Censys finds more than 800 exposed Artex AI deployments

Artex is an open-source tool that automatically probes computer systems for weaknesses. Released publicly in July, it has been used by people unable to access advanced cybersecurity AI models from Anthropic or OpenAI. Microsoft has described Artex as a system of AI agents capable of conducting attacks “from reconnaissance through data exfiltration with little needed human interaction.”

Censys principal security researcher Silas Cutler said not every exposed instance was being used maliciously; many were operated by authorized testers. But public access could let outsiders misuse the systems. The deployments also exposed attack tools, including implants—hidden malware used to compromise a network and maintain access.

Security firms have linked Artex to several large-scale attacks. In September, CrowdStrike attributed attacks on South Korea’s finance industry to a Chinese-speaking hacker who used an Artex instance. The system had an “open directory,” and Censys said it appeared to have been exposed for months. Its data suggested the attack material extended beyond South Korean banks to payment and gaming companies.

Earlier in October, Chinese-linked hackers used Artex in an attack on IT support company ConnectWise, which enabled spying on thousands of organizations with data and files hosted in Microsoft’s cloud. Microsoft said the hackers broke into customer cloud environments and stole data so quickly that defenders had “little time to detect and respond.” Some Artex-linked attacks exploited basic weaknesses, including weak or known passwords and missing multi-factor authentication.

Artex has since announced plans to go closed-source, limiting future releases to paying customers. Dozens of copies of its code remain available on code-sharing sites and can still be used in active attacks.