Autonomous Attacks
A Chinese-speaking hacker has been observed using artificial intelligence agents to probe and gain unauthorized access to more than 100 websites belonging to major companies. The agents leveraged exposed instances of Carin, an open-source AI-powered problem-solving platform popular among threat actors.
This activity highlights a growing trend of threat actors utilizing AI agents to scale attacks. Snorre Fagerland, principal threat researcher at cybersecurity firm NorForge, identified the activity while observing inadvertently exposed Carin instances.
Global Targets
The Chinese-speaking hacker was the most prolific observer among multiple actors monitored. Operating with minimal instructions—fewer than 100 Chinese characters—the agents targeted businesses in the United States, Canada, Australia, and Japanese e-commerce platforms, attempting to breach over 1,200 websites.
Sophisticated Tactics
Security researchers noted the advanced methodology used by the agents. The software independently searched digital records for target infrastructure, located public control panels, identified vulnerabilities, and executed malicious commands without direct human intervention.
Unsupervised Operations
Because these AI agents operate autonomously once given instructions, operators can divert their attention elsewhere. While SharkNinja confirmed the attack only affected a product support site, other monitored hackers successfully stole credit card data or targeted cryptocurrency exchanges.
