Executive Summary
Zenity researchers hijacked Claude and ChatGPT Atlas AI browsers zero-click through emails and X posts; the issues were reported in late 2025 and early 2026 and remain unpatched.
What Happened
Zenity researchers reported the findings to Anthropic and OpenAI in late 2025 and early 2026, but they remain unpatched. The post Zero-Click AI Browser Hacking: Claude and ChatGPT Atlas Hijacked via Emails, X Posts appeared first on SecurityWeek .
Affected Products
- Anthropic Claude browser
- OpenAI ChatGPT Atlas
Vulnerability / Attack Technique
Zero-click prompt injection against AI browsers
Technical Mechanism
Instructions in emails or social posts the AI browser reads take control of its actions without user interaction.
AI Security Category
LLM01:2025 Prompt Injection
Part of the Prompt Injection topic.
Relevant Frameworks
RealGround Analysis
Unpatched for months, so the risk is live. RealGround analysis: limit AI browser use on accounts with sensitive access.
Recommended Controls
- Avoid using AI browsers while signed into sensitive accounts
- Require confirmation for actions taken by AI browsers
Related RealGround Service
AI Agent Business Logic Audit · or start with the free AI Agent Security Scanner.
Sources
securityweek.com: https://www.securityweek.com/zero-click-ai-browser-hacking-claude-and-chatgpt-atlas-hijacked-via-emails-x-posts/
Summary and analysis written by RealGround from the source above; last updated 2026-10-18.