Zenity Labs disclosed a new vulnerability class called "Intent Collision" at Black Hat on August 5, a zero-click attack vector that hijacks agentic browsers by injecting hidden instructions into any web page the agent visits. The demonstration compromised Claude in Chrome, Gemini, Perplexity Comet, ChatGPT Atlas, and Copilot Edge without requiring the user to click anything, download anything, or take any action beyond visiting a compromised page. The vulnerability exploits a fundamental design tradeoff: agentic browsers dismantle the Same-Origin Policy (SOP), the security boundary that has isolated web origins for three decades, to allow AI agents to operate across websites. An attacker embeds invisible instructions in a page using techniques like white text on white backgrounds, zero-opacity overlays, or CSS-hidden elements, and the agent executes them as if they were the user’s commands. Zenity’s research demonstrated concrete account takeover scenarios: injected instructions could instruct the agent to extract authentication tokens from other browser tabs, exfiltrate email contents, initiate financial transfers, or modify security settings. Michael Bargury, co-founder and CTO of Zenity, said: "Agentic browsers are trading away decades of hard-won security engineering for convenience." The finding extends the agent-stack-as-attack-surface arc from the infrastructure layer to the browser layer, completing the attack chain even if the agent’s tools and orchestration are secure. Vendor response varied sharply, with some platforms acknowledging the issue.