Not All Attackers Are Malicious: When Safety Degrades Without Harmful Intent
A talk on how safety can silently degrade when models are adapted after deployment, even without malicious intent.
Feb 9, 2026
Who Defines Safety for T2I Diffusion Models?
A lecture on who defines safety for text-to-image models, and why human authority matters in generative AI.
Nov 20, 2025
From Text to Vision: Ensuring Responsibility and Safety in Modern AI
A lecture on attacks, defenses, and responsible safety mechanisms for language and multimodal AI.
May 27, 2025