FBI Dallas
@FBIDallas · 7h
The #FBI issued a FLASH advisory warning that cybercriminals and state-aligned hackers are exploiting ARTEX, a super intelligence-powered penetration testing framework, to automate attacks against businesses and organizations worldwide. ARTEX can accelerate intrusions from initial targeting to privileged access in under three hours.
It was used in the October ConnectWise supply-chain compromise, which exposed Microsoft 365 customer environments and involved attempted authentication against more than 2,250 downstream tenants.
The FBI is urging businesses and organizations to strengthen access controls, monitor suspicious automated activity, and report potential compromises to http://ic3.gov.
ARTEX is open-source code that does not run on its own. It has no language model inside it.
Read that 5 times.
The framework sends goals written in plain language to external models and receives the steps those models generate.
The FBI advisory lists the models it connects to: Claude from Anthropic, GPT from OpenAI.
The same advisory states that Chinese models, mainly DeepSeek, appeared most often on the servers used in the attacks it tracked. But clearly this would be the perfect fake. Why? Blame open source. DeepSeek by their own definition is far weaker than Anthropic and OpenAI models, right?
Because the program only works when it can reach one of those services, the attack speed the FBI described depends on the availability and pricing of those cloud APIs.
Removing the original GitHub repository in October 2026 did not change that dependency; copies and the commercial version still call the same external models. The result is a free piece of software whose practical power is limited by the policies and access controls of the companies that host the models.
ARTEX Is simple software and not AI and is already duplicated in many other versions.
ALL get real power from Anthropic and OpenAI. Not free open source “weaker” AI models.
Are we clear?