New Technique Medium TD-2026-0098 For CISO, SOC / IR, AppSec, AI Eng AI-assisted
Google GTIG reports threat actors using AI agent frameworks to automate vulnerability exploitation
Google Threat Intelligence Group says threat actors now use agentic AI workflows to automate vulnerability discovery and exploitation. GTIG identified an exposed staging server hosting a multi-agent framework called exploit_agent, used in attacks on French media, entertainment, and education websites. GTIG projects that sophisticated, fully autonomous agentic attacks are likely to emerge within six months.
Why it matters
Attacks against internet-exposed systems with low-complexity flaws can now run with little human input, shrinking the time defenders have to patch. Organizations face a higher volume of exploit attempts and AI-generated proof-of-concept code that must be verified.
Do this
- Prioritize patching internet-facing web applications and PHP components, and shorten patch SLAs for flaws with public exploit code.
- Hunt web server logs for high-rate, automated requests probing many endpoints or known vulnerable paths.
- Review exposed staging and development servers and remove public access to them.
- Validate AI-generated proof-of-concept exploits against your own test environment before acting on them.
- Monitor for signs of LLMjacking, such as unexpected API usage or spend on AI accounts, and rotate exposed keys.
Signals
- NoCISA KEV
- 25/100Priority
- 80%Confidence
Evidence
“In the past quarter, GTIG identified threat actors using LLM frameworks to fully automate vulnerability discovery and exploitation without a human in the loop for the first time”
“The researchers estimate that in the next three months threat actors will be able to launch unsophisticated attacks that require "little to no human interaction" and rely mostly on multiple AI agents.”
“Google warns that the deployment of such frameworks in attacks has dramatically decreased the time defenders have to patch their systems, while it also produces false leads.”
Sources
Affected / related
AI-driven automated vulnerability exploitation, exploit_agent, CAIRN, ARTEX, CyberStrikeAI, HexStrike AI, UNC6836, Google Threat Intelligence Group
Published 2026-11-16 14:00 UTC · updated 2026-11-16 20:05 UTC · AI-assisted summary, checked against quoted sources · How we score