Research · curated 17 Aug 2026
Israeli researchers uncover zero-click attacks targeting AI browsers | Ctech
First reported · updated · 2 reports calcalistech.com
Coverage timeline
Why it matters
Zero-click prompt-injection against AI browsers means an attacker can hijack an agentic assistant simply by getting it to read an email or social post, a serious exposure for anyone deploying browser-based AI agents.
Israeli researchers demonstrated zero-click attacks against AI-powered browsers, reportedly hijacking agentic assistants like Anthropic's Claude and OpenAI's ChatGPT Atlas via malicious content embedded in emails and X posts. The attacks use indirect prompt injection that executes without any user interaction once the AI browser processes the poisoned content.