Dr Petar Radanliev
Professional Masters Programme Project Supervisor, Department of Computer Science
About
Dr Petar Radanliev researches offensive security for agentic AI and autonomous multi-agent systems. He develops automated exploit pipelines and code-level red-teaming frameworks combining algorithmic fuzzing, prompt injection, tool-invocation abuse, model inversion and data poisoning. His work targets attack surfaces and coordination failures across Agent2Agent (A2A) networks, Model Context Protocol (MCP) integrations and quantum-resistant infrastructures, converting adversarial research into reproducible penetration-testing simulations for vulnerability discovery and secure deployment.
Expertise
- Algorithmic red teaming
- Agentic AI & multi-agent security
- Offensive cyber operations
- Adversarial machine learning
- Automated exploit pipelines
- Generative AI penetration testing
- Agent2Agent protocol vulnerabilities
- AI software supply chain attacks
- Post-quantum cryptographic break scenarios
- Distributed & blockchain system vulnerabilities
- AI-driven malware analysis.
Selected publications
- See a full listing of research papers and books here
- AI Security, The Most Dangerous Cyber-Attacks on Artificial Intelligence (Springer Cham, Series Title: Cognitive Technologies, 2026)
- Post-Quantum Security for AI: Resilient Digital Security in the Age of Artificial General Intelligence and Technological Singularity (Pearson, Addison-Wesley Professional; 1st edition, 2025)
- AI-Powered Digital Cyber Resilience (Pearson, Addison-Wesley Professional; 1st edition, 2025)
- The Rise of AI Agents: Integrating AI, Blockchain Technologies and Quantum Computing (Pearson, Addison-Wesley Professional; 1st edition, 2024)
- Beyond the Algorithm: AI, Security, Privacy, and Ethics (Pearson, Addison-Wesley Professional, 2024)
Media experience
Dr Petar Radanliev has experience providing commentary on state-sponsored cyber operations and autonomous AI weaponisation, translating complex multi-agent vulnerabilities into national security insights as seen in his analysis for the BBC on agentic communication risks and The Times on large-scale espionage profiles.
Watch online
Dr Petar Radanliev has three YouTube channels:
- YouTube Channel on Airdrop Hunting for those interested in learning about and exploring new and emerging blockchain
- YouTube Channel on AI Security for those interested in learning about and exploring new and emerging Generative AI projects
- Oxford Research Channel for those interested in learning about research methods, referencing software and all matters related to academic writing and academic life
Languages
Welsh, Macedonian, Serbo-Croat, Greek, Bulgarian