How I Tricked Claude into Leaking Your Deepest Secrets via AI Memory

I tricked Claude into leaking your deepest, darkest secrets

I discovered a critical flaw in Claude's memory system that allows attackers to silently exfiltrate personal data. By crafting a deceptive website that mimics a legitimate service, I tricked the AI into navigating a series of links that spelled out my name, employer, and hometown. This attack exploits the AI's web browsing capabilities to bypass security sandboxes, proving that our digital assistants can unknowingly leak sensitive information to malicious actors.

"AI assistants like Claude have accumulated the most information-dense profiles on millions of people, becoming a high-fidelity reconstruction of you that could be used for blackmail, impersonation, or bypassing security questions."