ABSTRACT
On the global market, several nations are racing to achieve a global innovation advantage in AI as it is understood that AI is a foundational technology that can boost competitiveness, increase productivity, protect national security, and help solve societal challenges. Comparing China, the European Union, and the United States in terms of their relative standing in the AI economy by examining six categories of metrics: talent, research, development, adoption, data, and hardware, the United States leads in absolute terms, with China coming in second, and the European Union lags further behind. This order could change in the coming years depending on a range of policy actions that can propel each nation or region to improve its AI capabilities [11, 14]. AI technology developments significantly impact electronic and component systems, semiconductor design, and production, as the amount of data processed and stored by AI applications continues to increase. Semiconductor architectural improvements are needed to address data use in AI-integrated circuits, and improvements in semiconductor design for AI are requested to enhance overall performance, speed, memory capacity, with increased energy efficiency. In this context, major initiatives have started globally to address the development of the semiconductor industry, such as the European Chips Act, which aims to bolster Europe’s competitiveness, resilience, and help achieve both the digital and green transition [12]. In this context, edge AI represents a paradigm shift in deploying and utilising AI technologies, marking a transformative evolution from centralised data processing systems to decentralised, edge-oriented solutions. The edge AI deployments are characterised by massive scale and heterogeneity. A single system may comprise many devices with diverse hardware and software stacks, creating significant challenges for deployment, interoperability, and management. These devices are often deployed in uncontrolled environments, making them vulnerable to physical tampering and environmental hazards, which introduces a class of security threats not typically considered in secure data centres. This tightly coupled system of trade-offs, forced by a resource-scarce environment, makes a holistic systems engineering approach a must.
