It’s the hacking equivalent of the Terminator: an intelligent, autonomous weapon that can stalk targets with ruthless efficiency.

And now, anyone can get one.

The tool is called ARTEX, a Chinese-made framework that puts military-grade AI hacking tools in the hands of garden-variety crooks — who no longer need deep pockets, fancy hardware or even much computer knowledge to wreak havoc.

Since it was unleashed in July, ARTEX is suspected of helping cyber criminals target organizations around the world, including ConnectWise Cloud Backup that left thousands of Microsoft 365 customers exposed, the FBI and cybersecurity organizations allege.

Microsoft attributed that attack to China-aligned hacking outfit Storm-0940, which used ARTEX “to automate attacks targeting hundreds of organizations globally across a broad range of industries.”

“Imagine giving a Terminator a mission in plain English,” Christopher Walcutt, chief information security officer at cybersecurity firm DirectDefense, told The Post.

“It identifies its targets, figures out how to reach them, executes the mission, and adjusts its approach when something doesn’t work.

“You don’t have to understand how any of that happens or know how to operate its weapons.”

ARTEX is an “automated penetration framework” that probes every facet of target systems in moments.

The technology has existed for years — developed to help test and secure computer systems — but it required massive resources and technical knowledge to use.

But with ARTEX, two-bit hackers can carry out attacks with little more than a smartphone and a dream, experts say.

“You no longer need years of technical experience to carry out an attack that once required a skilled team. … You can run it on your phone and connect it to other pieces of infrastructure that you set up, or that are even publicly available,” Walcutt said.

ARTEX uses natural language inputs, meaning someone can tell it what they want done in plain speech — like asking ChatGPT for a lasagna recipe.

It can deploy AI “worker agents” to perform multiple tasks at once, analyze what those agents learn, and spin up new, smarter minions to eventually pilfer credit card data, download confidential files, install malware, or pull off any number of other digital heists.

The framework was released in July on GitHub by a Chinese software developer — ostensibly as a diagnostic tool to find security flaws, according to researchers at cybersecurity firm Datadog.

It was taken down after two months, but by then, the AI genie was out of the bottle.

“As of early November 2026, researchers identified over 1,600 publicly exposed ARTEX console instances, with total deployments likely significantly higher when including systems behind firewalls or in private networks,” the FBI wrote in a recent memo.

ARTEX isn’t the only Terminator-style killbot on the net: Google’s Threat Intelligence Group identified several other penetration frameworks likely being used by bad actors.

One, “PentAGI,” was developed to run security testing in secure, black-box environments — but now the beast is loose and has been used for “active intrusions and persistence across multiple organizations,” the group said in a recent whitepaper.

Corporations and governments aren’t defenseless: They can and do use the same sorts of frameworks to test and eventually shore up their own systems against AI-powered attacks.

But it can be faster for the bad guys to exploit weaknesses than for their targets to update and patch them, said Scott Schober, cybersecurity expert, CEO of Berkeley Varitronics Systems and author of “Senior Cyber.”

“Unfortunately, I think sometimes the cyber criminals are ahead of the curve exploiting these advances in AI before the good guys can get around to doing it. And they’ve just got to get in once,” Schober told The Post.

“Once they get in, it becomes headlines, it becomes a data breach, it becomes a major problem or an outage. The good guys have to be fighting it every single day for many, many attackers going after their systems,” he said.

The scramble to update and patch has created problems of its own — including the wave of recent breakdowns at organizations ranging from the FAA to JPMorgan Chase, which experts have blamed on frantic attempts to patch security holes.

“Organizations need to understand where they’re exposed and whether their defenses can keep pace. AI doesn’t necessarily create new vulnerabilities. It makes the ones we already have much easier to find and exploit, and gives organizations far less time to react,” Walcutt said.

Share.
Exit mobile version