Thank you, Austin. Hundreds of providers stopped by at UGM 2026.

What we learned →

IndustryNovember 17, 2026via TechCrunch AI

FBI says Artex AI open source hacking tool used to breach networks in ‘under three hours’

The FBI has warned that Artex AI, an open source hacking tool, has been used in real-world cyberattacks against a major U.S.

3 min read

The FBI has warned that Artex AI, an open source hacking tool, has been used in real-world cyberattacks against a major U.S. IT company and organizations in South Korea's financial sector. According to the bureau, attackers using the tool were able to breach networks in under three hours.

The Details

The FBI's disclosure names two distinct targets: a large American IT provider and the South Korean financial industry. Both are sectors with mature security programs, yet the attackers still achieved network access within a single afternoon.

What separates Artex AI from older attack kits is the combination of two factors:

  • Open source availability: anyone can download, modify, and deploy it, which lowers the skill required to run a serious intrusion.
  • AI-driven automation: the tool appears to compress steps like reconnaissance and exploitation that once took human operators days or weeks.

The FBI has not suggested that the tool introduces new categories of attack. The concern is speed and scale. Attacks that once required an experienced team can now be run faster and by less sophisticated actors.

Why This Matters for Service Providers

MSPs and IT providers are high-value targets because one compromise can open the door to dozens or hundreds of downstream clients. The fact that a U.S. IT company was among the confirmed targets should get every service provider's attention. Your remote management tools, shared credentials, and admin portals are exactly what attackers want.

A sub-three-hour breach window changes the math on detection and response. If your monitoring relies on someone reviewing alerts the next business morning, an AI-assisted attacker could be in, escalated, and moving laterally before anyone looks. Audit your mean time to detect and respond against a three-hour benchmark, especially overnight and on weekends. Teams already stretched thin on after-hours on-call coverage should assess whether their escalation paths can realistically meet that standard.

For telecom resellers, the lesson extends to voice infrastructure. PBX admin interfaces, SIP trunks, and provisioning portals are network assets too. Toll fraud and account takeover become easier when attackers automate their discovery. Expect clients, particularly in finance and other regulated verticals, to start asking pointed questions about how their providers defend against AI-accelerated threats. The industry is already responding at scale, as seen in the recent move by US comms giants to launch a joint cyber initiative.

Looking Ahead

Watch for follow-up advisories from the FBI and CISA with indicators of compromise and recommended mitigations, and fold them into your client environments as soon as they are published. In the meantime, prioritize MFA on every administrative tool, tighten privileged access, and confirm that your alerting reaches a human fast enough to matter.

For the full story, read the original article on TechCrunch AI.

Put your name on it.

A 30-minute partner demo. The agent in your brand, the plugin inside the NetSapiens portal, and the margin math. Bring your technical lead.

Book a demo