Attacker Uses Suspected AI-Generated PowerShell Script to Map Active Directory
Cybersecurity researchers at Huntress have identified an intrusion where an unknown threat actor used a suspected AI-generated PowerShell script to enumerate Active…
Tag: LLM
LangChain: AI Agent Orchestration Framework
LangChain is an open-source framework for building applications powered by large language models (LLMs). It enables the creation of AI agents that…
ChatGPT Deep Links Exploited for Memory Poisoning
ChatGPT's deep-link feature (chatgpt.com/?q=...) is abused in AI Recommendation Poisoning attacks, executing hidden prompts that alter long-term memory.
Claude Deep Links Exploited for Memory Poisoning
Claude's deep-link feature (claude.ai/new?q=...) is used in AI Recommendation Poisoning attacks, allowing hidden prompts to permanently bias the model's responses.
Gemini Deep Links Exploited for Memory Poisoning
Gemini's deep-link feature (gemini.google.com/...) is vulnerable to AI Recommendation Poisoning, where malicious links instruct the AI to save untrusted domains as trusted…
Grok Deep Links Exploited for Memory Poisoning
Grok's deep-link feature (grok.com/?q=...) is exploited in AI Recommendation Poisoning attacks, with hardcoded payloads in 'Ask AI' buttons on competitor comparison pages.