PRESS RELEASE
Fujitsu and Singapore University of Technology and Design collaborate on GenAI-powered cyber twins to strengthen critical infrastructure resilience
Research will explore how virtual testing environments can help power-sector operators simulate cyberattacks, evaluate defences and assess their suitability for real-world systems
SINGAPORE, 17 November 2026
Fujitsu Limited and the Singapore University of Technology and Design’s (SUTD) iTrust, Centre for Research in Cyber Security, are pleased to announce a research collaboration to develop a generative AI-powered cyber twin framework for critical infrastructure, with an initial focus on the power sector. The 15-month project aims to help operators test and refine defences against cyberattacks affecting information technology (IT), operational technology (OT) and the connections between them, without experimenting on live operational plants.
The project addresses a practical challenge in critical infrastructure security: understanding whether a defensive measure will work without disrupting the physical systems it is intended to protect. A cyber twin provides a controlled environment that reproduces relevant system behaviour continuously, allowing researchers to examine possible attacks and test potential responses away from live operations.
Researchers will combine generative AI (GenAI) with physics-based models, protocol-level emulation and validated control logic to build twins spanning both IT and OT environments, leveraging iTrust's Electric Power and Intelligent Control (EPIC) testbed for initial development and evaluation. They will then correlate threat intelligence and use AI-driven attack generation to simulate attacks against the twins, and explore using them as honeypots to gather further threat intelligence.
The team will generate, test and refine candidate mitigation strategies, then investigate whether measures validated on one power-sector system remain effective on another, and under what conditions. A central research objective is to establish measurable criteria for when a cyber twin represents its physical counterpart well enough to support a particular cybersecurity assessment by comparing selected results with physical testbed experiments.
Together, these capabilities aim to establish a repeatable process connecting virtual testing, threat intelligence and mitigation evaluation, laying the groundwork for more informed decisions about real-world cyber defences. Meanwhile, Fujitsu Research is advancing its work in Physical AI Security and will explore the potential to leverage and integrate the technologies and outcomes developed through this joint research into its broader Physical AI Security framework.
For this project, Fujitsu will contribute multi-agent GenAI-based twin-generation workflows – in which multiple AI agents coordinate to support cyber twin generation – alongside AI-driven attack simulation, IT-side threat-intelligence tooling and researcher expertise. Meanwhile, SUTD's iTrust will contribute its power-sector testbed and cyber twin capabilities, OT threat research and mitigation expertise.
"Critical infrastructure operators need ways to assess cyber defences without putting essential services at risk," said Mr Tong Ker Yang, ASEAN CTO, Fujitsu Asia Pacific. "By combining our AI and security research with SUTD's cyber-physical systems expertise, we aim to develop an evidence-based approach to testing defensive measures and understanding where they can be applied. This collaboration reflects our commitment to working with research partners on practical industry challenges."
"Working with Fujitsu, we will investigate how generative AI can support cyber twin development, threat intelligence simulation and how to establish confidence in twin-based testing," said Mr Mark Goh, Assistant Director at SUTD iTrust. "The goal is to build knowledge and methods that support safer, more resilient systems."
Media contacts
Date: 17 November 2026
City: SINGAPORE
Company: Fujitsu Limited

