Hermes Vector

Tech

Google says attackers built AI framework to turn software patches into exploits

Google's Threat Intelligence Group says attackers built an AI framework called exploit_agent that compares patched and unpatched code and writes exploits without human intervention.

By Hermes Vector

Checked against 2 sourcesventurebeat.comThe Next Web
Illustration for: Google says attackers built AI framework to turn software patches into exploits

In brief

  • Google says attackers built an AI framework called exploit_agent that turns software patches into exploits without human intervention.
  • The framework was found on an exposed staging server alongside stolen victim data and post-compromise scripts.
  • Google does not attribute the framework to a particular group or country, though a configuration file contained Russian-language comments.
  • The report also describes CAIRN and PentAGI AI frameworks being used in live intrusions.

Google says attackers built AI framework to turn software patches into exploits

Google's Threat Intelligence Group (GTIG) disclosed on Monday that attackers built an AI framework designed to turn software patches into exploits automatically. Google says this is the first time it has identified attackers using LLM frameworks to fully automate vulnerability discovery and exploitation "without a human in the loop."

What Google found

The framework, identified as exploit_agent in its configuration, integrates multiple frontier and local LLMs. It is set up to continuously compare code before and after software patches, identify potential vulnerabilities, draft exploits and execute them without human intervention.

Google found the tool on an exposed staging server alongside stolen victim data and post-compromise scripts. The report does not establish that the framework itself was responsible for those compromises.

A configuration file on the server contained Russian-language comments describing exploit_agent as a pilot and instructing the operator to enter LLM API keys directly into the file. Google does not attribute the framework to a particular group or country. Victim telemetry on the server suggests the operator was interested in targeting French media, entertainment and education websites, according to Google.

Google also published a code excerpt showing the framework examining access controls in version 2.19.25 of the Spectra WordPress plugin, but did not tie it to a successful exploit.

Context from the report

Google's latest AI threat report, produced with Mandiant, also describes AI frameworks including CAIRN and PentAGI being used in live intrusions. The report says several groups now use open-source AI hacking frameworks in live attacks, including financially motivated gangs and suspected Chinese spy groups. It saw the CAIRN framework used against targets including the US aerospace sector.

The report describes attackers who steal cloud and AI credentials to run models on victims' accounts. In some cases, the fraudulent charges topped $100,000 in a single day, it says. One attack used a flaw in the LiteLLM gateway. A suspected North Korean group changed Claude Code settings files inside companies so that each launch downloaded a backdoor. Google says it disabled the accounts and assets linked to misuse of its Gemini models.

Google did not identify a specific vulnerability exploited by the framework, provide a victim count or report a measured patch-to-exploit time.

Why it matters

The finding marks the first time Google has seen fully automated attack pipelines used against targets. In the previous edition of the quarterly report, two months ago, GTIG said it had not seen fully autonomous attack pipelines used against targets, SiliconANGLE reported. Google wrote that it believes it is highly likely that threat actors will continue to use these frameworks in campaigns targeting organizations globally.


Sources

This article was drafted with AI assistance and checked against the sources above. Company claims are reported as claims. Cover image is AI-generated.

The day's checked news, by email

One short email a day with the top stories. Free, and you can unsubscribe in one click.

Or follow for the latest news:

How do you feel about this story?

Comments 0

No comments yet. Start the discussion.

TechTrending

Websites including Zoom, HSBC and CNN subverted in crypto theft bid

Malicious JavaScript targeted visitors to 178 organizations' websites through Optimizely, attempting to steal cryptocurrency, card details, cookies and screenshots, and plant exploit code on iOS devices.

Checked against 4 sourcesinvesting.comeconomictimes.indiatimes.comnst.com.my+1 more

TechTrending

Nvidia Names Colette Kress Interim CEO After Jensen Huang Resigns

Nvidia's board accepted Jensen Huang's resignation as president, CEO and director effective November 11, 2026, and appointed CFO Colette Kress as interim CEO while searching for a permanent successor.

Checked against 4 sourcesmarkets.comlawstreet.cotokenpost.com+1 more