Artificial Intelligence (AI) is widely recognised as one of the most transformative technologies of our time. From healthcare and financial services to manufacturing and logistics, organisations are exploring how AI can improve productivity, reduce costs, and enhance decision making. Data from the USA indicates benefits to businesses that adopt AI, and many are considering where the UK stands. Yet while much attention can be focused on the technical capabilities of AI, an important consideration is the interaction of the economic and the technical case.
A critical question emerges: What conditions are needed for firms to successfully adopt AI at scale?
A new research project led by Professor Anne Green at the City-Region Economic Development Institute (City-REDI), working with Lead Researcher Ian Taylor, aims to answer this question within one of the UK’s most strategically important sectors: defence.
The project, ‘AI Investment in the Defence Industry: Prospects for Fine-Tune Training and Data Sharing’, examines a challenge that sits at the heart of AI adoption. While powerful foundation models are now widely available, organisations often need to “fine-tune” these systems using their own specialist data before they can deliver meaningful productivity improvements. In many industries, however, the required data is proprietary, commercially sensitive, or difficult to access.
The defence sector faces these challenges more acutely than most. Defence manufacturers and suppliers operate within complex supply chains, handle highly sensitive information, and must comply with strict security requirements. At the same time, governments around the world are seeking to increase defence production capabilities in response to a rapidly changing geopolitical landscape. Productivity improvements within the sector have therefore become both an economic and national security priority, with potential implications for government budgets
The central question for the research is: Does restricted access to data limit the case for AI adoption in defence? If firms cannot access sufficient training data, many potentially valuable AI applications may never become commercially viable. In answering this question the research team is seeking to understand what other factors are constraining AI adoption in defence, whether these factors are separate to or linked to the access to data.
Crucially the research provokes the question of whether mechanisms can be found that enable secure data sharing across organisations to realise the potential benefits AI adoption more widely across the sector. The implications of enhancing the productivity in the UK defence industry are multiple for society. National and security could be enhanced, government budgets could go further and defence exports could become more competitive. Through augmentation the skills of the labour force could be bolstered, protecting regional economic activity. Many of these considerations for AI adoption go beyond the technical and into the economic and behavioural.
To investigate these issues, the researchers will speak to experts from industry, government, academia and policy to assess the current state of AI adoption, identify barriers to implementation, and explore future scenarios for collaboration and innovation. The ambition is for the findings to help inform strategic investment decisions by defence firms and policymakers. The research will also contribute to wider debates about innovation, industrial strategy, and the future competitiveness of the UK’s defence sector.
Lead Researcher Ian Taylor notes:
‘AI technologies are technically complex but some of the questions around adoption are far more than simply technical challenges. One of my core interest areas is collaboration and at the heart of the AI adoption question is the behaviour of people and organisations. The peculiar conditions of the defence industry accentuate the considerations that exist across the wider economy. Collaboration between the actors in the defence ecosystem may hold the potential to address some of these issues and this research is striving to shed light on how we might move towards this.’
At a time when both economic resilience and national security are increasingly dependent on technological innovation, understanding how AI can be deployed effectively and responsibly in defence has never been more important. This research will help chart a path towards a more productive, competitive, and innovative defence industrial base for the UK.
This blog was written by Anne Green, Professor of Regional Economic Development and Co-Director, and Ian Taylor, Associate, City-REDI, University of Birmingham.
Disclaimer:
The views expressed in this analysis post are those of the authors and not necessarily those of City-REDI or the University of Birmingham.