Forty-four AI agents split up the work and went after at least 16 Japanese organizations
- At a glance: 44 AI agents split up an attack on at least 16 Japanese organizations
- Date: 2026-10-23
- Reported by: Trend Micro (JA), Trend Micro research (EN), eSecurity Planet
- Status: Source: Trend Micro's analysis of files left on attacker servers, without victim-side logs. "AI-driven" is the analyst's assessment.
- Last checked: 2026-10-29
On October 23, 2026, Trend Micro published an analysis of attackers abusing CyberStrikeAI, an open-source AI-agent penetration testing tool, against Japanese organizations. The attackers' exposed management console was in Simplified Chinese.
Separate agents handled reconnaissance, vulnerability research, exploitation, command and control, persistence, data exfiltration and report generation. One record reads "agent 8/44 in charge of credential discovery", pointing to a setup of 44 agents.
Within what could be identified, at least 19 organizations were targeted, at least 16 of them in Japan, plus targets in Taiwan and Hong Kong. Trend Micro calls this a lower bound.
Trend Micro stresses that attempts are not breaches. For most targets the attackers' own records show no session, no web shell and no exfiltration. A full intrusion chain was clearly established at only one company, through an old weakness: an unauthenticated public admin page. The attackers are unidentified, and it is not confirmed whether they are the same actors behind the earlier ARTEX-related attacks.
Source: Trend Micro's analysis of files left on attacker servers, without victim-side logs. "AI-driven" is the analyst's assessment.
Sources
Find this archive useful? Supply the resistance ♥
See today's AI threat brief →