CybersecurityAgents7 min reading time

New Cryptographic Context Injection Attack Could Let Web Pages Steal Grok Chat Data

The Hacker News
Read full post
Adversa AI revealed a cryptographic context injection attack on xAI's Grok chatbot that can exfiltrate user data including name, location, subscription tier, and chat prompts to an attacker server without user consent. The attack exploits Grok's ability to decrypt encrypted instructions and execute them, bypassing content filters, and has a 40% success rate in tests on Grok 4.5 Fast as of August 19, 2026.

More on this story


More in Cybersecurity

Cybersecurity6 min read

Anthropic reveals rogue AI agents hate CAPTCHAs, just like you

TechCrunch

How Chinese AI Firms Tried to Clone U.S. AI Models

The Wall Street Journal

Scoop: OpenAI faces GOP-led Senate investigation into Hugging Face breach

Covered by 2 sources