TL;DR
The FBI says hackers have used the open-source AI security-testing tool Artex in attacks on organizations since at least August. The agency reported that some attacks reached high-level access in less than three hours, while separate Microsoft and Amazon findings describe related activity and its limitations.
The FBI says hackers are using Artex, an open-source AI-powered security-testing tool, to break into organizations in attacks reported since at least August. In an advisory issued November 13, the agency said some intrusions took less than three hours from choosing a target to gaining high-level access, leaving defenders little time to respond.
Get privacy and security gear delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
The FBI said the activity has affected or targeted banks, government agencies, technology companies and healthcare providers. It cited attacks against South Korean financial institutions that exposed personal information belonging to about 67,000 people. The advisory describes Artex as a tool intended to automate security testing, but says attackers have used its capabilities to support intrusions.
Microsoft’s analysis says Artex agents can examine an organization’s defenses, exploit flaws and extract data. Automation can let attackers try multiple routes into a network at once. Microsoft also reported a separate campaign in which attackers breached more than 2,250 organizations’ Microsoft 365 systems, using stolen ConnectWise backup credentials alongside Artex. Microsoft cautioned that a successful sign-in does not by itself establish that data was accessed or stolen.
Amazon Web Services described another credential-focused scheme in research dated November 13. Amazon Threat Intelligence said it found the activity using decoy systems posing as AI services. Amazon linked the campaigns to Artex with moderate confidence, but said it could not independently verify results the agents reported from systems outside its monitoring. The FBI said the tool’s original GitHub repository was removed by its developer in October, although copies remain available.
AI Speeds Up Cyberattacks FBI Warns
The FBI says attackers are using the open-source tool Artex to speed intrusions. For defenders, the warning is about faster access and the need to investigate what happened after a suspicious sign-in.
How automation changes the response
The FBI’s under-three-hour figure describes some attacks, not an average for all incidents.
Artex agents can examine systems and identify weaknesses.
Automation can test multiple routes into a network at once.
Some reported intrusions moved from target selection to high-level access in less than three hours.
What the findings establish
The reporting points to real risks while leaving important questions open.
- The FBI says Artex has been used in attacks since at least August.
- Microsoft reported successful sign-ins in a separate campaign involving more than 2,250 organizations.
- Amazon linked activity to Artex with moderate confidence.
- A successful sign-in alone does not prove data was accessed or stolen.
- Amazon could not independently verify agent-reported results outside its monitoring.
- The available reporting does not link Artex to the FAA ground stop or Microsoft outage.
Actions for organizations
The FBI recommends reducing exposure and watching for signs of account or data misuse.
- Limit access to sensitive systems.
- Review software permissions and replace exposed credentials.
- Monitor unusual account activity and large data transfers.
- Investigate what suspicious sign-ins and connected services could access.
How Automation Shrinks Response Time
The FBI’s warning points to a practical risk for organizations: automation can compress the time between identifying a target and obtaining access. Security teams may receive alerts during an intrusion, but a fast sequence of automated attempts can leave less time to investigate and contain activity. The agency’s under-three-hour figure describes some attacks, not a measured average for all incidents.
The reported cases also show why access and confirmed data theft should not be treated as the same event. Microsoft said attackers successfully signed in to affected Microsoft 365 systems, while warning that this alone does not prove information was taken. For organizations responding to a possible breach, confirming what an account or connected service accessed is a separate investigative step.
The findings are part of broader concerns about AI agents that can perform tasks with limited human supervision. In this case, the concern is not that Artex independently caused every reported intrusion, but that attackers can use automated security-testing capabilities to speed up activities such as probing systems and trying to exploit weaknesses.
Products Worth Considering

Ledger Nano Gen5 Crypto Wallet with Touchscreen, Black

Ledger Flex Crypto Wallet with 2.8-Inch E Ink Touchscreen

LG 27G810A-B 27-inch UltraGear 4K UHD IPS Gaming Monitor
As an affiliate, we earn on qualifying purchases.
The FBI issued its cyber advisory on November 13, 2026. On November 17, the agency said the advisory had been released the previous Friday and was not a new announcement; it added that the guidance remained current. The activity described in the advisory dates back to at least August, rather than beginning on the advisory date.
The warning came on the same day that communications problems prompted a nationwide Federal Aviation Administration ground stop and one day after a major Microsoft 365 outage began. Microsoft attributed its disruption to technical problems, and the FAA said it had found no indication of a cyberattack. An FBI official told ABC News that Artex did not appear connected to either event or to other recent disruptions, including banking outages involving JPMorgan Chase. Those outages should not be conflated with the Artex activity.
Amazon’s research described attackers relying on default passwords, exposed credentials and outdated software—weaknesses that AI-assisted automation can help locate and exploit more quickly. The report’s link between that activity and Artex was qualified as moderate confidence, rather than a definitive attribution.
“The advisory was released last week because as we head to the end of the year…”
— The FBI, in its November 17 statement
Attribution and Data Access Questions
The FBI advisory and company research do not establish the full scope of Artex-related activity, the number of organizations where attackers successfully obtained access, or how much information was taken. Microsoft’s reported figure covers breaches of more than 2,250 organizations’ Microsoft 365 systems in a separate campaign; it does not mean data was stolen from every affected organization.
Amazon said its connection of the activity to Artex carried moderate confidence and that it could not independently verify agents’ reported results on systems outside its monitoring. The available information also does not establish that Artex was involved in the FAA ground stop, the Microsoft 365 service disruption, or the cited banking outages. The FBI official told ABC News the tool did not appear connected to those incidents, while the FAA and Microsoft described non-cyber causes for their respective disruptions.
Although the original Artex GitHub repository was removed in October, the FBI said copies remain available. The advisory does not quantify how widely those copies are being used or whether the reported activity has changed since the research and warning were published.
Steps Organizations Are Advised to Take
The FBI recommends limiting access to sensitive systems, reviewing software permissions, replacing exposed credentials and monitoring for unusual account activity or large data transfers. These steps address the access pathways and warning signs identified in the advisory and related research.
Organizations should continue investigating suspicious sign-ins and connected services rather than treating a login alert alone as proof of data theft. The FBI has said its November 13 guidance remains current, but the information provided does not announce a new advisory or a specific forthcoming update. Further details about the scale of the activity, affected organizations and verified data loss remain dependent on ongoing investigations and disclosures.
Source: CNET
Key Questions
What is Artex?
Artex is an open-source AI-powered security-testing tool designed to help identify security weaknesses. The FBI says attackers have used it to support intrusions; the original GitHub repository was removed in October, but copies remain available.
How quickly have some reported attacks moved?
The FBI said some attacks took less than three hours from selecting a target to gaining high-level access. That figure applies to some reported attacks and is not an average for all incidents.
Does a successful Microsoft 365 sign-in prove data was stolen?
No. Microsoft cautioned that a successful sign-in does not necessarily mean attackers accessed or stole data. Investigators need to determine what the account and connected services could reach and what activity followed.
Was Artex linked to the FAA ground stop or Microsoft outage?
The available reporting does not link Artex to either event. The FAA said it found no indication of a cyberattack, Microsoft attributed its disruption to technical problems, and an FBI official told ABC News that Artex did not appear connected to those incidents.
What does the FBI recommend organizations do?
The FBI advises organizations to restrict access to sensitive systems, review software permissions, replace exposed credentials, and monitor for unusual account activity or large data transfers.
Source: CNET
