Censys Finds Over 800 Publicly Exposed Artex AI Tool Instances

More than 800 Artex deployments were publicly reachable

Censys has identified more than 800 distinct deployments of Artex, a Chinese-made open-source AI penetration-testing tool, that were accessible from the web without a password. The findings raise concern because the software can automate the probing of computer systems for weaknesses and has been linked to several cyberattacks since its public release in July.

Artex is designed to run on users' own servers. Censys principal security researcher Silas Cutler said not every exposed instance is operated maliciously; many may support authorised penetration testing. However, the software does not need to be internet-accessible, leaving the purpose of hundreds of public deployments unclear.

Exposure can reveal attack infrastructure

Microsoft has described Artex as a system of AI agents able to conduct cyberattacks from reconnaissance through data exfiltration with little human interaction. Censys said publicly accessible instances can expose attack tooling, including implants: hidden malicious software used to compromise a victim environment and maintain persistent access.

The discovery illustrates a basic but consequential operational failure. A powerful tool intended for security testing can be left online without authentication, making it available to anyone who finds it. That exposure can place both the operator and potential victims at greater risk of misuse.

Artex has appeared in several reported attacks

Security companies have connected Artex to broad-scale attacks, including activity against South Korea's finance industry in September. CrowdStrike attributed those attacks to a Chinese-speaking hacker who used an Artex instance and had left an open directory hosting the system. Censys said the instance had been publicly exposed for months and appeared to contain attack data extending beyond South Korean banks to payment and gaming companies.

Artex was also used in an October attack on IT support company ConnectWise. Microsoft said Chinese-linked hackers used the tool to enter customer cloud environments and steal data so quickly that defenders had little time to detect and respond. The affected organisations included customers whose data and files were hosted in Microsoft's cloud.

Basic controls remain central to defence

Many attacks associated with Artex exploited simple weaknesses, including guessed weak or known passwords and the absence of multi-factor authentication. The case shows that advanced AI-enabled tooling does not remove the importance of reducing routine exposure and correcting configuration errors.

Artex has said it will move to a closed-source model, 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. For businesses, the practical implication is to audit internet-facing systems, remove unnecessary public access, enforce multi-factor authentication and investigate exposed security tools as a priority.