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

The FBI has issued a flash cybersecurity advisory warning that hackers are using autonomous, AI-driven tools to launch intrusions across the U.S. and globally, with one open-source platform, Artex, linked to attacks moving from initial targeting to active compromise in under three hours in some cases. The campaigns cited include a widespread breach affecting customers of American IT company ConnectWise and a series of attacks across South Korea's financial industry. The advisory names financial services, government, technology, defense, telecommunications and internet providers, and law firms among the sectors targeted in recent months.
Artex belongs to a newer class of open-source, AI-powered penetration testing platforms, software built to probe a network's defenses. According to the FBI, hackers can connect Artex to hosted frontier large language models such as those from Anthropic and OpenAI, as well as to open-source models and models from outside the U.S. The tool automates reconnaissance: it rapidly scans a victim's network for weak points, then exploits simple security bugs — weak protections and exposed or stolen passwords — to gain a foothold. From there it helps the attacker move laterally toward databases, mailboxes and other sensitive data.
The FBI describes Artex intrusions as "noisy," generating large volumes of traffic and data that can trip a targeted company's defensive alerts. That detail matters for defenders: the tool is not stealthy so much as fast. The practical problem is not detection but reaction time. A human security team watching a legitimate alert still has to triage it, confirm it, and decide on a response, and the advisory's claim of compromise within under three hours suggests the tool compresses the window between first contact and real damage to a span shorter than many incident-response processes.
The ConnectWise case ties back to earlier findings from Microsoft, which the FBI confirmed. Hackers used Artex to breach ConnectWise and reach the Microsoft 365 cloud environments of at least 2,250 of its customers, then stole data from those affected customers. Microsoft has said the hackers are likely Chinese. A ConnectWise spokesperson did not respond to a request for comment on the FBI's claim. IT management vendors like ConnectWise are an attractive target in this model because a single compromise can fan out to many downstream customers — the supply-chain dynamic that made earlier incidents involving management software so damaging.
The South Korean attacks targeted a number of financial institutions and were attributed in large part to weak cybersecurity defenses, including the absence of protections such as multi-factor authentication. Those intrusions led to the theft of personal data belonging to at least 68,000 people. The FBI noted Artex has been used in several cyberattacks since its first public debut in July.
Security company Barracuda also disclosed that it was breached in an Artex-linked attack carried out by financially motivated criminals. In that case the tool found an exposed credential belonging to a third-party service that Barracuda had not revoked after ending work with the vendor. That access let the hackers reach stores of credentials Barracuda uses to collect data from customer systems. Barracuda said its logs were not detailed enough to rule out access, so it chose to assume the worst and changed all of its customer credentials. Its chief information security officer, Arve Kjoelen, wrote that defenders who rely on spotting malicious activity and then responding will often arrive after the attacker has finished. A Barracuda spokesperson told TechCrunch the company directly notified and advised a minority of XDR customers to revoke and recreate integration credentials, without specifying how many.
After the tool was used in these attacks, its developer made the software closed-source in an effort to prevent further abuse. The FBI says the earlier open-source version still circulates and remains widely accessible for use in attacks. Separately, researchers last week found hundreds of publicly exposed Artex systems online that had been placed on the internet without a password, likely through mistake or misconfiguration — including the system used to target South Korea. This suggests the offensive tooling is itself a liability: an exposed instance could hand powerful capabilities to anyone who finds it, which is a familiar pattern for internet-facing services left unauthenticated.
The broader picture is that AI is being used less to invent novel exploits than to industrialize the discovery and exploitation of ordinary weaknesses — unpatched systems, reused or leaked credentials, missing MFA. This is my own reading of the pattern, but it follows the cases described: the Korean failures are attributed to basic defensive gaps, and Barracuda's entry point was a stale credential rather than a sophisticated flaw. Two trade-offs stand out for security teams. First, agentic scanning is loud, so signal exists; the constraint is staffing and automation on the defense side, since a queue-based response model cannot match a sub-three-hour intrusion. Second, credential hygiene across third parties and integrations is exactly the kind of unglamorous control that AI-driven reconnaissance punishes hardest.
Why it matters: Companies running Microsoft 365 environments through IT management providers and financial institutions with weak authentication are the most exposed groups named in the advisory, and developers building integrations should assume leaked or stale credentials will be found and used quickly. For product and security teams, the practical shift is toward revoking unused third-party access, enforcing MFA, and automating detection-to-response rather than relying on analysts to chase alerts manually — my inference, based on how these intrusions are described. The open-source Artex code continuing to circulate means the risk outlasts the developer's decision to close the project.
Based on reporting from the original publisher. Visit the source for full context and later updates.
Publisher excerpt
The FBI says that the open source hacking tool was used in cyberattacks targeting a U.S. IT giant and the South Korean financial industry.