What Is the Role of Singapore University of Technology and Design’s (SUTD) Cybersecurity Lab in AI Security

In a world where artificial intelligence shapes everything from daily apps to critical systems, keeping AI safe from cyber threats is more important than ever. Singapore University of Technology and Design, or SUTD, leads the way through its iTrust Centre for Research in Cyber Security. This lab focuses on protecting cyber-physical systems, which often rely on AI for smart operations. iTrust combines design thinking with advanced tech to tackle risks in AI-driven environments. Their work helps secure things like water treatment plants and autonomous vehicles, where AI meets the real world. This blog explores iTrust's role in AI security. We cover their research, testbeds, and training efforts. Even if you are new to these ideas, the explanations stay simple and clear.

Oct 7, 2025 - 15:13
 5

Table of Contents

Overview of SUTD's iTrust Cybersecurity Lab

iTrust, established in 2012 by SUTD and Singapore's Ministry of Defence, serves as a key hub for cyber security research. It targets threats to cyber-physical systems, blending fields like control theory, artificial intelligence, and software engineering. These systems use AI to connect digital controls with physical processes, such as smart grids or self-driving cars. iTrust's mission is to understand attacks and create defenses that work in real life. As a National Satellite of Excellence, it receives funding from the Cyber Security Agency to advance secure critical infrastructure.

The lab hosts advanced facilities, including testbeds that mimic real-world setups. This allows researchers to simulate AI-integrated systems and test security measures safely. iTrust also runs programs like the ST Electronics-SUTD Cyber Security Laboratory, a $44.3 million partnership focused on next-gen solutions. Their interdisciplinary approach ensures AI security research addresses both tech and human factors, making systems resilient against evolving threats.

For beginners, cyber-physical systems are like smart factories where AI decides actions based on data. iTrust ensures these do not fail due to hacks, protecting society from disruptions.

AI's Role in Modern Cybersecurity

Artificial intelligence powers cybersecurity by spotting patterns in huge data sets that humans might miss. In AI security, the focus shifts to protecting AI itself from attacks like data poisoning, where bad info tricks models into wrong decisions. iTrust uses AI foundations to build defenses for systems where AI runs operations, such as anomaly detection in networks.

  • Threat Detection: AI analyzes traffic to flag unusual behavior in AI-controlled devices.
  • Mitigation Strategies: Develops tools to counter attacks on AI models in IoT or CPS.
  • Automation: Speeds up responses to incidents in AI-integrated environments.
  • Resilience: Ensures AI systems recover from manipulations without halting operations.

At iTrust, AI integrates into research thrusts like CPS and IoT security. For example, machine learning helps protect battery systems or autonomous vehicles. This dual role, using AI for security and securing AI, makes the lab vital in Singapore's Smart Nation push.

Think of AI as a smart guard dog: it learns threats but needs protection from poisoned food, which iTrust researches.

Research Contributions to AI Security

iTrust drives projects that secure AI in critical setups. Their work on cyber-physical protection uses AI to model threats and defenses. Key efforts include machine learning for physical layer security in batteries and predictive maintenance in water systems.

  • CPS Reconnaissance: Uses causal graphs and design principles to spot AI vulnerabilities.
  • IoT Firmware Analysis: AI infers device functions to enhance dynamic security checks.
  • Attestation Techniques: Verifies AI-driven control systems against tampering.
  • Incident Response: Automates recovery in AI-managed infrastructures.

Research draws from global partners like MIT and focuses on translatable tech. For AI security, they address risks in interconnected systems, ensuring models resist adversarial inputs. Outputs include guidelines for maritime OT and energy grids, where AI optimizes operations. This builds secure AI ecosystems, reducing breach impacts.

The lab's emphasis on design science integrates AI securely from the start, preventing flaws in deployment.

Testbeds and Practical Applications

iTrust's testbeds simulate AI-integrated environments for realistic testing. The Secure Water Treatment (SWaT) testbed models AI-controlled processes, allowing attack simulations and defense trials. Others like WADI and EPIC focus on water distribution and power systems, incorporating AI for intelligent control.

  • MariOT: Hybrid platform for maritime AI security, validating defenses without real-ship risks.
  • RESILIOT: Tests IoT devices with AI analysis for firmware vulnerabilities.
  • Cyber Twins: Digital replicas for training AI responses to threats.
  • EPIC: Secures AI in electric power grids.

These facilities support remote access for researchers, fostering innovation in AI security. Applications extend to autonomous vehicles and blockchain, where AI ensures safe transactions. By experimenting here, iTrust translates theory to practice, enhancing Singapore's defenses.

Testbeds act like virtual proving grounds, where AI security measures are battle-tested before real use.

Education and Training Initiatives

iTrust supports SUTD's programs blending AI and cybersecurity. The Design x AI x Tech certification trains non-tech workers in secure AI practices. Master's in Security by Design teaches AI threat modeling.

  • Workshops: Hands-on with testbeds for AI defense strategies.
  • Student Projects: Involves undergrads in AI security research.
  • Industry Training: Custom sessions for executives on AI risks.
  • Scholarships: Supports AI-cyber focus via Future Communications program.

These build talent pipelines, with students gaining skills in AI anomaly detection and response. iTrust's involvement ensures practical, AI-centric education.

Training demystifies AI security, preparing professionals for hybrid threats.

Collaborations and Broader Impact

iTrust partners with industry like ST Electronics and agencies like MPA for AI-secure innovations. Global ties with MIT advance AI in CPS security.

  • Government: Informs policies via CSA funding.
  • Industry: Develops products for AI-protected infrastructure.
  • Academia: Joint projects on AI threats.
  • International: Shares testbed insights.

Impact includes resilient Smart Nation systems and trained experts. iTrust's work influences standards, ensuring AI security scales nationally.

Collaborations amplify research, turning AI security into widespread protections.

Summary of Key Projects

Here is a table summarizing iTrust's projects related to AI security.

Project Focus AI Security Role
Machine Learning for Battery Security Physical layer protection Uses AI to detect cyber threats in energy systems.
CPS Reconnaissance Threat modeling AI graphs for vulnerability analysis.
Automated Incident Response Recovery in ICS AI automates defenses.
MariOT Testbed Maritime OT Validates AI security tech.

Conclusion

SUTD's iTrust lab plays a pivotal role in AI security by researching threats, building testbeds, and training experts. Through AI-integrated defenses for CPS and IoT, it safeguards critical systems. Collaborations extend impact, fostering a secure digital future for Singapore and beyond.

Frequently Asked Questions

What is iTrust at SUTD?

iTrust is SUTD's cyber security research centre focusing on threats to cyber-physical systems using AI and other tech.

How does iTrust use AI in research?

It applies AI from control theory for threat mitigation in CPS.

What are cyber-physical systems?

Systems blending digital AI controls with physical processes like water treatment.

What testbeds does iTrust have?

SWaT, WADI, EPIC, and MariOT for simulating AI-secure environments.

How does iTrust contribute to AI security?

Through projects like ML for battery protection and CPS reconnaissance.

What partnerships does iTrust have?

With ST Electronics, MPA, and globals like MIT.

What education programs link to iTrust?

Design x AI x Tech certification and Master's in Security by Design.

What is the SWaT testbed?

A secure water treatment simulation for AI threat testing.

How does iTrust train professionals?

Via workshops and cyber twins for hands-on AI security.

What funding supports iTrust?

CSA as National Satellite of Excellence.

What is MariOT?

Maritime testbed for AI-secure ship systems.

How does iTrust address IoT security?

Through RESILIOT lab analyzing AI in devices.

What is NSoE DeST-SCI?

Excellence program for secure infrastructure with AI focus.

Does iTrust research autonomous vehicles?

Yes, integrating AI safety and security analysis.

What tools does iTrust develop?

Attestation for AI control systems and honeypots.

How does iTrust impact Smart Nation?

By securing AI in critical infrastructure.

What students work at iTrust?

PhDs and undergrads on AI-cyber projects.

What is EPIC testbed?

Electric power system for AI security research.

How does iTrust use design thinking?

In secure AI system development from start.

What future threats does iTrust target?

AI manipulations in CPS and IoT.

What's Your Reaction?

like

dislike

love

funny

angry

sad

wow

Ishwar Singh Sisodiya I am focused on making a positive difference and helping businesses and people grow. I believe in the power of hard work, continuous learning, and finding creative ways to solve problems. My goal is to lead projects that help others succeed, while always staying up to date with the latest trends. I am dedicated to creating opportunities for growth and helping others reach their full potential.