Amro Abdalla is a Ph.D. student in Computer Science at Georgetown University, where his research focuses on generative AI, diffusion models, multimodal AI, and the safety and governance of AI models. His work explores methods for aligning generative models with safety objectives, including defending them against malicious fine-tuning while preserving their safe capabilities. Amro is also interested in bridging the gap between technology and policy, aiming to equip policymakers with the tools needed to anticipate, understand, and govern emerging AI risks.
