CrowdStrike Tracks at Least 6 AI Agent Adversary Groups

CrowdStrike is tracking at least six agentic adversary groups that use AI agent tools to carry out intrusion activity, Adam Meyers said in an interview in Seoul, according to ZDNet Korea. Meyers also said the company announced Falcon Guardian, which it describes as the next step of AI detection and response, or AIDR.

Meyers said CrowdStrike detected twice as many new groups of this kind in the last three months as in the previous three months. On its platform, the company says AI-originated detections have reached 2.5 times human-originated detections over the past six months.

CrowdStrike also said it has established an in-house frontier AI research organization, the Cyber Superintelligence Lab, and has begun releasing security-focused models: Red Tempest for offense and Blue Solano for defense. At its annual Fal.Con conference in September, the company announced the Red Tempest and Blue Solano models.

On the 8th, Meyers held a video briefing with Korean media about an analysis report. From late September to early October, a targeted campaign hit Korean financial institutions; CrowdStrike reconstructed it from artifacts including Claude Code session records, ARTEX configuration files and Claude memory files. CrowdStrike assessed the campaign as likely carried out by a Chinese-speaking cybercrime group using ARTEX and DeepSeek, at moderate confidence.

CrowdStrike was co-founded in 2011 by George Kurtz, Dmitri Alperovitch and Gregg Marston and is headquartered in Austin, Texas. The company listed on Nasdaq in 2019, entered the Korean market in 2020 and names AhnLab and Genians as Korean competitors. For the fiscal year ended January 31, 2026, it reported revenue of 4.81 billion dollars and about 10,700 employees as of 2026.