Research Lab

Decentralized Cybersecurity & Artificial Intelligence Lab
(DCAILab)

Building secure, privacy-preserving, and intelligent systems at the intersection of decentralized cybersecurity and artificial intelligence to protect critical infrastructures.

Explore Research

Research Areas

Our work spans critical areas of cybersecurity and Artificial Intelligence, focusing on building robust, privacy-preserving systems for healthcare, industrial IoT, and smart environments.

🤖

AI-driven and Autonomous Cybersecurity

Designing closed-loop autonomous cybersecurity systems that continuously perceive threats, reason under uncertainty, and take adaptive defensive actions with minimal human intervention.

🛡️

Security of Agentic and Autonomous AI Systems

Studying the security, reliability, and controllability of agentic and autonomous AI systems, including adversarial manipulation, unsafe emergent behavior, verification, and deployment-time guarantees.

🔗

Federated and Decentralized Learning

Developing federated and decentralized learning systems that enable cross-organizational collaboration while addressing trust, robustness, auditability, and adversarial behavior.

🧠

Federated Unlearning & Machine Forgetting

Designing secure and auditable machine unlearning frameworks that enable the selective removal of data, knowledge, or behaviors from AI systems. Our research explores federated unlearning, agentic AI unlearning, privacy compliance, machine forgetting, verifiable deletion, and trustworthy post-deployment AI governance.

🔐

Privacy-Preserving Machine Learning

Designing privacy-preserving learning frameworks with formal guarantees for sensitive domains, integrating differential privacy, secure computation, and cryptographic safeguards into real-world AI systems.

⚙️

Cyber-Physical and Industrial IoT Security

Protecting cyber-physical and industrial IoT systems through AI-driven detection, adaptive defense, and resilient control mechanisms for critical infrastructure.

🔍

Threat Intelligence & AI Governance

Integrating AI-driven threat intelligence with governance frameworks that support explainability, accountability, and coordinated decision-making in security operations.

🌐

Blockchain & Web3 Agent Security

Studying the security and trustworthiness of blockchain-enabled autonomous agents in Web3 ecosystems, focusing on adversarial behavior, economic manipulation, and governance.

Current & Industry Projects

Ongoing funded and collaborative projects at DCAILab focused on trustworthy AI, federated learning, cybersecurity, privacy-preserving systems, and AI governance.

NCC Funded
Co-Principal Investigator (Co-PI)
Cyber-Synthetic Risk: A Generative Multi-Agent Approach to Forecasting Advanced Cyberattack Impacts on Enterprise Ecosystems
Technical Focus: Generative AI for Cybersecurity, Multi-Agent Systems, Cyber Risk Forecasting, Threat Intelligence, Enterprise Resilience, Adversarial Simulation Sponsor: National Cybersecurity Consortium (NCC) Industry Partners: Fortinet Canada  ·  InCloud Security Inc.  ·  Riskview Systems  ·  Toronto Community Housing Period: 2026 – 2027 Grant: NCC Research Grant
This NCC-funded project develops an AI-driven cybersecurity forecasting platform capable of simulating advanced cyberattacks, including zero-day exploits and ransomware campaigns. The platform combines generative AI, adversarial simulation, multi-agent attacker-defender modeling, and adaptive cyber risk analytics to help organizations proactively understand, quantify, and mitigate emerging cyber threats before they materialize.
  • Develop generative cyber threat simulation engines using advanced AI models.
  • Build multi-agent attacker-defender environments for enterprise cybersecurity forecasting.
  • Design adaptive cyber risk analytics and decision-support systems.
  • Integrate real-time threat intelligence and operational enterprise data.
  • Validate cybersecurity forecasting capabilities with industry and public-sector partners.
  • Advance cybersecurity technologies from TRL-4 to TRL-6.
Project Details →
Awarded
Co-Investigator & Technical Lead
AI for the Global Majority (AI4GM) Initiative
Technical Focus: Federated Learning, Privacy-Preserving AI, and Cybersecurity Safeguards Partners: Geneva Graduate Institute  ·  International Telecommunication Union (ITU)  ·  Microsoft Period: December 2025 – July 2026
Strengthening Societal Resilience through Inclusive AI Governance and Capacity Building for the Global Majority. The project develops federated AI architectures for health and climate applications, privacy-preserving machine learning safeguards, AI governance tools, risk monitoring frameworks, and policy recommendations.
  • Lead the technical architecture of a privacy-preserving federated monitoring and AI-literacy dashboard.
  • Design federated learning pipelines for health and climate-related AI use cases.
  • Develop privacy and security safeguards including secure aggregation, differential privacy, audit mechanisms, adversarial testing, and residual-risk documentation.
  • Assess federated learning vulnerabilities, including membership inference, gradient inversion, model poisoning, data leakage, and misuse risks.
  • Translate technical findings into accessible AI-literacy modules and policy recommendations.
Project Link →

Mission & Vision

The Decentralized Cybersecurity and Artificial Intelligence Lab (DCAILab) is dedicated to advancing the frontiers of secure and intelligent systems. Our research combines cutting-edge AI techniques with robust cybersecurity frameworks to address the most pressing challenges in protecting critical infrastructures.

We focus on developing privacy-preserving machine learning algorithms, secure federated learning systems, and autonomous cybersecurity solutions that can defend against sophisticated threats while maintaining user privacy and system integrity.

Our interdisciplinary approach brings together expertise in artificial intelligence, cryptography, distributed systems, and cybersecurity to create innovative solutions for healthcare, industrial IoT, smart cities, and emerging Web3 technologies.

15+
Research Projects
60+
Publications
10+
Collaborations
7
Focus Areas

Meet the Researchers

Our diverse team of researchers, engineers, and students working together to advance cybersecurity and AI.

Principal Investigator

Dr. Abbas Yazdinejad
PI – Lab Director
Leading research in decentralized AI and cybersecurity for critical infrastructure.

Students (Supervised / Co-supervised)

Ali Mohammadi Ruzbahani
PhD Student, University of Calgary
Research focus: Blockchain and Decentralized AI
Faisal Popalzai
MSc Student, University of Regina
Research focus: Cybersecurity and AI Context Drift
Humayoun Karimi
MSc Student, University of Regina
Research focus: Cybersecurity and AI
Sana Azizi
MSc Student, University of Regina
Research focus: Shadow AI Security
Open PhD Position
PhD Student, University of Regina
Research focus: Agentic AI Security
Open Postdoctoral Fellow Position
NCC Project  ·  Fall 2026
Research focus: AI-Driven Cyber Risk Forecasting & Enterprise Resilience

Research Collaborators

Dr. Gelan Zewdie
Research Collaborator
Dalla Lana School of Public Health, University of Toronto
Research focus: AI for Health & Health Informatics
Maral Niazi
Research Collaborator
Balsillie School of International Affairs, University of Waterloo
Research focus: AI Governance

Recent Publications

Explore our latest research contributions to the fields of cybersecurity and artificial intelligence.

Loading publications...

Talks & Outreach

Invited talks, panels, and public engagement activities by DCAILab.

Invited Talk
Agentic AI in Cybersecurity: Autonomy, Attack, Defense, and Global Governance
Balsillie School of International Affairs · Jan 27, 2026
Official Page Video
Invited Talk
Toward Human-Aware Autonomous Cyber Defense: Cognitive–Physiological Intelligence for Adaptive Security Operations
IEEE Control Systems Society (CSS) - Rising Star Symposium on Cyber-Physical Systems Security, Resilience, and Privacy · Tue Mar 31, 2026
Official Page Zoom Registration
Invited Talk
SPS Webinar: A Robust Privacy-Preserving Federated Learning Model Against Model Poisoning Attacks
The IEEE Signal Processing Society, 27 April 2026, 12:00 PM - 1:00 PM (ET)
Official Page

Open Positions

We are recruiting highly motivated students to work on cutting-edge research in Agentic AI, Federated Learning, and Cybersecurity for Critical Infrastructure.

PhD Position

Fully Funded PhD — Winter 2027

Research topics include Agentic AI security, privacy-preserving machine learning, federated unlearning, and AI governance for critical infrastructure systems.

  • Strong background in AI / ML / Cybersecurity
  • Programming experience in Python
  • Interest in research publications and innovation
MSc Position

MSc Position — Spring 2027

Work on applied AI-driven cybersecurity projects and contribute to real-world systems involving IoT, smart grids, and decentralized AI.

  • Background in Computer Science or Engineering
  • Interest in AI and cybersecurity applications
  • Motivation for hands-on research

News

Recent highlights including publications, invited talks, workshops, grants, and student achievements.

Updates Coming Soon

We are currently preparing recent announcements and activities from DCAILab. Please check back soon for updates.

Call for Papers

Selected conferences, workshops, and special issues closely aligned with the research interests of DCAILab in cybersecurity, federated learning, privacy-preserving AI, cyber-physical systems, and autonomous security.

Workshop
The 3rd International Workshop on Autonomous Cybersecurity (AutonomousCyber 2026)
Venue: In conjunction with the 31st European Symposium on Research in Computer Security (ESORICS 2026)
Location & Date: Sapienza Università di Roma, Rome, Italy — 14–18 September 2026
The workshop focuses on autonomous and AI-driven cybersecurity systems, agentic security, intelligent threat detection, adaptive defense, cyber resilience, autonomous security operations, and trustworthy AI for cybersecurity.
Official CFP
Special Collection
Data-privacy-preserving federated learning in cyber-physical systems
Journal: Scientific Reports (Nature Portfolio)
Submission Deadline: 27 February 2027
This collection welcomes original research on privacy-preserving federated learning frameworks for cyber-physical systems, critical infrastructure, industrial automation, smart transportation, industrial IoT, and secure distributed AI.
View Collection

Contact Us

Interested in collaboration, joining our team, or learning more about our research?

📍

Location

University Campus
Research Building, Room 301

🔬

Research Opportunities

We welcome graduate students, postdocs, and visiting researchers.