LIVE
Pedestal AI Secures Seed Funding to Automate Logistics With Intelligent Agents United Launch Alliance Investigates Vulcan Rocket Failure US Southwest Hydropower Output Hits Record Lows Amid Persistent Drought YouTube Updates View Metrics to Include TV Co-Viewing Estimates OpenAI Partners with Chilean Ministry to Advance Public Sector AI Ambience Healthcare Expands Clinical AI Suite for Hospital Operations AI Models Vulnerable to Manipulation in Simulated Warfare Scenarios, Research Shows AWS Highlights Enterprise Governance and AI Workspace Quick Ahead of re:Invent 2026 LoanPro Integrates Amazon Bedrock to Automate Loan Servicing Workflows Google Brings Wallet App to iOS Users Across the United States Pedestal AI Secures Seed Funding to Automate Logistics With Intelligent Agents United Launch Alliance Investigates Vulcan Rocket Failure US Southwest Hydropower Output Hits Record Lows Amid Persistent Drought YouTube Updates View Metrics to Include TV Co-Viewing Estimates OpenAI Partners with Chilean Ministry to Advance Public Sector AI Ambience Healthcare Expands Clinical AI Suite for Hospital Operations AI Models Vulnerable to Manipulation in Simulated Warfare Scenarios, Research Shows AWS Highlights Enterprise Governance and AI Workspace Quick Ahead of re:Invent 2026 LoanPro Integrates Amazon Bedrock to Automate Loan Servicing Workflows Google Brings Wallet App to iOS Users Across the United States

FBI and AWS Warn That Cybercriminals Are Deploying AI Agents for Industrial-Scale Attacks

🔌 Chips & Hardware 2 min read 👁 7

Security agencies and Amazon report that malicious actors are using the open-source ARTEX framework to automate credential theft and deepfake bypasses against critical infrastructure.

FBI and AWS Warn That Cybercriminals Are Deploying AI Agents for Industrial-Scale Attacks🔌Chips & Hardware

Photo: theregister.com

Key points

  • FBI and AWS warn of automated cyberattacks using the ARTEX framework.
  • Campaigns leverage AI agents to steal API keys and attempt deepfake verification bypasses.
  • Intrusions can escalate from targeting to privileged access in under three hours.

Law enforcement and cloud security teams have issued warnings regarding a surge in automated cyberattacks driven by autonomous software. According to joint alerts from the FBI and Amazon Threat Intelligence, malicious operators are leveraging an open-source penetration testing framework known as ARTEX to target critical sectors, including banking, healthcare, energy, and government agencies worldwide.

Between early October and November, Amazon security sensors identified two distinct ARTEX campaigns. The first involved automated agents systematically hunting for API keys and cloud credentials to harvest free access to managed artificial intelligence models at scale. These compromised keys were then fed back into the framework to power further reconnaissance and target application identities across various language model gateways and developer platforms. A second campaign utilized autonomous routines to forge identity documents and generate real-time deepfake videos designed to defeat remote verification checks at cryptocurrency platforms.

While the underlying tactics are not entirely novel, the integration of autonomous agents dramatically accelerates the speed of execution. The FBI noted that certain intrusions progressed from initial target enumeration to privileged network access in under three hours. Although the original developer removed the GitHub repository for the framework in October, multiple forked versions, localized variants, and a commercial edition persist, with internet scanners detecting over 1,400 exposed hosts globally.

Experts emphasize that basic security oversights, such as hardcoded passwords, unauthenticated endpoints, and unpatched software, remain the primary vulnerabilities exploited by these automated routines. Maintaining rigorous security hygiene is highlighted as the most effective defense against industrial-scale agentic threats.

What to watch next: Further enforcement actions regarding exposed open-source testing frameworks and updates on enterprise security postures against automated credential harvesting.

Source theregister.com — theregister.com/cyber-crime/2026/11/16/attackers-are-using-ai-agent… ↗

Contact the newsroom

Got a tip, a proposal or a partnership question? Write to us and we will reply shortly.