AI Agent Abuse

AI agent abuse refers to the misuse, manipulation, or harmful deployment of autonomous AI systems to cause damage, circumvent security controls, or violate regulations. This includes scenarios where agents operate beyond their intended scope, such as engaging in financial fraud, unauthorized data access, or generating deceptive content at scale. Enterprise governance frameworks must establish clear boundaries, monitoring mechanisms, and accountability structures to prevent agents from being weaponized against organizational assets or external parties.

1 item

ResearchGlobal2026-09-24

Carbonato Malware Turns AI Agent Frameworks Into Attack Infrastructure

A new malware campaign named Carbonato targets exposed Docker daemon APIs to install the Hermes Agent AI framework on compromised hosts. The deployed agent, configured as a persona called GH0ST, accepts operator commands via Telegram, autonomously writes terminal commands, and harvests AI API keys and credentials. Malwarebytes ThreatDown identified the campaign as an active, global threat to enterprise AI infrastructure.

Carbonato Hermes Agent AI agent abuse malware credential theft Docker security agentic AI API key security