The use of artificial intelligence in public policy is moving beyond administrative automation and entering a new phase of standardized, data-driven decision-making. The Electronics and Telecommunications Research Institute (ETRI) said its AI-based policy intelligence standard for supporting public decision-making has received final approval from the International Telecommunication Union Telecommunication Standardization Sector (ITU-T), establishing an international technical framework for digital government.
The significance of the standard is not that AI will replace policy judgment. Rather, it lies in defining common rules for how data scattered across public institutions and the private sector should be collected, managed, analyzed and applied to policy decisions. In other words, the competitiveness of AI-driven administration will depend not only on algorithmic performance, but also on data connectivity and trust architecture.
ETRI said the standard, titled “Data Management Interface for Public Decision-Making Framework Based on Intelligent Edge Computing” (ITU-T Q.5036), was formally adopted as an international recommendation at the ITU-T Study Group 11 meeting on protocols and test specifications, held in Geneva from March 3 to 11. The standard defines the data management structure, interfaces and data exchange protocols needed to implement AI-based policy intelligence.
Until now, public data has often been difficult to integrate because government agencies have used different management systems and security policies. In policy areas such as disaster response, economic planning, welfare, transportation and the environment, multiple ministries and private data sources are often required at the same time, yet the data frequently remains disconnected. The new standard is meaningful because it offers an internationally unified framework for reducing this fragmentation through common data management interfaces.
Policy intelligence refers to technologies that combine public data and AI to analyze policy effects, compare alternatives and support decision-making. For example, it can be used to assess the economic impact of fiscal spending, forecast the effects of policy measures, or evaluate the costs and risks of changes in social systems. While conventional policymaking has relied heavily on expert judgment and historical statistics, policy intelligence expands the basis for decision-making by combining real-time data with AI analysis.
A key feature of the new standard is its foundation in intelligent edge computing. Rather than sending all data to a central server for processing, the framework allows data to be analyzed and processed in real time at the edge, including on-site or distributed devices. This makes it particularly relevant for policy areas that require fast responses, such as disaster management, traffic congestion control, environmental monitoring and regional economic analysis.
One of the biggest challenges in deploying policy AI is data reliability. Even the most advanced AI model cannot produce policy-ready analysis if the input data is incomplete or if standards differ across institutions. This is why the standard also defines data pipelines, workflows and data governance. Only when the full process from data collection and storage to analysis and use is standardized can AI-based policy decision-making become repeatable, verifiable and reliable.
Another important issue is the connection between public and private data. Digital government cannot be built solely on internal administrative data. It requires secure links with a wide range of private-sector data, including information related to economic activity, industrial trends, mobility, consumption, the environment and finance. The new standard presents interfaces that enable distributed public and private data to be connected safely and used across institutions, creating a foundation for future policy intelligence services.
The adoption of an international standard also carries significance beyond technology. If public AI systems are developed separately by individual countries or agencies, interoperability will remain limited. By contrast, an international standard can support cross-border policy data linkage, joint analysis and broader digital government cooperation. ETRI’s role in proposing the standard and leading it through final approval shows that South Korea is taking part in setting the rules for public AI infrastructure.
Lee Yeon-hee, head of ETRI’s Fiscal and Economic Policy Intelligence Research Center, said the approval demonstrates that Korea’s AI policy intelligence technology for public decision-making has been recognized at the global level. Yoo Tae-wan, principal researcher at ETRI, also said AI- and data-based policy decision-making technologies are core components of digital government and are expected to contribute to public service innovation and improved policy efficiency.
The achievement was made through the “AI and Data-Based Fiscal and Economic Digital Twin Platform Development” project, supported by the Ministry of Science and ICT and the Institute of Information & Communications Technology Planning & Evaluation, as well as ETRI’s internal project on “High-Reliability Policy Intelligence Technology Based on AI and Big Data.” As AI adoption expands in public policy, the central issue is shifting from whether governments use AI to how they operate data under trusted rules and governance structures.
