Skip to content
Threat Feed

Tag

Vlm

1 brief RSS
high advisory

Manipulation of Vision-Language Models via Imperceptible Image Perturbations

Cisco researchers discovered that attackers can manipulate vision-language models (VLMs) by using pixel-level perturbations in images to embed malicious instructions, which are unreadable by humans but interpreted by AI, leading to potential data exfiltration or other unauthorized actions.

GPT-4o +5 ai vlm perturbation defense-evasion ai-security
2r 1t