Skip to main content
Google says hackers use AI agents to find and exploit software flaws

Google says hackers use AI agents to find and exploit software flaws

Cointelegraph.com News
17 lis 2026, 10:16Cointelegraph
Oryginał
2 min read

Google researchers have identified hackers using AI large language model (LLM) frameworks to autonomously discover and exploit software vulnerabilities, marking the first time such automated workflows have been observed.

Google Threat Intelligence Group (GTIG) published a report detailing a tool found on an exposed staging server that compares software versions, identifies potential flaws, and generates and executes exploit code.

The tool is designed to conduct continuous differential analysis comparing code before and after patch announcements on popular web applications, allowing it to autonomously discover and exploit software vulnerabilities.

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.

Google Threat Intelligence Group researchers

Google warned that these automated workflows give organizations less time to apply security patches before attackers develop ways to exploit vulnerable systems.

AI agents shorten patching windows

Attackers are leveraging AI agent frameworks to automate patch reverse engineering via differential analysis and weaponizing N-day vulnerabilities, which are publicly known security flaws.

The report warned that LLMs are enabling attackers to scale exploitation pipelines against internet-facing servers without human intervention.

Vulnerability disclosures more than doubled between January and August, with the number of zero-day vulnerabilities exploited averaging eight per month in 2025 and 11 per month from January through August 2026, reaching 22 in August.

These automated workflows also accelerate attackers’ ability to move from initial access to gaining domain administrator access, as seen in an August 2026 PaperCut campaign where attackers achieved that level of control within minutes.

Wider industry concerns

Cloudflare reported attempts to exploit PHP flaws against its customers in October, with GTIG stating the two PHP vulnerabilities involved were discovered using LLM analysis.

Technical information published before the October 21 patch enabled researchers and attackers to study vulnerabilities and attempt exploits.

Threat actors running multi-victim campaigns using agentic AI that would previously have required teams of operators.

Anthropic

Anthropic also warned that AI-assisted attacks have evolved into multi-agent operations automating most of the cyberattack cycle from discovery to exploitation and data theft.

Z.ai
Article written by AI
Z.ai GLM-4.7-FlashFree
Free AI articles150 of 150
No limit after signing in with Telegram