AI Security
Adversarial AI, model security, poisoning, prompt and tool misuse, assurance and secure deployment.
Research · Innovation · Education · Impact
Dr Zia Ush Shamszaman
Cybersecurity · AI Security · Trustworthy AI · Multi-Agent Systems
I study how intelligent and increasingly autonomous systems can remain secure, trustworthy and strategically resilient in adversarial environments.
Lecturer in Cybersecurity Education
Department of Informatics · King's College London
Dr Zia Ush Shamszaman
Cybersecurity & Trustworthy AI
King's College London
Senior Member, IEEE
Fellow, Advance HE
EPSRC Peer Review College
Industry & Knowledge Exchange
Research question
How can we design intelligent systems in which secure and trustworthy behaviour remains strategically sustainable, even when agents face adversarial incentives?
Research programme
From securing AI models to governing adaptive AI ecosystems
Agentic & Multi-Agent AI
Trust, cooperation, competition, coordinated pressure, strategic behaviour and security among interacting autonomous agents.
Evolutionary Game Theory
Adaptive attacker-defender dynamics, incentives, commitment, cooperation and security mechanisms that evolve over time.
Federated & Collaborative AI
Distributed learning under poisoning, backdoor, collusion and strategically unreliable clients.
AI Governance & Safety
Responsible deployment, accountability, assurance, media and regulatory incentives, and trustworthy AI adoption.
Cyber-Physical & High-Stakes AI
Healthcare AI, IoT/AIoT, secure connected systems and applied intelligent decision support.
Selected publications
Recent work shaping my current research direction
Strategic Commitments Shape Collective Cybersecurity under AI Inequality
Chaos, Solitons & Fractals · Evolutionary game theory, AI security, differential access to defence and strategic commitment.
Publication details →Co-evolutionary Dynamics of Attack and Defence in Cybersecurity
Knowledge-Based Systems · Adaptive attacker-defender populations, defence intensity and evolutionary stability.
Publication details →Can Media Act as a Soft Regulator of Safe AI Development?
Artificial Life Conference Proceedings · AI safety, media influence and evolutionary incentives for responsible development.
Publication details →Funded work
Research translated into projects and practice
CyberSafeAI North East
Responsible AI workforce development integrating AI safety, security and governance.
£299,666
CyberPathway
Cybersecurity skills development, widening participation and employer-facing professional education.
£143,000
OPERA
Collaborative cybersecurity research and innovation.
£60,000
CyberThemis
Cybersecurity and AI-focused collaborative research and innovation.
£60,000
Qnergy-Sec
Quantum-resilient security research addressing emerging cyber risk.
£30,000
SafeSCMS & iShield
Principal-Investigator-led cybersecurity research and innovation projects.
£61,000 combined
Education
Research-led teaching at scale
Cybersecurity, AI, ethics, distributed systems and postgraduate research, informed by curriculum leadership, online and international delivery, and large-cohort supervision.
Teaching & supervision →Academic contribution
Leadership, service and impact
Externally funded innovation, editorial leadership, peer review, professional service, industry engagement and public-facing impact.
Leadership → Service → Impact →Profile at a glance
Research, education and external engagement
My work connects cybersecurity, AI security, trustworthy AI and multi-agent systems with education, industry collaboration and applied research.
£900k+portfolio of funded projects and knowledge exchange activity
100+MSc research projects supervised
6current PhD researchers recorded in my portfolio
566Google Scholar citations in the uploaded profile record
Academic profile
Research that connects theory, systems and external value
Trustworthy and secure AI
Research spanning AI security, adversarial behaviour, multi-agent cooperation, evolutionary game theory, federated learning, explainability and responsible AI.
Research themes →Cybersecurity and AI education
Teaching and module leadership across King's College London, Teesside University, University of Galway and Hankuk University of Foreign Studies, including large-scale postgraduate research education.
Teaching profile →From funded research to practice
Projects and partnerships connect research with SMEs, professional communities, healthcare, workforce development and regional cyber ecosystems.
Impact and engagement →Latest insights
Research commentary & professional posts
New research commentary and professional posts will appear here automatically when published through the Insights editor.
Collaborate
Research collaborations, doctoral supervision, industry partnerships and invited talks
I welcome conversations around cybersecurity, trustworthy AI, AI security, agentic systems, game-theoretic security and responsible AI deployment.
Get in touch