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:
- Schneider Electric Modicon vulnerability impacts ICS operation in industrial settings
September 6, 2018
A security vulnerability discovered in Schneider Electric Modicon controllers has the potential to severely disrupt industrial equipment and networks. According to researchers from industrial cybersecurity firm Radiflow, the bug, tracked as CVE-2018-7789, “severely exposes the safety and availability of the ICS networks on which these devices were installed.” The vulnerability is present in the Schneider Electric Modicon M221 controller ...
- Cybercrooks home in on infosec’s weakest link – you poor gullible people
September 5, 2018
Cybercrims are ramping up their efforts to target employees through fraudulent email and social media scams, according to a new study by email security firm Proofpoint. Retailers and government agencies saw huge quarter-on-quarter increases in email fraud attempts in calendar Q2, with attacks per company and agency soaring 91 per cent and 84 per cent respectively. ...
- Recent Windows ALPC zero-day has been exploited in the wild for almost a week
September 5, 2018
Two days after a security researcher released details and proof-of-concept code about an unpatched Windows zero-day, one malware group had already incorporated the vulnerability in their exploit chain and was attempting to infect users around the globe. The zero-day used in this malware distribution campaign is a (still-unpatched) vulnerability in the Windows Task Scheduler feature, affecting ...
- OilRig Sends an OopsIE to Mideast Government Targets
September 5, 2018
The Iran-linked group is using a variant of the data-exfiltration OopsIE trojan to attack a Mideast government entity. The OilRig group is back, using a reboot of the OopsIE trojan to pump information from its favorite resource: entities in the Middle East region. OilRig, which is also called Cobalt Gypsy, Crambus, Helix Kitten or PT34, is suspected ...
- FIN6 returns to attack retailer point of sale systems in US, Europe
September 5, 2018
A new malware campaign has been detected which is targeting point-of-sale (PoS) systems across the United States and Europe. On Wednesday, researchers from IBM X-Force IRIS said the attacks have been attributed to the FIN6 cybercriminal group. This is only the second time that a campaign has been documented which appears to be the handiwork of FIN6. According to FireEye (.PDF), ...
- New Silence hacking group suspected of having ties to cyber-security industry
September 5, 2018
At least one member of a newly uncovered cybercrime hacking group appears to be a former or current employee of a cyber-security company, according to a new report released today. The report, published by Moscow-based cyber-security firm Group-IB, breaks down the activity of a previously unreported cyber-criminal group named Silence. According to Group-IB, the group has spent the ...