Chinese developer makes ARTEX AI agent closed-source after Korean bank hack

Chinese developer ARTEX has transformed their AI agent into a closed-source project following cybersecurity firms' identification of its use in recent cyberattacks against South Korean banks. The decision was announced by Autumn-27, the developer behind the project, on Thursday, stating that no further updates or maintenance support will be provided.

The original purpose of ARTEX was to assist enterprises and organizations in conducting security risk testing and enhancing security capabilities. However, the developer explicitly mentioned their opposition to any illegal use of the software, claiming no responsibility for conduct that violates laws and regulations. ARTEX's GitHub page has been taken down, as confirmed by Reuters.

US cybersecurity firm Crowdstrike attributed the recent cyberattacks on South Korean banks, aimed at stealing customers' personal data, to a China-based 26-year-old individual who utilized the ARTEX AI agent and Anthropic's Claude Code. ARTEX, an open-source AI agent designed to automate penetration testing, connects to external language models such as ChatGPT, Claude, and DeepSeek.

At least nine South Korean banks have been targeted by cyberattacks since late September, leading to an ongoing police investigation and President Lee Jae Myung's call for robust response measures. The Chinese foreign ministry spokesperson, Mao Ning, denied knowledge of the case during a press briefing, reaffirming China's consistent stance against hacking activities.