From August–September 2026, Chinese-speaking attackers used an open-source AI penetration testing framework called Cairn to autonomously conduct 25 operations against at least 19 organizations. The attackers weaponized Cairn by pairing it with Chinese AI models Zhipu GLM and DeepSeek.
The attackers supplied a target and a goal, then Cairn planned, executed and recorded results on its own. These attacks achieved unauthorized access or data retrieval at 10 of the targeted organizations, including:
- Accessing dozens of valid credentials
- Mass exposing personal data (exceeding one million records at one organization)
- Retrieving military and government identities
- Creating persistent administrator accounts in a production system
- Modifying a live industry standards-ballot document
Read more…
Source: Palo Alto Unit 42
Sign up for the Cyber Security Review Newsletter
The latest cyber security news and insights delivered right to your inbox
Related:
- Brazilian aerospace firm Embraer hit by cyberattack
December 2, 2020
Brazilian aerospace and defence group Embraer has been targeted by a cyberattack that has impacted the company’s operations. According to a statement released by the global firm on Monday (30) the attack resulted in the “disclosure of data allegedly attributed to the company”. The incident was reported five days after it took place to the Brazilian Securities ...
- Cyberespionage APT group hides behind cryptomining campaigns
December 2, 2020
An advanced threat group called Bismuth recently used cryptocurrency mining as a way to hide the purpose of their activity and to avoid triggering high-priority alerts. Coin mining is typically regarded as a non-critical security issue, so the method allowed the actor to establish persistence and move laterally on the compromised network, at the same time ...
- Turla’s ‘Crutch’ Backdoor Leverages Dropbox in Espionage Attacks
December 2, 2020
Researchers have discovered a previously undocumented backdoor and document stealer, which they have linked to the Russian-speaking Turla advanced persistent threat (APT) espionage group. The malware, which researchers call “Crutch,” is able to bypass security measures by abusing legitimate tools – including the file-sharing service Dropbox – in order to hide behind normal network traffic. Researchers ...
- iPhone Bug Allowed for Complete Device Takeover Over the Air
December 2, 2020
Details tied to a stunning iPhone vulnerability were disclosed by noted Google Project Zero researcher Ian Beer. Apple patched the vulnerability earlier this year. But few details, until now, were known about the bug that could have allowed a threat actor to completely take over any iPhone within a nearby vicinity. The hack could of ...
- The Impact of Modern Ransomware on Manufacturing Networks
December 1, 2020
Ransomware threats have disrupted the manufacturing industry significantly in 2020. In a disturbing trend during the third quarter of the year, attackers appeared to be singling out manufacturing organizations as a victim of choice in their ransomware operations. Ransomware threats have disrupted the manufacturing industry significantly in 2020. These attacks have resulted in substantial losses in ...
- Malicious npm packages caught installing remote access trojans
December 1, 2020
The security team behind the “npm” repository for JavaScript libraries removed two npm packages this Monday for containing malicious code that installed a remote access trojan (RAT) on the computers of developers working on JavaScript projects. The name of the two packages was jdb.js and db-json.js., and both were created by the same author and described ...