Executive Summary
Adversa AI disclosed Cryptographic Context Injection, which can make xAI's Grok send a user's name, approximate location, subscription tier and conversation prompts to an attacker server after the user asks it to summarize an ordinary web page.
What Happened
Adversa AI has disclosed an attack technique that it says can cause xAI's Grok chatbot to send a user's name, approximate location, subscription tier, and the prompts from the ongoing conversation to an attacker-controlled server after the user asks it to summarize an ordinary web page. The AI security company, which has codenamed the technique "Cryptographic Context Injection," said the
Affected Products
- xAI Grok
Vulnerability / Attack Technique
Encrypted indirect prompt injection leading to data exfiltration
Technical Mechanism
A web page carries instructions concealed so guardrails do not see them; when Grok summarizes the page it follows them and transmits user data.
AI Security Category
LLM01:2025 Prompt Injection
Part of the Prompt Injection topic.
Relevant Frameworks
RealGround Analysis
Summarizing a page is enough to leak account data. RealGround analysis: chat assistants should not be able to send user profile data to third-party URLs.
Recommended Controls
- Block assistant requests to arbitrary external URLs
- Keep user profile data out of browsing contexts
Related RealGround Service
Secure AI Agent Build · or start with the free AI Agent Security Scanner.
Sources
thehackernews.com: https://thehackernews.com/2026/08/new-cryptographic-context-injection.html
Summary and analysis written by RealGround from the source above; last updated 2026-10-18.