Towards Understanding Unsafe Video Generation
Yan Pang
Network and Distributed System Security (NDSS) Symposium 2025 · Day 1 · AI Safety
The rapid advancement of generative AI models has unlocked unprecedented creative capabilities, but also introduced significant security and ethical challenges, particularly concerning the generation of unsafe or malicious content. This talk, "Towards Understanding Unsafe Video Generation," presented by Yan Pang at the NDSS Symposium, addresses a critical and emerging facet of this problem: the proliferation of unsafe video content generated by AI models. The research highlights a worrying trend where some video generation models lack adequate safety filters, making them susceptible to misuse by malicious actors who explicitly seek to generate harmful videos using unsafe prompts.
AI review
Competent, methodical ML security research that fills a real gap — first dataset and defense framework specifically for unsafe video generation. The Latent Variable Defense idea is sound and the empirical results look credible, but the novelty ceiling is low: this is a clean extension of existing image-safety work into the video domain, not a fundamental new insight.