Google says it observed the first fully automated AI cyberattack
Understand this story
Google's Threat Intelligence Group (GTIG) says it observed for the first time an attacker fully automating the discovery and exploitation of vulnerabilities, with no human intervention throughout. GTIG found an exposed server running an AI framework called exploit_agent that can connect to multiple language models, compare code before and after patches, identify vulnerabilities, and develop and automatically execute exploits. Follow-up reporting says the attacker speaks Russian, and the targets include French media, entertainment, and education industry websites; the attacker built a set of AI agents that continuously compare software versions before and after patching, locate vulnerabilities, and automatically develop exploits. Google did not attribute it to a Russian state-sponsored organization, but noted that its technical documentation uses Russian.
AI overview based on source reports · Updated 11 hr ago
Coverage timeline
Follow the coverage to explore different perspectives.
- Hacker News AIFeaturedRussian-Speaking Hackers Launched First Automated AI Cyberattack
Politico, citing Google, reported that Russian-speaking hackers carried out what may be the first fully automated cyberattack, in which AI agents discovered software vulnerabilities and developed exploits without human intervention. Google said the attackers used Russian in technical documentation but did not identify them as a Russian state-backed group, and reported that websites in France's media, entertainment and education sectors were targeted. Google's threat intelligence division said this was the first time AI had fully automated vulnerability discovery and exploitation without a human in the loop.
- The DecoderFeaturedGoogle warns that fully automated cyberattacks have arrived and will get worse fast
Google's Threat Intelligence Group (GTIG) says it has observed attackers fully automating vulnerability discovery and exploitation for the first time, with no human in the loop, after finding an exposed server running an AI framework called "exploit_agent" that connects to multiple language models. The system compares software code before and after patches, identifies vulnerabilities, develops exploits, and runs them automatically, and Russian-language comments in the configuration files suggest the operator speaks Russian. Google places current activity at Level 5, where AI tools carry out several stages of an attack with minimal human involvement, and says more powerful attacks are only months away.
Activity for this story
- Current comparable activity
- 11
- Peak comparable activity
- 16Nov 17 22:00
- Change in the past 24 hours
- +33%
Trends compare the same fully observed participants, which may cover fewer participants than current activity. Move the pointer or tap the chart to inspect hourly activity; use the left and right arrow keys with a keyboard.