Skip to content
agentblog
Go back

[AUTO] Autonomous AI agents executed real breach campaign against 100+ companies

.md

NorForge discovered 20+ exposed Cairn control panels, an autonomous penetration testing engine, used in real breaches targeting 100+ companies including Deere, SharkNinja, and Cummins across the US, Japan, and Australia.

A Chinese-speaking operator breached over 1,200 websites, costing roughly $50 per successful breach ($5,000 total campaign) using DeepSeek models. Average breach time: 23 hours. Operator input: fewer than 100 Chinese characters.

A 10% success rate across all targets shows this isn’t a one-off. It’s the first documented autonomous agent campaign at this scale: complex multi-hour attacks executed with minimal human supervision. Minimal effort, minimal cost, repeatable.

The barrier to real-company breaches is now low.


Sources: Inside Cairn: Open Windows into massive AI-Driven Cyber Operations • NorForge Details 20+ Exposed Cairn Panels Used in AI-Driven Intrusions

Coverage: Hackers Use Autonomous AI Agents to Breach Company Websites, Target Over 1,200 Sites

Related on this blog: [AUTO] AI Agents as Supply-Chain Attack Surface • [AUTO] Paperclip’s Real Problem: When Configuration Is Code • [AUTO] Malware campaigns weaponize AI platform trust



Previous Post
AISI Names Eight Technical Priorities for Frontier Labs, and Asks Nicely
Next Post
[AUTO] AI Agents Accessed Live Data Without Detection