AI · 147 giorni fa

Cisco research demonstrates VLMs can be manipulated by imperceptible pixel perturbations embedding instructions in images (e.g., banners/previews), creating a covert channel that humans and simple filters miss.

Cisco research demonstrates VLMs can be manipulated by imperceptible pixel perturbations embedding instructions in images (e.g., banners/previews), creating a covert channel that humans and simple filters miss.

Cronologia

• 7 mag Segnalato da SecurityWeek

Fonti

1 fonte che coprono questa storia

• SecurityWeek 7 mag

  Attackers Could Exploit AI Vision Models Using Imperceptible Image Changes

  Cisco’s AI security researchers have analyzed ways to target vision-language models (VLMs) using pixel-level perturbation.

Part of the PlainSec briefing for 2026-05-07

Editions

• Briefing for 2026-05-07