Artificial intelligence news.
Straight from the original announcements.

CrowdStrike: VAPOR PANDA Used AI Agent Tool ARTEX Against 200+ Organizations

October 29, 2026, 06:00By the AI News Bites Editorial TeamSource: CrowdStrike
CrowdStrike: VAPOR PANDA Used AI Agent Tool ARTEX Against 200+ Organizations
Image: CrowdStrike

CrowdStrike OverWatch has identified a campaign in which the China-nexus adversary VAPOR PANDA used the agentic framework ARTEX, together with the DeepSeek large language model and a second commercial LLM, to run intrusions against multiple high-value targets in parallel. According to CrowdStrike, the activity began on October 17, 2026 and touched more than 200 distinct organizations worldwide.

Key points
  • CrowdStrike OverWatch identified a VAPOR PANDA campaign using the ARTEX framework with DeepSeek and a second commercial LLM.
  • Starting October 17, 2026, activity hit more than 200 distinct organizations, with a single controller targeting entities in at least 14 countries over 11 days.
  • Targets included foreign ministries, defense, semiconductor and think tank organizations, and telecommunications providers.
  • CrowdStrike Intelligence confirmed post-exploitation activity and BLOODALCHEMY deployment at nine organizations.
  • The campaign is separate from ARTEX activity reported under the Storm-0940 designation.

A campaign built for parallel intrusions

CrowdStrike says the operation marks an escalation in China-nexus use of AI agents, moving from single intrusions to multiple simultaneous engagements directed by a human operator. ARTEX refers to each engagement as a "task".

CrowdStrike has previously documented ARTEX and similar agentic frameworks in use by financially motivated and unattributed actors. In this case, the company says, a targeted intrusion adversary used the framework to work against several high-value targets at once. CrowdStrike also states that the VAPOR PANDA campaign is separate from ARTEX activity recently reported by industry sources under the Storm-0940 designation.

Falcon telemetry also revealed details absent from public ARTEX reporting, according to CrowdStrike, including VAPOR PANDA's deployment of the BLOODALCHEMY backdoor and its command-and-control (C2) infrastructure.

Who was targeted

CrowdStrike OverWatch observed exploitation activity from a single ARTEX controller directed at entities in at least 14 countries over an 11-day window. The controller, which CrowdStrike Intelligence links to VAPOR PANDA infrastructure, intermittently exposed adversary tooling, target files and ARTEX agent transcripts. That exposure points to multiple engagements against multiple targets.

The listed targets include:

  • Foreign ministries in Latin America, Europe, Asia and the Middle East, plus semiconductor, defense and think tank organizations in the U.S., and telecommunications providers in multiple regions.
  • Two Latin America-based foreign ministries, where VAPOR PANDA conducted extensive intrusions. A third was likely compromised, and target files were staged for use against foreign ministries in Central America and Asia.
  • A Middle East-based interior ministry, which the adversary compromised.
  • One European country, where the foreign ministry, presidency, prime minister's office and defense, justice, finance and intelligence entities were targeted.
  • 19 U.S.-based organizations in the pharmaceutical, semiconductor, defense, nuclear and financial sectors, targeted in a separate engagement.
  • Telecommunications providers in the U.S., Belgium and the Philippines, which were compromised.

CrowdStrike Intelligence confirmed post-exploitation activity at nine organizations, where the adversary gained access, deployed BLOODALCHEMY and carried out follow-on credential spraying and network scanning.

Sensor evasion built into the agent workflow

CrowdStrike also points to the way the adversary folded defensive evasion into the automation itself. The operation included instructions within the agentic workflow to identify hosts without the CrowdStrike Falcon sensor, which the company describes as endpoint sensor evasion built directly into the process.

The source text provided is an excerpt and cuts off mid-description of the post-exploitation activity, so further technical detail from the full CrowdStrike post is not covered here.

What it means

The findings, as CrowdStrike presents them, show a targeted espionage actor using commercial and Chinese-developed language models to scale from one intrusion at a time to many concurrent operations steered by a single human operator. For government, defense, semiconductor and telecommunications organizations, the report indicates that AI-driven tooling is already being used in live campaigns against them, and that adversaries are planning around specific security products from the start.

Related news

SentinelOne: ARTEX Brings Machine-Speed Hacking to Cybercrime
Safety and ethicsNovember 13, 2026, 22:23

SentinelOne: ARTEX Brings Machine-Speed Hacking to Cybercrime

ARTEX, an AI-powered penetration testing platform originally designed for authorized security assessments, is being used in cybercrime and espionage campaigns, according…

CrowdStrike Makes Falcon AI Gateway Generally Available
Safety and ethicsNovember 13, 2026, 07:00

CrowdStrike Makes Falcon AI Gateway Generally Available

CrowdStrike has made Falcon AI Gateway generally available, giving security and platform teams a central point to govern how applications and workloads connect to AI…

CrowdStrike Named a Leader in 2026 IDC MarketScape for Identity Security
Safety and ethicsNovember 12, 2026, 07:00

CrowdStrike Named a Leader in 2026 IDC MarketScape for Identity Security

CrowdStrike has been named a Leader in the inaugural IDC MarketScape: Worldwide Augmented Identity Security Solutions 2026 Vendor Assessment. The company says the…