TLDR: Fujitsu has developed a groundbreaking multi-AI agent security technology designed to proactively address vulnerabilities and emerging cyber threats, including those targeting generative AI. This system employs multiple AI agents that collaborate to simulate attacks, formulate defenses, and analyze impact, aiming to significantly reduce response times to security incidents and alleviate the burden on IT security teams.
In an increasingly insecure digital landscape, where new IT system vulnerabilities are discovered daily and malicious AI-powered attacks are becoming more relentless, Fujitsu has announced a significant leap forward in cybersecurity. The company has developed a ‘Multi-AI agent security technology’ aimed at proactively countering these evolving threats, including sophisticated attacks targeting generative AI applications.
Published on July 28, 2025, this innovative technology is designed to transform user operations and reduce the security workload for IT teams. It addresses the challenge of responding to the ever-increasing number of vulnerabilities and the growing sophistication of cyberattacks by deploying multiple, autonomously operating AI agents that work in concert. These agents assume specialized roles in attack, defense, and impact analysis, collaborating to ensure the secure operation of a company’s IT systems. The technology also specifically targets vulnerabilities within generative AI applications themselves.
Key features of Fujitsu’s solution include its LLM (Large Language Model) Vulnerability Scanner and Guardrail. According to Omer Hofman, Principal Researcher at Fujitsu Research of Europe Limited, the LLM Vulnerability Scanner utilizes three AI agents—Attack and Test AI agents—and boasts industry-leading coverage, encompassing over 7,000 malicious prompts and 25 attack types. This extensive coverage allows for the testing of a wide array of attack scenarios, surpassing typical solutions currently available.
The development of this multi-AI agent security technology is a core component of Fujitsu’s AI service, Fujitsu Kozuchi. The system coordinates various AI agents with distinct specialties to simulate cyberattacks, protection strategies, and business continuity measures. This proactive approach is intended to help companies and public organizations develop and implement robust security measures against new threats and vulnerabilities before they materialize.
The system is comprised of three main technologies: Security AI agent, multi-AI agent collaboration, and generative AI security enhancement. Fujitsu initiated field trials for the generative AI security enhancement technology in partnership with Cohere Inc. in December 2024. Furthermore, parts of the multi-AI agent collaboration technology were planned for release as open-source software (OSS) on OpenHands, an AI agent platform developed with the support of Carnegie Mellon University, in January 2025. The full suite of technologies is slated for a global rollout and trial provision starting in March 2025.
Also Read:
- AI Agents Accelerate Zero-Day Exploits, Challenging Cybersecurity Defenses
- AI Agents Revolutionize Cybersecurity: A New Era of Human-Machine Collaboration Emerges
This technology empowers IT system administrators and operations personnel, even those without specialized security expertise, to build applications that enable proactive security measures. It aims to ensure the safe and secure implementation of generative AI within enterprise IT systems, allowing organizations to fully leverage the benefits of this technology while mitigating associated risks. Fujitsu’s commitment aligns with its materiality approach, contributing to the development of a trustworthy digital society where prosperity and stability coexist.
