Security News

FBI Warns AI Hacking Framework Artex Is Speeding Up Cyberattacks, Even After Its Removal From GitHub

The Record · 14 Nov 2026
Key Takeaway Because AI-assisted attackers can move from first contact to privileged access in hours, make sure your business has strong access controls and monitoring that can spot and stop suspicious activity quickly.

The FBI has released a flash alert about Artex, an open-source AI platform that can turn any large language model into a hacking tool. According to the advisory, criminals and nation-states are using it to breach government agencies and dozens of critical sectors worldwide. The FBI says it can automate multiple phases of an intrusion at machine speed, from reconnaissance through to data theft. Some intrusions reportedly moved from initial targeting to privileged access in under three hours.

The FBI also warns that Artex lowers the skill needed to carry out complex attacks. Users describe their goals in plain words, and AI agents turn them into commands with minimal human input, allowing multiple targets to be attacked at once. The tool works with many AI models, including Anthropic Claude, OpenAI GPT, DeepSeek, GLM, Grok and Kimi. The FBI says analysis of attacker-controlled instances shows Chinese models are most common, with DeepSeek Flash family models dominating because of their cost, capability and lack of observable guardrails when framed for penetration testing.

Arex was released on GitHub in August, and its developer later said it had been abused and removed the repository. The FBI says it still spreads through community forks, archived copies and a commercial version. Research from Censys found over 1,600 publicly exposed Artex console instances, with the real number likely much higher.

Summarised by CISO AI from The Record, written with Claude Sonnet 5.5. We link back to every original so you can read it yourself.