China-made AI tool ARTEX used to attack Japanese companies, with about 20 targets

GMO Cybersecurity by Ierae said on October 19 that attacks targeting Japanese companies were confirmed to have used the China-developed AI autonomous penetration testing system ARTEX, with about 20 companies listed as attack targets. Within its observable scope, the company found that the attack infrastructure almost exclusively targeted Japanese companies, 9 of which were websites in the same industry; it also newly discovered 6 AI attack infrastructures mainly targeting Japanese companies, involving industries including digital gifts, retail distribution, education, printing, finance, insurance, e-commerce, media and logistics. Among the confirmed Japanese companies, 8 disclosed information leaks in September–October 2026; the company believes the same attack group caused multiple victim incidents and has shared the information with law enforcement agencies. ARTEX was open-sourced in July 2026 and was taken down by its developer in October after reports of abuse; CrowdStrike also reported the use of ARTEX in attacks on South Korean financial institutions. The company noted that about 20 is the number of targets of the initial attack infrastructure, not meaning all companies were successfully breached, and that not all of the 6 additionally discovered infrastructures were confirmed to use ARTEX.