AI Risk Wire

Cases › security breach › LR-2026-0893

Threat actors weaponize ARTEX autonomous penetration-testing agents for intrusions and data exfiltration

Also listed as: AI cyberattacks

Who did it
Outside attacker
Who was hit
South Korean financial institutions and about 67,000 individuals whose personal data was exfiltrated; more than 2,250 downstream tenants of ConnectWise Cloud Backup
AI agent used
ARTEX
AI company and model
Autumn-27
Reported
14 Nov 2026
Severity
Moderate
Sector
Finance and Insurance
  1. ARTEX, an LLM-driven multi-agent platform, turns natural-language objectives into machine commands run by specialized agents with little human input after the goal is set; threat actors used it to run intrusions.
  2. Threat actors used ARTEX to automate intrusions from reconnaissance through data exfiltration, reaching privileged access in some cases in under three hours.
  3. One campaign reached South Korean financial institutions and exfiltrated personal data on about 67,000 individuals; another was a supply chain compromise of ConnectWise Cloud Backup exposing more than 2,250 downstream tenants.

Source: Cyber Security News · 4 more

What was done

The developer pulled the ARTEX repository on October 8 after abuse reports, but forks, archived copies and ARTEX PRO keep it available; the FBI urges organizations to inventory peripheral internet-facing systems and restrict administration endpoints.

Withdrawn or disabled (Developer or vendor) · Other stated action (Regulator or court) · Withdrawn or disabled (Developer or vendor)

Fix reported: “The developer pulled the ARTEX repository on October 8 after abuse reports; the FBI urges organizations to inventory peripheral internet-facing systems, restrict Kudu/SCM and cloud administration endpoints, and apply least privilege to multi-tenant application consent grants.” · Network egress control

Risks AI used to run a cyber attack (97 incidents) · AI system attacked or breached (54 incidents)

No source states: money lost or records, tools it had, company using the AI.

Similar incidents

All cases of this kind ›

Evidence

Sources (5)

  1. FBI Warns Hackers Are Using ARTEX AI Agents to Breach Networks in Under 3 Hours · Cyber Security News · 2026-11-14
  2. FBI warns that the Artex AI hacking tool can reach privileged access in under three hours · MLQ.ai · 2026-11-14
  3. FBI Warns of ARTEX AI Attack Framework — CASRAI · CASRAI · 2026-11-14
  4. FBI FLASH on ARTEX: Agentic AI Hacking Frameworks Move From Pen Testing Tool to Mainstream Threat · OODA Loop · 2026-11-14
  5. FBI FLASH on ARTEX: Agentic AI Hacking Frameworks Move From Pen Testing Tool to Mainstream Threat — OODAloop - EUROPE SAYS · EUROPE SAYS · 2026-11-15
Show every field, with its source and grade

The AI system

Developer
Autumn-27
D inferred by us from the described behaviour, reason shown
Software around the model
ARTEX
C one secondary source states it
How much it acted alone
it acted alone, unwatched
D inferred by us from the described behaviour, reason shown
Stage of use
in live use
D inferred by us from the described behaviour, reason shown
How it was connected
agent with tools
D inferred by us from the described behaviour, reason shown
What it was for
autonomous penetration testing
D inferred by us from the described behaviour, reason shown

What happened

Earliest source we hold
2026-11-14
D inferred by us from the described behaviour, reason shown
Themes
penetration testing framework, supply chain compromise, data exfiltration, open source abuse
D inferred by us from the described behaviour, reason shown
Who was responsible
an outside party
D inferred by us from the described behaviour, reason shown
Who caused it
an outside attacker
D inferred by us from the described behaviour, reason shown
What was done
The developer pulled the ARTEX repository on October 8 after abuse reports; the FBI urges organizations to inventory peripheral internet-facing systems, restrict Kudu/SCM and cloud administration endpoints, and apply least privilege to multi-tenant application consent grants.
D inferred by us from the described behaviour, reason shown
What went wrong
security breach
D inferred by us from the described behaviour, reason shown
AI agent
an AI agent acted
D inferred by us from the described behaviour, reason shown

Impact

Severity
moderate
D inferred by us from the described behaviour, reason shown
Kind of harm
privacy, others' security
D inferred by us from the described behaviour, reason shown
Who was affected
South Korean financial institutions and about 67,000 individuals whose personal data was exfiltrated; more than 2,250 downstream tenants of ConnectWise Cloud Backup
C one secondary source states it
Kinds of data exposed
personal information
D inferred by us from the described behaviour, reason shown
Why this severity
The ConnectWise Cloud Backup supply chain compromise exposed more than 2,250 downstream tenants, meeting the many-third-parties test.
D inferred by us from the described behaviour, reason shown

Context

Industry
Finance and Insurance
D inferred by us from the described behaviour, reason shown
Country
South Korea
D inferred by us from the described behaviour, reason shown

LR-2026-0893 · version 7 · published 2026-11-16 after automatic checks; not yet checked field by field by a person