AI ClickFix Hijacking Computer-Use Agents Using ClickFix

# AI ClickFix Hijacking Computer-Use Agents Using ClickFix

https://gcdradar.vercel.app/incidents/ai-clickfix-hijacking-computer-use-agents-using-clickfix · AIR-2026-0349 · closed 2026-04-14

- Lane: Research
- Severity: 1 (Research / proof of concept)
- Fix status: Fix status unknown
- First reported: 2026-03-31 (happened 2025-05-24)
- AI role: Attack on AI
- Products: Claude Computer-Use Agent, ChatGPT

## Summary

Researchers demonstrated that AI computer-use agents can be manipulated via prompt injection and social engineering to execute arbitrary code on a victim machine. The red-team exercise highlights risks to automated agents interacting with untrusted web content.

## What to do

No action recorded; read the sources below and the vendor's advisory.

## Latest updates

- 2026-03-31: AI ClickFix: Hijacking Computer-Use Agents Using ClickFix (https://embracethered.com/blog/posts/2025/ai-clickfix-ttp-claude)
- 2026-03-31: AI ClickFix: Hijacking Computer-Use Agents Using ClickFix (https://atlas.mitre.org/studies/AML.CS0055)

## Sources

- 2026-03-31 MITRE ATLAS: AI ClickFix: Hijacking Computer-Use Agents Using ClickFix https://atlas.mitre.org/studies/AML.CS0055
- 2026-03-31 Embrace The Red: AI ClickFix: Hijacking Computer-Use Agents Using ClickFix https://embracethered.com/blog/posts/2025/ai-clickfix-ttp-claude

Verify against the vendor's advisory before changing production. Data CC BY-SA 4.0, https://gcdradar.vercel.app.