Energy Efficient Computing - from innovation to impact

What is neuromorphic computing?

Modern computer systems tend to employ computational architectures which inherently use a large amount of power and time to transport data to and from the processor. For large scale tasks, such as those employed to operate on large data sets for training AI, this causes lengthy compute times and often require the use of data centres. This additional computational power requires a significant amount of energy to run – the International Data Centre Authority stated in May 2026 that almost 6% of the UK’s energy consumption was from data centres alone – as well as the vast amount of cooling required to keep data centres operational.

The human brain on the other hand is naturally energy efficient, consuming less than 20 Watts to function, that’s about the same amount as a computer monitor on standby. Mimicking the function of the human brain, neuromorphic computing, also known as brain-inspired computing, comprises artificial electronic neurons and synapses framed into a computational architecture. This technology brings about advantages beyond energy efficiency too, such as increased computational speed, cognitive learning and adaptability, and the functionality to compute at the edge i.e. process data on chip. As a result, neuromorphic computing has the potential to disrupt the AI hardware market.

What are the potential applications?

While the early applications of neuromorphic computers were to understand the processes of the human brain, edge computing and edge AI are the most likely commercial applications. In the consumer and health sectors, the low power consumption combined with edge computing abilities means that neuromorphic chips could be used in wearables, and smart biosensor-based technologies. Similarly, this edge computing ability is particularly suited to applications which require real-time decision making without the need to be connected to a data centre, such as in autonomous vehicles. Other areas being explored include robotics, event-driven cameras and satellite technologies.

What are the challenges?

To enable development and deployment of this technology, expertise is needed across multiple scientific and engineering disciplines including neuroscience, computer science and materials. Like many technologies that are needed to be deployed at scale, development of cost-effective manufacturing processes and access to facilities is needed, as well the ability to integrate with other technologies – standardisation of aspects of technology development may help with this.

How is the UK addressing these challenges?

The UK Government’s AI Hardware plan, announced in June 2026 and totalling over £1.1 billion of investment, sets out a path to enable the UK to develop AI hardware at pace. This would strengthen existing capabilities and recognises the need for the development of novel compute technologies such as neuromorphic, photonic and quantum computing systems.

In December 2024, Innovate UK Business Connect brought together experts from across the UK neuromorphic computing sector for a day of networking and shared learning on topics including; bioinspired computing architectures; applications in robotics and autonomous vehicles; and neuromorphic chip integration and packaging. This event enabled cross-discipline discussions and networking, catalysing collaborations for neuromorphic computing development.

Since 2025, UK Research and Innovation (UKRI) has funded the following centres and networks to help deliver the UK’s ambition for future compute technologies:

• NeuMat – Funded by EPSRC, NeuMat is connecting academia and industry to drive the development of materials for neuromorphic computing and AI hardware
• NeuroWare – an Innovation and Knowledge centre, funded by Innovate UK & UKRI to support the development and translation of neuromorphic computing hardware by bringing together industry, academia and policy makers
• NeuroSYNC – the UK’s Multidisciplinary Centre for Neuromorphic Systems and Computing, funded by EPSRC, is aiming to advance the development of sustainable neuromorphic computing technologies and AI systems
• Aston-Hartree Neuromorphic Centre of Competence – a collaboration between Aston University and the Science and Technologies Facilities Council (STFC) to support of the development of technologies and their adoption in industry.

Are you interested in this topic?

Would you like to discuss innovation in the area, make connections or get further information? Then reach out to Innovate UK Business Connect colleagues  Amy Romaniuk, Knowledge Transfer Manager – Emerging and Enabling Technologies and Simon Yarwood Knowledge Transfer Manager – Industrial Technologies