of Privacy and Security (Mis) Conceptions? Findings of a Representative Survey in 12 Countries. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (pp. 1-23).
Dr Krikamol Muandet

Krikamol Muandet is a machine learning scientist. He is leading the Rational Intelligence (RI) Lab at CISPA. His research aims at developing artificial intelligence systems that go beyond pattern recognition by enabling them to distinguish genuine causal relationships from spurious ones (causality), recognize the limits of their own knowledge (uncertainty), earn and maintain human trust (human-centric design), and learn to collaborate effectively with one another (collective intelligence). To achieve this vision, his research requires a seamless integration of data, people, and AI. The approach is interdisciplinary, combining insights from statistics, computer science, machine learning, and economics.

Five Selected Publications 

  1. A. Singh, S. L. Chau, S. Bouabid, and K. Muandet, “Domain generalisation via imprecise learning,” in UCML, 2024.
  2. S. L. Chau, K. Muandet, and D. Sejdinovic, “Explaining the uncertain: Stochastic Shapley values for Gaussian process models,” in NeurIPS, Curran Associates Inc., 2023.
  3. J. Park, S. Buchholz, B. Schölkopf, and K. Muandet, “A measure-theoretic axiomatisation of causality,” in NeurIPS, Curran Associates Inc., 2023.
  4. J. Park and K. Muandet, “Towards empirical process theory for vector-valued functions: Metric entropy of smooth function classes,” in International Conference on Algorithmic Learning Theory, vol. 201, PMLR, 2023.
  5. R. Zhang, M. Imaizumi, B. Schölkopf, and K. Muandet, “Instrumental variable regression via kernel maximum moment loss,” Journal of Causal Inference, vol. 11, no. 1, 2023.