GTIG AI Threat Tracker: Agentic Threat Activity Crosses a Significant Threshold
GTIG AI Threat Tracker: Agentic Threat Activity Crosses a Significant Threshold
Google Threat Intelligence Group and Mandiant report that threat actors have moved beyond manual prompting to deploying autonomous multi-agent offensive frameworks such as ARTEX, CAIRN, CyberStrikeAI and PentAGI, and that GTIG for the first time identified a framework, dubbed exploit_agent, fully automating vulnerability discovery and exploitation without a human in the loop. The report also covers LLMJacking for reselling access to stolen LLM credentials and cloud resources, full LLM integration into information operations including turnkey Influence-as-a-Service platforms, and espionage use of LLMs for target research and supply chain attacks. GTIG notes these frameworks currently trade stealth for velocity but are highly capable of achieving operational objectives.
It offers a detailed, telemetry-grounded survey of how agentic AI is being operationalized across cyber crime, espionage and influence operations, useful for defenders tracking the shift to machine-speed intrusions.
Source: Hacker News AI · cloud.google.com