Executive Summary
Artificial intelligence (AI) is emerging as a core driver reshaping the
global economy, industry, and society. While advancements in hardware—
such as computing power and data storage capacity—have received
considerable attention, the role of software in making AI innovation
practically feasible has been relatively underexplored.
However, software extends far beyond the mere implementation of
complex algorithms; it constitutes the foundation that determines the
efficiency, scalability, and accessibility of AI technologies. Advanced software
frameworks and programming languages enable developers and researchers
to rapidly design, refine, and deploy sophisticated AI models. Contemporary
AI systems rely on the large-scale collection and processing of data, as well
as intensive computation leveraging high-performance computing resources.
AI models function as the core engines of diverse applications, forming the
backbone of a wide range of AI services and products. In this sense,
software enables the creation of AI, while AI, in turn, amplifies the value of
software—illustrating a process of mutual and reinforcing growth.
The fundamental importance of software carries clear policy
implications. Investment in hardware infrastructure or data alone is
insufficient; it must be accompanied by strategic support for areas such as
open-source software ecosystems, standardization and interoperability, and
robust software security frameworks. Furthermore, policies that promote
integrated education and workforce development in software and AI,
technological innovation, and international cooperation will serve as critical
instruments for sustaining long-term AI competitiveness and fostering
inclusive growth. Ultimately, AI policy should be designed and implemented
based on the recognition that the competitiveness of the software
ecosystem directly translates into the competitiveness of AI.
This report examines the roles and illustrative cases of software
underpinning each stage of the AI development lifecycle and explores
software policy directions aimed at strengthening the overall AI ecosystem.