Google observes that malicious actors are now using autonomous multi-agent offensive frameworks, accelerating intrusions in several stages. These groups also harness the power of large language models.
Multi-agent frameworks for faster attacks
Google revealed Monday that malicious actors have adopted autonomous multi-agent offensive frameworks, abandoning manual prompts from artificial intelligence. Among the tools cited are ARTEX, CAIRN, CyberStrikeAI and PentAGI. According to Google, these systems make it possible to carry out intrusions in several stages with increased operational speed.
The evening letterFree · 7 p.m.
Do you want the best AI tools before the rest?
We test and decipher the new AI tools every evening, in 5 minutes. Free.
Included upon registration: our selection of the best AI guides & comparisons.
Choose your pace
Free · No spam · Unsubscribe in 1 click
Exploitation of large language models
At the same time, Google indicates that the capabilities of large language models are also exploited by malicious actors.

