NVIDIA Research Exposes Critical VLM Security Flaws in AI Vision Systems

NVIDIA researchers demonstrate how adversarial image attacks can manipulate vision language models, turning traffic light recognition from 'stop' to 'go' with imperceptible changes.