Skip to content
#

ai-threat-detection

Star

Here are 3 public repositories matching this topic...

Language: All
Filter by language
All 3 Shell 1 TypeScript 1

NotSooShariff / adversarial-vision

Star 1

Adversarial Vision is a research-backed interactive playground exploring how pixels can become prompt injections. It demonstrates how hidden text, subtle contrast shifts, and adversarial visual cues can manipulate multimodal AI models like ChatGPT, Perplexity, or Gemini when they “see” images.

research computer-vision nextjs cybersecurity image-manipulation red-team ai-safety adversarial-attacks ai-security prompt-injection llm-security multimodal-ai ai-threat-detection
  • Updated Oct 25, 2025
  • TypeScript

Improve this page

Add a description, image, and links to the ai-threat-detection topic page so that developers can more easily learn about it.

Curate this topic

Add this topic to your repo

To associate your repository with the ai-threat-detection topic, visit your repo's landing page and select "manage topics."

Learn more