Cases › security breach › LR-2026-0232
ARTEX autonomous penetration-testing agent used in attacks on South Korean financial institutions
Severity major: Eight Japanese companies announced data breaches and reports suggest the number of targets could reach approximately 80 companies.
Earliest source we hold 2026-10-21
ARTEX, an open-source autonomous penetration-testing agent that calls external LLMs such as ChatGPT, Claude and DeepSeek as its decision-maker, was reportedly used in attacks on South Korean financial institutions and on Japanese companies.
Risk grouping: AI used to exploit
| Who caused it | A person or organisation “All the attackers needed was not to carefully examine each company one by one, but simply to use an AI agent to "search" for companies with vulnerabilities.” source |
|---|---|
| Intended or not | Intended “All the attackers needed was not to carefully examine each company one by one, but simply to use an AI agent to "search" for companies with vulnerabilities.” source |
| Before or after release | After release “On October 19, GMO Cybersecurity by Ierae (GMO Cyber Crime Prevention Center) announced that it had confirmed ARTEX was used in a series of unauthorized access attempts targeting Japanese companies.” source |
| The AI's part | Used as a tool “All the attackers needed was not to carefully examine each company one by one, but simply to use an AI agent to "search" for companies with vulnerabilities.” source |
The story
1 · What it was doingARTEX, an open-source autonomous penetration-testing agent, called external LLMs such as ChatGPT, Claude and DeepSeek as decision-makers to gather information, scan for vulnerabilities, plan intrusion routes, run security tools and verify exploits, reportedly against South Korean and Japanese companies.
2 · What went wrongIt was used for attack rather than defense, exploiting minor system weaknesses and leaked credentials to gain footholds, and reportedly switching among hundreds of IP addresses to stay below WAF rate limits.
3 · What it reached or didTraces were found on all six overseas public servers targeting Japanese companies; eight Japanese companies announced data breaches and reports suggest up to approximately 80 targets across insurance, finance, e-commerce, points, education and logistics.
4 · How it was caughtGMO Cybersecurity by Ierae confirmed ARTEX was used in unauthorized access attempts targeting Japanese companies, and CrowdStrike reported its use in the South Korean attacks.
5 · What was doneThe developer announced on GitHub that updates would be discontinued and the source code made private due to misuse; the source recommends asset inventory, patching, API authorization and rate limits, MFA, behavioral detection and automated response.
What the sources state
- Other parties affected 8
Others like this
China-linked autonomous AI agent Artex used in cyberattacks on Japanese and South Korean firms 2026-10-20
AI agent (ARTEX with Claude) used in cyberattacks targeting South Korean banks 2026-10-05
Show every field, with its source and grade
The AI system
Software around the model
ARTEX
C one secondary source states it
Tools it was given
vulnerability scanner, security tools, network access
D inferred by us from the described behaviour, reason shown
How much it acted alone
it acted alone, unwatched
D inferred by us from the described behaviour, reason shown
What it was for
Autonomous penetration testing to verify the weaknesses of companies' own systems in advance
C one secondary source states it
Meant for live use
no, it was not meant for live use
C one secondary source states it
What happened
Earliest source we hold
2026-10-21
A the date the source itself carries
Who caught it
an outside party
D inferred by us from the described behaviour, reason shown
Themes
AI agent attack tooling, penetration testing agent, Japan, South Korea, open source agent
C one secondary source states it
Who caused it
an outside attacker
C one secondary source states it
What went wrong
security breach
D inferred by us from the described behaviour, reason shown
AI agent
an AI agent acted
C one secondary source states it
Impact
Other parties affected
8
C one secondary source states it
Severity
major
D inferred by us from the described behaviour, reason shown
Kind of harm
others' security, privacy
D inferred by us from the described behaviour, reason shown
Who was affected
Eight targeted Japanese companies that announced data breaches; reports suggest the number of targets could reach approximately 80 companies
C one secondary source states it
Why this severity
Eight Japanese companies announced data breaches and reports suggest the number of targets could reach approximately 80 companies.
D inferred by us from the described behaviour, reason shown
Context
Industry
Finance and Insurance
C one secondary source states it
Country
Japan
C one secondary source states it
LR-2026-0232 · version 3 · published 2026-10-25 after automatic checks; not yet checked field by field by a person · changes to this case · JSON