Research Blog

A research team from the University of Hull has secured third place in the Cambridge Edge AI Innovation for Sustainability Challenge 2026 with SustainGRF, a project exploring how smartwatch data and federated learning could support research into the relationship between everyday physical activity and bone health.

It is evident that in the AI era, security cannot be bolted on at the end. It must be a core feature from the first design decision. Let’s define what a "secure product" is, why AI makes trustworthy computing both harder and more necessary, and a practical checklist product teams can use.

Anthropic’s refusal to provide pre-release access exposes the risks of relying on privileged model access for frontier AI assessment. Summary Anthropic’s decision not to provide Claude Mythos 5.1 to the UK AI Security Institute (AISI) for pre-release testing has been discussed mainly as a problem of voluntary cooperation, regulation and geopolitics. It is also a technical warning. Some of the strongest methods for understanding and controlling advanced models depend on priv

Multi-agent AI systems are rapidly moving from research prototypes into enterprise workflows. Instead of relying on a single Large Language Model (LLM), these systems use multiple specialised AI agents to plan tasks, retrieve information, evaluate evidence and generate final answers. This architecture is increasingly attractive for enterprise AI, Agentic RAG, automated decision support, AI assistants and business-critical workflows.

DEIS research presented at Computing Conference 2026 explores how generative AI can create formal safety artefacts, including fault trees, while keeping engineering evidence transparent and open to verification.

Dr Zhibao Mian presented DEIS research on generative AI for formal safety artefacts during the Applied AI session at Computing Conference 2026.

Artificial intelligence (AI) systems are becoming increasingly integrated into critical areas such as healthcare, transportation, finance, and manufacturing. As these systems take on more complex and impactful roles, ensuring their safety and reliability becomes essential. SafeML, a framework designed for safety monitoring of AI, offers a promising approach to address this challenge. This post explores the concept of SafeML, its key features, methodologies, and applications a

Professor Yiannis Papadopoulos discusses how safe AI, optimisation and large language models could improve fleet service, maintenance and repair.