New Attack Technique Tricks AI Browsers Using a Simple ‘#’
Security researchers at Cato CTRL have discovered a new indirect prompt injection technique called HashJack, which weaponises legitimate websites to manipulate AI browser assistants. The attack…