The integration of artificial intelligence features into web browsers has created a new security challenge, cyber risk specialist Unit 42 by Palo Alto Networks wrote in a Monday (March 2) blog post.
This new browser architecture introduces an AI assistant that could allow attackers to issue commands to the browser core, according to the post. In addition, the integration of an AI side panel brings about vulnerabilities that could be exploited by less-privileged website or browser panels, per the post.
In order to perform tasks, AI browsers must be granted privileged access to the browsing environment so they can see what the user sees on the screen and get information from the website, according to the post.
“However, this expanded capability and privileged access introduce a new and widened attack surface,” the post said. “This creates security implications that are not present in traditional browsers.”
We’d love to be your preferred source for news.
Please add us to your preferred sources list so our news, data and interviews show up in your feed. Thanks!
Unit 42 said this while outlining a security vulnerability in Google’s Gemini feature in its Chrome browser that was discovered and fixed. This security vulnerability, CVE-2026-0628, could have allowed malicious extensions with basic permissions to take control of the Gemini Live in Chrome browser panel, escalate privileges, and perform actions such as access the victim’s camera and microphone, take screenshots of websites and access local files and directories, according to the post.
Unit 42 uncovered the security vulnerability, disclosed it to Google and helped the company remediate it, per the post. Google released a fix in early January before the information about the security vulnerability was published.
Advertisement: Scroll to Continue
“While AI browsers or AI features implemented into existing browsers can improve the user experience, it’s important to continue monitoring for potential security flaws,” the post said.
It was reported in February that Google’s Chrome web store is being plagued by a wave of malicious browser extensions that claim to offer the abilities of an AI assistant but steal users’ personal information in secret.
PYMNTS reported in October 2025 that the autonomy that makes AI browsers powerful also makes them risky.
Google said in a December 2025 blog post that it designed new tools it will use to improve the safety of agentic browsing with Chrome.
Add as Preferred Source
