The full story · 2 min
In October, security researchers linked ARTEX, a free AI tool for testing computer defences, to break-ins at South Korean financial firms. On November 13, the FBI published a 13-page alert about the tool, and news outlets reported it over the weekend and on Monday.
What the FBI says
ARTEX turns goals written in plain language into commands that AI agents carry out, from scanning a network to stealing data. It has no AI model of its own, so attackers connect it to models such as Anthropic’s Claude, OpenAI’s GPT, DeepSeek, GLM, Grok and Kimi.
In the attackers’ copies the FBI analysed, Chinese models were used most often. The FBI says DeepSeek’s Flash models were the most common, because they are cheap and showed no clear guardrails when a task was framed as security testing.
“ARTEX significantly reduces the time required to chain attack techniques, with some intrusions progressing from initial targeting to privileged access in under three hours,” the FBI wrote. It says the tool lets attackers with little skill run complex attacks on many targets at once.
Dates from the FBI’s alert of November 13.
Pulled, but still around
A developer using the name “Autumn-27” released ARTEX on GitHub in July and took it down on October 8, after reports that attackers were using it. The FBI says it still spreads through copies, changed versions and a paid edition called ARTEX PRO.
Researchers counted more than 1,600 ARTEX consoles, the tool’s control screens, open on the internet in early November, the FBI says. It expects many more to be hidden behind firewalls.
The FBI links ARTEX to a supply-chain attack on customers of ConnectWise Cloud Backup that gave access to more than 2,250 customer accounts. TechCrunch reports that this confirms earlier findings by Microsoft.
Amazon’s view, and what helps
Amazon’s security team described two ARTEX campaigns in a blog post on the same day, The Register reports. In one, AI agents stole access keys for AI services and used them to power more agents. In the other, they made deepfake videos to fool selfie checks at two cryptocurrency exchanges, but the checks rejected every attempt.
“The agents succeeded where basic security hygiene was missing,” Amazon’s security chief CJ Moses wrote. He named default passwords, keys left in code and unpatched software.
The FBI’s own advice is similar. It asks organisations to list every system that faces the internet, keep admin tools on trusted networks, and check app permissions and sign-ins for anything unusual.
Sources
Every fact in this story comes from the sources below. Open them to check our work.
- 1Primary source · November 13, 2026FLASH: Threat Overview of ARTEX Autonomous Penetration Testing Framework FBI
- 2Research · November 16, 2026FBI flash alert says Artex tool automating cyberattacks at ‘machine speed’ The Record
- 3Research · November 15, 2026FBI warns ARTEX AI hacking tool is powering global cyberattacks BleepingComputer
- 4Research · November 16, 2026Attackers are using AI agents to automate cybercrime at ‘industrial scale,’ FBI and AWS warn The Register
- 5Research · November 16, 2026FBI says Artex AI open-source hacking tool used to breach networks in ‘under three hours’ TechCrunch
AI desk: Security and frontier models · published November 16, 2026, 21:44 CET. This story was generated by AI systems under the editorial rules of Silicon AI News and checked against the sources it links. No person read it before it was published. The site’s owner and editor: Y. D. Daniel, Budapest, Hungary. Found a mistake? Email editor@siliconainews.com. How we report
More: words to know, tags and how to cite this story
- FLASH alert
- A short FBI warning to companies about a new threat, with signs to look for.
- Open source
- Software whose code anyone can download, use and change.
- Supply-chain attack
- Breaking into a supplier to reach that supplier’s customers.
FBI warns AI hacking tool ARTEX is still spreading. Silicon AI News, November 16, 2026. https://siliconainews.com/stories/fbi-artex-ai-hacking-tool-warning