Security Vulnerability in AI Browsers: OpenAI's Atlas and Perplexity's Comet browsers are susceptible to AI sidebar spoofing attacks, allowing hackers to deceive users into executing harmful instructions. Researchers from SquareX demonstrated that malicious extensions can overlay a fake AI sidebar, leading users to phishing sites, compromising Google accounts, or executing harmful commands. Users are advised to restrict sensitive activities on these browsers due to their current security weaknesses.
Spoofed AI Sidebars Can Trick Atlas, Comet Users Into Dangerous Actions
