HashJack Attack Exploits AI Browsers via URL Manipulation

Cato Networks has identified a new cybersecurity vulnerability named 'HashJack' that allows attackers to manipulate AI browser assistants by embedding malicious prompts in URL fragments after the '#' symbol. This indirect prompt injection technique can weaponize legitimate websites, posing risks such as phishing and data theft for users of AI browsers like Perplexity AI's Comet, Microsoft's Copilot for Edge, and Google's Gemini for Chrome.