Skip to the content
AI and Cybersecurity Learning and Practice Center Test automation Open the Learning Center

WeaknessASC-2026-0025Other weakness

Prompt injection through images, audio and video

MediumWorkarounds only

Multimodal models read text in images, spoken words and frames. Instructions hidden in a screenshot, a photo or an audio clip work like text injection and are harder for people to spot. Trail of Bits showed images whose hidden text appears only after the model's own downscaling step.

How to detect it

  • Images or audio from untrusted sources passed to an agent with tools
  • Text in images that addresses the AI

Workarounds

  • Treat media from outside as untrusted input, just like text
  • Do not give agents that process outside media access to tools that act or send data
  • Show the user the image as the model sees it after any resizing, and require confirmation before text in an image can start a sensitive action

There is no reliable filter for instructions embedded in images or audio.

How to investigate

  1. Review the media the agent processed before the unexpected action
  2. Extract text from the media (OCR, transcription) to find the instructions

Sources

Open it in the Learning Center's vulnerability database