Dr Koorosh Aslansefat
AI Safety, LLM Assurance and Dependable AI Researcher
Email:
Office:
Assistant Professor of Computer Science, University of Hull
Co-Lead, Dependable Intelligent Systems Research Centre
Deputy Lead, Centre for Responsible AI
About:
Dr Koorosh Aslansefat develops methods for evaluating, explaining and assuring artificial intelligence systems before deployment and during operation. His research focuses on AI safety, LLM and agentic AI assurance, explainable AI, runtime monitoring, failure propagation and probabilistic safety modelling for safety-critical and high-consequence applications.
Biography
Dr Koorosh Aslansefat is an Assistant Professor of Computer Science at the University of Hull, Co-Lead of the Dependable Intelligent Systems Research Centre and Deputy Lead of the Centre for Responsible AI.
His research investigates how artificial intelligence systems can be evaluated, monitored and assured before deployment and during operation. His work combines AI safety, explainability, runtime monitoring, uncertainty evaluation, probabilistic modelling and system safety engineering. His current research has a particular focus on large language models, agentic AI, autonomous systems and AI-enabled engineering systems.
He has led or contributed to projects funded by Innovate UK, UKRI, EPSRC, the European Commission, the Alan Turing Institute, the IET and industrial partners. His applied research spans healthcare, manufacturing, offshore renewable energy, robotics, transport, planning law and professional decision support.
Dr Aslansefat has developed and contributed to research methods and software including SafeML for runtime ML monitoring, SMILE and XWhy for model-agnostic explainability, SafeLLM for domain-specific LLM monitoring, SafeDrones for runtime reliability assessment and HDFT for hierarchical dynamic fault-tree analysis.
He received the Alan Turing Institute Postdoctoral Enrichment Award for his work on SafeML and the IET Leslie H. Paddle Award for outstanding research. He also contributes to postgraduate supervision, peer review, editorial activity, research funding assessment, international collaboration and AI standardisation.
Research Interests
AI safety and assurance Dependable and trustworthy AI Responsible AI Explainable AI Runtime monitoring and uncertainty evaluation LLM and agentic AI assurance Autonomous systems and safe, secure robotics Reliability and safety engineering Probabilistic modelling, especially Markov models Data-driven fault detection, diagnosis and prognosis Optimisation of AI methods and evolutionary algorithms
At a Glance
Figures as of September 2026. Latest citation metrics: Google Scholar. |
Recent PapersHis ten most recent journal and conference papers, newest first.
Full publication lists: Google Scholar · Research Gate · ORCID · Web of Science |
Funded ProjectsHis funded research portfolio spans responsible AI, trustworthy LLMs, safety-critical systems, robotics, healthcare AI, digital twins, EdgeAI, and offshore renewable energy.
Led as PI or Academic Lead
Co-Investigator
Awards and Fellowships
Earlier Funded Research
Funding Review Activity
|
Postgraduate StudentsHe supervises and mentors postgraduate researchers working on trustworthy AI, responsible LLMs, safety-critical ML, fairness, explainability, digital twins, EdgeAI, generative AI, and dependable autonomous systems. Current First Supervision
Current Co-Supervision
Completed or Near-Completion
|
Honours and ServiceSafeML, his method for runtime safety monitoring of machine learning classifiers, is recommended in DIN SPEC 92005, the German specification on uncertainty quantification in machine learning. Awards and Recognition
Editorial Roles
Conference Organisation
Selected Invited Talks
Teaching
|
Academic Background and Research Profile
2023 – Present
2021 – 2023
2018 – 2023
Assistant Professor in Computer Science, University of Hull. Teaching Responsible AI, Applied AI, Real-Time Dependable Systems and Safety-Critical Systems. Co-Lead of the Dependable Intelligent Systems Research Centre and Deputy Lead of the Centre for Responsible AI. Fellow of Advance HE.
Research Associate in the EU H2020 SESAME project (Secure and Safe Multi-Robot Systems), Department of Computer Science and Technology, University of Hull. Research on runtime safety monitoring, multi-robot dependability and safe AI.
PhD in Dependability Modelling and Assessment, University of Hull. Research focused on AI safety monitoring, dynamic fault trees, SafeML and probabilistic safety assessment for machine learning systems. Awarded the Alan Turing Institute Postdoctoral Enrichment Award and the IET Leslie H. Paddle Award.
Let's Connect
Publications
Useful URLs
Related DEIS Research
