Google GTIG report: 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 adversaries have moved beyond manual prompting to deploying autonomous multi-agent offensive frameworks such as ARTEX, CAIRN, CyberStrikeAI and PentAGI during live intrusions. GTIG observed CAIRN used against US aerospace targets and a PentAGI-derived platform scanning organizations across Asia between late August and early October 2026, and it identified an exposed staging server running an exploit_agent framework that performs differential patch analysis and executes exploits without human intervention. The report also details LLMJacking, where stolen cloud and LLM credentials are resold through LLM Mixers at $0.10 to $25 per million tokens, with single-day fraudulent charges exceeding $100,000 USD in some incidents.
Same event; the pick carries“Google says Russian-speaking hackers ran first fully automated AI cyberattack”
Source:Hacker News · AI · cloud.google.com