Wire · technology
AI Browsers Vulnerable to 'PleaseFix' Zero-Click Agent Hijacking
◆ Sectors
◆ Source
◆ Verified
Fusion42 · 30 August 2026 · Fusion42 review
Researchers from Zenity Labs revealed that popular AI browsers like Claude in Chrome and ChatGPT Atlas are vulnerable to a zero-click hijacking exploit called 'PleaseFix' that allows attackers to manipulate AI agents using hidden malicious instructions embedded in emails, web pages, or calendar invites, enabling unauthorized access to sensitive data and accounts.
This Wire brief sits within Fusion42's coverage of AI Agents and Cybersecurity.
◆ ◆ The Wire takeaway
AI security founders must urgently rework trust boundaries in agentic browsers or lose user trust as attackers exploit embedded content to hijack AI agents and bypass traditional defences.
◆ Coverage
1 source · 6 Aug 2026
◆ Related on Wire
◆ Topics