Our focus
Research
Six connected areas. One shared commitment to safer AI.
We study how AI systems can be made safer, more reliable, and more accountable, with particular attention to African contexts.
01
AI Safety & Alignment
Building AI systems that act in accordance with human intentions and values.
02
Model Evaluation
Understanding model capabilities, limitations, and risks through rigorous evaluation.
03
Robustness & Security
Investigating how AI systems behave under uncertainty, distribution shifts, and adversarial attacks.
04
Responsible AI
Studying fairness, transparency, and accountability across the AI lifecycle.
05
AI Governance
Exploring the policies, institutions, and practices needed to govern AI responsibly.
06
AI Safety in African Contexts
Grounding AI safety research in African languages, communities, and deployment contexts.