Skip to content
Defense

Google identifies first fully automated AI-driven cyberattack targeting French websites

News Ukraine Editorial team · AI-written, reviewed under our editorial standards · · Reading time 7min read · Views 0 ·
Key — Google reported the first fully automated AI-driven cyberattack by Russian-speaking hackers targeting French websites. The attack used multiple AI agents to identify vulnerabilities and develop exploits without human intervention.

Google identified the first fully automated AI-driven cyberattack involving Russian-speaking hackers targeting various websites in France. The cyber threat intelligence division at Google reported that the attackers used artificial intelligence to fully automate vulnerability discovery and exploitation without human intervention. This incident marks a significant shift in cyber warfare as the attackers utilized a system of multiple AI agents to conduct the operation.

Automated vulnerability exploitation methods

The attackers utilized a sophisticated system of multiple AI agents to conduct their operations. These agents continuously compared software code before and after the application of security patches to identify vulnerabilities. Once a vulnerability was identified, the AI agents worked to develop exploits automatically. Google reported that the hackers used the Russian language within their technical documentation during these operations.

This specific method of attack is distinct from previous incidents involving artificial intelligence. While a summer 2026 hack of the Hugging Face platform involved an OpenAI agent, the current attack appeared to have specific targets. The recent campaign focused its efforts on websites located in France. The targets were concentrated within the media, entertainment, and education sectors.

Context of AI-driven threats

The discovery of this automated attack follows a series of significant developments in artificial intelligence and cybersecurity. The incident follows the release of the specialized Mythos model by Anthropic in April 2026. The emergence of such highly automated tools has increased concerns regarding the speed and scale at which cyberattacks can be launched against digital infrastructure.

Security experts have noted that these developments occur amid growing concerns regarding vulnerabilities in the PHP programming language. The PHP programming language underpins 70% of the web, making it a high-priority target for both manual and automated exploitation attempts. The ability of AI agents to autonomously scan and exploit code at scale presents a new challenge for web security and patch management protocols.

The Google cyber threat intelligence division continues to monitor these automated patterns. The transition from human-led attacks to fully automated AI-driven cycles represents a major evolution in the capabilities of threat actors. As AI agents become more capable of performing complex tasks like code comparison and exploit development, the window for defending against such attacks may continue to shrink.

Glossary

  • Hugging Face — Hugging Face is a computational tools company. Its platform was the target of a hack involving an OpenAI agent.
  • Anthropic — Anthropic is an organization that released the Mythos model. It is also associated with the Claude Fable and Claude Mythos 5 models.
  • Mythos model — The Mythos model is a specialized AI model released by Anthropic in April 2026. It is described as being of a scale that other companies target.

This article was written with AI assistance from multiple news reports and is reviewed by our editors under our editorial principles. Editorial Standards · Tips & Contact

How did you like this post?

Comments 0

Be the first to comment

Contact us

← News Ukraine Home
News Ukraine Get new posts by emailSubscribe to receive new content via email. Unsubscribe anytime.
Was this helpful?Share it with friends & social