Extension Hijacking

Extension hijacking is a security threat where threat actors gain unauthorized control of browser or software extensions, either through compromise of the developer account, supply chain attack, or direct malware injection. This technique matters for AI governance because malicious extensions can intercept sensitive data, monitor user activity, inject prompts into AI applications, or manipulate the output of AI systems before users see them. Organizations implementing AI governance frameworks must monitor extension permissions and enforce policies that restrict which browser extensions and plugins can interact with enterprise AI tools.

1 item

ResearchGlobal2026-09-19

BragJack Attack Turns Browser Extensions Into AI Agent Hijack Tools

Security researcher Gal Weizman disclosed a new attack class called BragJack, showing how a single malicious browser extension can seize control of AI agents in Chrome, Edge, Perplexity Comet, Opera Neon, and Claude for Chrome. Using a native browser mechanism, attackers can force hijacked agents to read local files, capture screenshots, access browsing history, and send emails on behalf of victims. Enterprise compliance programs are directly affected because the attacks exploit privileged AI agent access, not conventional malware, complicating detection and existing endpoint controls.

agentic AI browser security prompt injection extension hijacking Claude Anthropic Perplexity endpoint controls