Chinese Hackers Deploy ARTEX AI in Attacks on Over 200 Organizations

Chinese hackers have employed ARTEX, an open-source, multi-agent AI penetration-testing framework, in a series of attacks targeting over 200 organizations across 14 countries. The operation commenced on October 17 and focused on government ministries, telecommunications providers, and entities within the pharmaceutical, semiconductor, defense, nuclear, and financial sectors.

The threat actor successfully compromised at least nine organizations and executed password spraying attacks and credential scanning activities against at least 12 additional high-value targets.

Researchers at CrowdStrike noted that this campaign represents a significant escalation in the use of AI agents by China-linked groups, evolving from single intrusions to multiple simultaneous engagements directed by a human operator.

Confirmed victims include telecommunications providers in the United States and Belgium, government organizations in Greece, the Middle East, and Latin America, as well as pharmaceutical firms in the U.S.

CrowdStrike tracks the threat actor under the name Vapor Panda. In one incident involving a Greek government entity, the AI agents forged an administrator token and accessed mailboxes by exploiting authentication keys from a 2022 mail-server directory export.

ARTEX AI was initially developed by Chinese developers as an open-source project but was later made proprietary after its use in real-world attacks, including a campaign against South Korean financial institutions.

In this latest cyberespionage operation, Vapor Panda also utilized DeepSeek and an unidentified commercial language model to coordinate planning and worker agents across multiple intrusions simultaneously.

CrowdStrike reported that the attackers issued instructions in Chinese and intervened at critical decision-making points during the operations. The AI agents conducted reconnaissance, exploited vulnerabilities, deployed web shells, obtained credentials, and sought alternative attack paths as necessary.

The operators specifically instructed the agents to avoid Windows machines protected by Falcon, CrowdStrike’s endpoint protection solution, and to concentrate on unprotected hosts.

In one instance, the AI agents processed over 1.4 billion input tokens, according to the researchers.

Additionally, CrowdStrike discovered that the threat actor deployed BLOODALCHEMY, a backdoor active since 2023, against government entities in Latin America and the Middle East.

Further ARTEX AI Incidents

The Dutch Institute for Vulnerability Disclosure (DIVD) has provided more details regarding a cyberattack it experienced in September 2026, which it attributed to an AI agent.

According to its latest update, the tool used in the attack on its network was ARTEX AI, which operated autonomously. DIVD explained in a LinkedIn post that the attack escalated from initial access to root privileges within 12 minutes, chaining a Zammad zero-day vulnerability with a second zero-day for local privilege escalation.

The agent executed numerous tasks concurrently, including reading a compromised mailbox, scanning the internal network, extracting login tokens, and attempting password resets.

DIVD recorded over 450,000 requests and 2,500 scripts associated with the attack, describing the activity as “super loud and messy.”

Source: BleepingComputer