Amazon finds hackers running ARTEX AI agents to steal AI model keys
Amazon's threat intelligence team says threat actors have been running cybercrime campaigns with ARTEX, an open-source agentic AI penetration testing framework, including a large-scale theft of credentials for AI models, according to an AWS security blog post. From October 3 to November 5, Amazon's MadPot decoy sensors captured requests from AI agents that contained the operating memory of these campaigns.
In the most prominent campaign, the agents found internet-exposed AI applications that store provider keys, such as LiteLLM, One API and Dify, extracted and validated the keys, and fed them into the framework's own model gateway, so stolen keys paid for the AI powering the search for more keys. Another campaign targeted cryptocurrency exchanges with real-time deepfake injection to bypass identity checks.
Amazon assesses with moderate confidence that ARTEX was used and that the activity comes from disparate Chinese-nexus groups, citing simplified Chinese prompts and China time-zone settings. It says it cannot verify the outcomes the agents reported, and has notified potentially affected customers and published indicators of compromise, The Hacker News reports.
- Nov 15 Japan's JPCERT/CC finds about 1,000 servers used as bases for AI-driven cyberattacks
- Nov 14 FBI warns ARTEX lets attackers automate intrusions at machine speed
- Nov 13 Traffic from AI hacking tools grew 30 times since ARTEX was taken down, Cloudflare says
We hope you enjoyed this article
Consider subscribing to one of our newsletters like Cybersecurity AI Weekly or Daily AI Brief.
Also, consider following us on social media:
More from Cybersecurity
Nov 15 Japan's JPCERT/CC finds about 1,000 servers used as bases for AI-driven cyberattacks Nov 15 Cursor says key exposed by Fireworks AI led to its breach and the malicious Claude Code extension Nov 15 AI tops the risks to US financial stability in Federal Reserve survey of market contacts Nov 15 Cohere completes AISLE acquisition and buys AI code review startup cubic Nov 15 Anthropic hired 33 counterintelligence staff this year as AI labs court former spiesCybersecurity AI Weekly
Weekly newsletter about AI in Cybersecurity.
Whitepaper
Tensordyne Napier: What If One Rack Could Do the Work of Nine?
Tensordyne
This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.
Read moreYou may also like
Check Point Finds ARTEX Agents Running Autonomous Cyberattacks
Hackers Use Stolen AI API Keys to Fund Autonomous Scans
Google Sees Attackers Using ARTEX and Other Agentic Frameworks in Intrusions
Censys finds hundreds of exposed deployments of the Artex AI hacking tool
Microsoft says Cambodia-based Storm-3358 uses ARTEX agents against financial firms
Daily AI Brief: the AI news that matters, in your inbox.