Published 3 hours ago • loading... • Updated 31 minutes ago
Google warns that fully-automated AI attacks are coming soon
Google said a suspected Russian-speaking operator used an AI agent to turn patches into exploits and steal data with no human in the loop.
A suspected Russian-speaking operator used an AI agent to turn security patches into working exploits with no human intervention, according to Google Threat Intelligence Group, marking a first such autonomous cyberattack.
Adversary tradecraft has shifted beyond manual LLM prompting to deploying autonomous multi-agent offensive frameworks, Google researchers said on Monday, enabling attackers to execute multi-stage intrusions at greater operational velocity.
The framework used Russian-language tooling to conduct continuous differential analysis of "popular web application software," comparing code before and after security updates to identify vulnerabilities and develop exploits automatically.
Artex poses a serious threat to networks, the FBI warned on Friday, noting the framework "reduces the time required to chain attack techniques," with some intrusions reaching privileged access in under three hours.
Attackers have also leveraged other AI-powered offensive frameworks, including Cairn, CyberStrikeAI, HexStrike, and PentAGI, which Google believes threat actors will continue deploying in campaigns targeting organizations globally.
Russian language hackers created an agent based on artificial intelligence, AI, who executed what might be the first fully automated cyber attack in history, announced Google, according to...