Skip to the content
AI and Cybersecurity Learning and Practice Center Test automation Open the Learning Center

NewsResearch

Google: attackers used an AI framework to find and exploit web app flaws with no human in the loop, a first in its tracking

Google Threat Intelligence Group's AI Threat Tracker, published on November 16, 2026, says threat actors are moving to autonomous workflows and multi-agent frameworks such as ARTEX, CAIRN, CyberStrikeAI and PentAGI. GTIG says it saw CAIRN and PentAGI used in the wild, CAIRN against targets including the US aerospace sector, and experimentation with CyberStrikeAI and HexStrike AI. In the past quarter it found, for the first time, attackers using LLM frameworks to fully automate finding and exploiting vulnerabilities: a framework dubbed exploit_agent, on an exposed staging server, compares web app software updates, drafts exploits and runs them without a person, with interest in French media, entertainment and education sites. GTIG says disclosures and exploits nearly doubled between January and August 2026, and that after PHP patched two flaws on October 21, hundreds of claimed but non-working proofs of concept, likely written with LLMs, were published. It also reports LLMs in routine use for influence operations, abuse of stolen model access (LLMjacking), and espionage groups tampering with coding assistant settings. GTIG judges it highly likely attackers keep using these frameworks, which for now trade stealth for speed.

Why it matters

The time between a patch and an exploit is shrinking, and some of the code that floods in after a disclosure is fake. Patch internet-facing web apps first, test a proof of concept before you trust it, and treat a coding assistant's settings file as something an attacker may change.

Sources

Read it with the rest of the news in the Learning Center