UK Launches TinyML Research Network to Advance Low-Cost, Low-Energy AI

TL;DR:

• The TinyML UK Network, funded by UKRI through the Engineering and Physical Sciences Research Council, brings together AI, electronics, and embedded systems researchers to advance AI that runs on small, low-power devices.
• Led by Nottingham Trent University with the University of Southampton and Imperial College London as co-leads, the network aims to build UK capability in decentralised AI.
• Practical applications already exist — from livestock health monitors to personal safety devices — positioning TinyML as a counterpoint to the dominant trend of ever-larger AI models.

A new UK-wide research network has been established to coordinate work on TinyML — an approach to artificial intelligence that enables machine learning models to run directly on small, low-power devices rather than relying on cloud servers.

The TinyML UK Network, led by Nottingham Trent University with the University of Southampton and Imperial College London as co-leads, is funded by UK Research and Innovation through the Engineering and Physical Sciences Research Council. It will connect researchers across AI, hardware engineering, and embedded systems with industry partners.

The Case for Smaller AI

While the AI industry’s focus has been overwhelmingly on scaling up — building larger models requiring more compute power and more energy — TinyML takes the opposite approach. By running specialised, adaptive AI models locally on sensors, wearables, and embedded systems, these devices can process data in real time without needing a cloud connection, keep sensitive information where it originates, and operate on minimal power.

The practical benefits are already visible. Livestock-monitoring devices can detect behavioural changes that indicate health issues, while personal safety devices identify abnormal motion or sound patterns on-device without recording or transmitting audio.

Why It Matters for the UK

Professor Eiman Kanjo, the network’s lead and professor of pervasive sensing and TinyML at Nottingham Trent, said AI adoption is accelerating “alongside concerns over energy consumption, infrastructure cost, resilience, privacy and sustainability.”

Those concerns have particular resonance in the UK, where energy costs for AI infrastructure are rising, data sovereignty questions persist post-Brexit, and rural connectivity gaps mean cloud-dependent AI simply does not work everywhere. TinyML’s ability to function without reliable internet makes it especially relevant for applications in agriculture, environmental monitoring, and remote healthcare — sectors where the UK has both strong research capability and pressing real-world needs.

Looking Forward

The network plans to run training programmes, competitions, and events for students, researchers, and SMEs, while building international links with global TinyML leaders. Its longer-term goal is to produce a UK roadmap for TinyML research and skills development. At a time when much of the AI conversation centres on compute-hungry foundation models, this network represents a bet that the future of practical AI may be small, cheap, and local.