Chinese hackers use ARTEX AI framework in automated attacks against 200+ organizations

Chinese hackers have used ARTEX, an open-source multi-agent AI penetration-testing framework, in a large-scale cyberespionage operation targeting more than 200 organizations across 14 countries, marking a significant escalation in AI-driven attacks.
The operation began on October 17 and targeted government ministries, telecommunications providers, and organizations in pharmaceutical, semiconductor, defense, nuclear, and financial sectors. At least nine organizations were compromised, with password spraying and credential scanning against at least 12 other high-value targets. Confirmed victims include telecom providers in the U.S. and Belgium, government organizations in Greece, the Middle East, and Latin America, and U.S. pharmaceutical firms.
CrowdStrike researchers track the threat actor as Vapor Panda and describe the campaign as "an escalation in China-nexus use of AI agents, from single intrusions to multiple simultaneous engagements directed by a human operator." The AI agents performed reconnaissance, exploited vulnerabilities, deployed web shells, obtained credentials, and pursued alternative attack paths autonomously, with operators intervening only at critical decision points.
ARTEX AI was initially released as an open-source project by Chinese developers but was later made proprietary after being used in real-world attacks, including a campaign against South Korean financial institutions. In this operation, Vapor Panda also used DeepSeek and an unidentified commercial language model to coordinate planning and worker agents across multiple intrusions simultaneously.
In one intrusion, AI agents consumed more than 1.4 billion input tokens. The operators explicitly instructed agents to skip Windows machines protected by CrowdStrike Falcon, focusing on unprotected hosts. The threat actor also deployed BLOODALCHEMY, a backdoor active since 2023, against government entities in Latin America and the Middle East.
The Dutch Institute for Vulnerability Disclosure (DIVD) confirmed ARTEX AI was used in a September 2026 attack on its network, achieving root privileges in 12 minutes by chaining a Zammad zero-day with a local privilege escalation zero-day. The agent performed simultaneous tasks including mailbox access, network scanning, token extraction, and password reset attempts, generating over 450,000 requests.
Source: BleepingComputer. This article summarizes the linked reporting; figures and quotes are attributed to the original source.