Threats · 15h ago
Palo Alto Networks’ Unit 42 says operators used the open-source AI penetration-testing framework Cairn in 25 operations against at least 19 organizations in August and September 2026, gaining access or retrieving data at 10. Unit 42 assesses with moderate confidence that the activity was China-nexus intelligence collection, tracked as CL-STA-1238.
Operators gave Cairn a target and a goal. Its AI model reviewed confirmed findings, chose a next move, and logged the result before continuing. That let one operator or a small team pursue different weaknesses at multiple targets in parallel. Outcomes included valid credentials, access to more than one million member records at one organization, persistent administrator accounts, and an altered live standards ballot.
For organizations with public-facing identity, voting, meeting, or cloud-storage systems, a single shared exploit or indicator cannot define the exposure: different weaknesses can leave different access paths and consequences at each target.
3 sources covering this story
Palo Alto tracks a massive Chinese AI attack that sadly lives up to industry fears
Vastly different attack vectors were used as the AI tailored its campaign to each target.
Attackers Use Autonomous AI to Target Semiconductor, Defense and Civil Society Sectors
Our analysis of how attackers used the open-source Cairn AI agent framework to autonomously breach semiconductor, defense and civic targets.
Attackers breached 10 organizations with an open-source AI pentest framework - Help Net Security
Chinese-speaking attackers ran autonomous AI cyberattacks using Cairn, gaining access or data at 10 of at least 19 targeted organizations.
Part of the PlainSec briefing for 2026-11-13