Insight

Overview of AI in Scotland

Scotland’s journey in artificial intelligence (AI) is marked by its strategic ambition to position itself as a leader in ethical and trustworthy AI. This theme echoes across various discussions and analyses on the topic and composes a core principle of the Scottish AI Alliance (a partnership between The Data Lab and the Scottish Government tasked with the delivery of the vision outlined in Scotland’s AI Strategy). Scotland has already put in place a number of foundational steps towards both realising this vision of ethical and trustworthy AI and also capturing the economic benefit of AI including, for example, Scotland’s AI Strategy and the Scottish AI Alliance. However, wider programmes such as the Digital Scale Up Level Up have called for AI supply side development to accelerate. To achieve this, Scotland has had to take stock and understand what its baseline is: who are the businesses, what are their capabilities, how healthy is the support landscape, where are the gaps and, crucially, where are the opportunities. This article aims to provide a brief response to these questions based on work undertaken by us in recent years.

AI Industry-base and Capabilities

There are over 200 AI companies in Scotland (based on research conducted by Optimat[1]), showcasing a diverse array of capabilities from machine learning to sub-fields like natural language processing (NLP) and computer vision. There is a large concentration of companies in the central belt but also pockets of strength in other regions, such as Aberdeen and Dundee. Over half of the company-base develop AI products with the remainder made up of consultancy service providers (25%) and platform/infrastructure solution providers (17%). This is a strength for Scotland as products are scalable and have potential for widespread, international adoption. Moreover, Scotland provides the ideal conditions for inward investors to thrive, underpinned by a strong base of capability and a supportive research and academic network that is heralded worldwide, e.g., The Data Lab innovation centre, University of Edinburgh’s AI Accelerator, and the Data-Driven Innovation (DDI) network. Further information on the support landscape is provided below. The mix of company sizes indicates a nascent industry with most companies classified as micro or small.

Perhaps of greater interest is the breadth of applications and the markets targeted with the solutions created by the Scottish industry-base. Instead of targeting a single industry vertical, most companies seek to address the issues of several markets at once, providing substantial horizontal benefits. Key industries that are targeted, include healthcare, agriculture and land use, financial services/fintech, energy, creative industries, built environment, and transport.

The Support Network

The AI ecosystem in Scotland is supported by a robust network of universities, business accelerators, innovation centres, sector organisations, government agencies, and other stakeholders. More than 70 academic, research and support stakeholders provide support to the industry-base, from funding to research collaboration, start-up support to skills development, and much more. Initiatives include an AI Accelerator at the University of Edinburgh, innovation centres such as The Data Lab and the Centre for Sensing, Imaging and IoT (CENSIS), The National Robotarium at Heriot-Watt University, a variety of university research departments and groups (e.g., Herriot-Watt University’s Intelligent Systems Lab, St. Andrews University’s AI Research Group, University of Edinburgh’s Machine Learning Group, and University of Strathclyde’s Human Centric AI Research Group), and access to RD&I funding via Scottish Enterprise, Highlands and Islands Enterprise, and Innovate UK.

Challenges and Limitations

Scotland’s AI journey is not without its challenges. The early stage of understanding AI applications suggests a need for a deeper, more nuanced engagement with the technologies that underpin AI, especially beyond the dominant field of machine learning. The broader adoption and integration of AI subfields such as NLP and computer vision could diversify the technological base, enhancing the robustness and applicability of AI solutions across different sectors. There is also an increasing concentration of companies across the central belt, therefore work is required to showcase the benefits of AI across Scotland and create the conditions for companies to thrive in regions that are seemingly less affiliated with the technology.

Regulation is another barrier that can stifle innovation, but it also forms a critical framework that ensures the responsible development and deployment of AI technologies. This has been most acutely acknowledged by the EU AI Act[2], the ‚ ‘world’s first comprehensive AI law‚Äù, which aims to ensure better conditions for the development and use of AI. Regulations not only safeguard privacy and data integrity but enforce standards relating to trust and transparency in AI applications and ensure they are ‚ ‘safe, transparent, traceable, non-discriminatory and environmentally friendly‚Äù. The EU AI Act will apply to UK firms which utilise AI systems within the EU, make AI systems available on the EU market, or participate in any other activity regulated by the Act.

In contrast to the EU’s AI Act which enacts comprehensive legislation, the UK Government announced in a white paper[3] in 2023 that it is prioritising the establishment of guidelines for the development of and utilisation of AI, including enhancing the authority of existing regulatory bodies (e.g., FCA) to provide guidance and oversee the use of AI. Guidance for regulators would revolve around five cross-sectoral principles, as follows:

• Safety, security and robustness
• Appropriate transparency and explainability
• Fairness
• Accountability and governance
• Contestability and redress

Regardless of the evolving regulatory landscape, the adherence to strict ethical guidelines is essential, as it not only protects citizens but reinforces Scotland’s commitment to prioritising and leading the global conversation on ethical AI.

Strategic Opportunities

Strategic investments in education, infrastructure, and awareness campaigns are crucial for overcoming existing limitations like talent shortages and lack of knowledge or comfortability with the technology. Such investments would not only supply the necessary human capital but also create an environment where AI can thrive, supported by state-of-the-art facilities and a well-informed, AI-literate populace.

The call for enhanced governance and ethics in AI deployment aligns with global movements towards responsible AI, emphasising the need for frameworks that ensure AI’s development and use remain transparent, ethical, and inclusive. This approach not only aligns with Scotland’s vision for AI but places it at the forefront of the global conversation on AI ethics.

Furthermore, the Scottish AI Alliance is making significant strides in implementing the vision for AI in Scotland, ensuring the Scottish Government’s strategy is delivered in an open, transparent and collaborative way. This is best demonstrated at the annual Scottish AI Summit which brings together the sector’s great minds and innovators, as well as enthusiasts, to receive updates on strategic progress and showcase Scotland’s leading role in the development of trustworthy, ethical, and inclusive AI. Continuing to capitalise on this expanding community only serves to bolster Scotland’s position as a growing force in AI across the UK, Europe and further afield.

In conclusion, Scotland stands at a pivotal point in its AI journey. The nation’s concentrated efforts in strategic cities, supported by a robust ecosystem of academic and business accelerators, sets a solid foundation for growth. However, realising the full potential of AI technologies will require sustained investments in talent development, infrastructure, ethical governance and communications campaigns to break down barriers such as the trustworthiness of AI. By navigating these challenges strategically, Scotland can strengthen its position as a leader in the ethical and innovative use of AI.

*The EU AI Act was approved in March 2024 and will be fully applicable 24 months after its entry into force. Some aspects of the Act will be applicable sooner. Find out more here.