Policy

MOEF AI Public Institution Strategy: 2025 AI Policy Roadmap

Background and Significance of MOEF's AI Public Institution Strategy

The Ministry of Economy and Finance (MOEF) has declared 2025 as the "Year of Public AI Transformation" and established a pan-government AI public institution strategy. This goes beyond simple digitalization expansion to pursue structural transformation that embeds AI across public service decision-making, administrative processing, and citizen convenience. The defining feature of this strategy is that MOEF leads it, leveraging its budget allocation authority and public institution management evaluation powers as accelerators for AI transformation.

In the global competitive context, this strategy carries significant meaning. The United States has mandated federal government AI adoption through executive orders, and the EU is governing high-risk AI use in the public sector through the AI Act. Within this landscape, Korea's establishment of a leading public AI model serves not only domestic administrative efficiency but can also become the foundation for exporting digital public governance to developing nations including Bangladesh.

340+
Target Public Institutions
Phase 1 (2025)
KRW 2.3T
Budget Investment
2025-2027 Three-Year Total
80%
AI Adoption Target
Major Institutions by 2027
127
Priority AI Projects
MOEF-Selected
40%
Processing Time Reduction
Expected Average Effect
120,000
Civil Servant AI Training
2025-2026 Target
35
Public-Private Partnerships
AI Consortiums
5 Projects
Bangladesh Cooperation
Including Digital ODA

2025 AI Policy Roadmap: Four-Phase Transformation Plan

MOEF's AI public institution strategy follows a four-phase roadmap: "Foundation Building, Expansion, Advancement, and Self-Sufficiency." Each phase has clear KPIs and budget execution criteria, with non-achievement reflected in public institution management evaluations as an enforcement mechanism. The year 2025 marks the foundation-building phase, with concentrated focus on public AI platform infrastructure construction and launching priority pilot projects.

MOEF AI Public Institution Transformation Roadmap (2025-2028)
2025: Foundation
Build G-AI platform, launch 127 pilot projects
2026: Expansion
AI integration across 340+ institutions, evaluation linkage
2027: Advancement
Full-scale predictive administration and AI decision support
2028: Self-Sufficiency
Self-sustaining AI ecosystem, global export model
01
Phase 1 (2025): Public AI Infrastructure Foundation
KRW 120 billion is jointly invested by MOEF, the Ministry of the Interior and Safety, and the Ministry of Science and ICT to build the Government AI Platform (G-AI). Key initiatives include constructing a public data AI learning hub, mandating basic AI training for civil servants (120,000 target), and establishing public institution AI adoption guidelines. MOEF introduces a new "AI-First Budget" principle with freshly created AI project budgeting standards.
02
Phase 2 (2026): AI Expansion Linked to Management Evaluation
AI adoption status across 340+ public institutions is incorporated into MOEF management evaluation indicators. AI utilization levels (adoption rate, usage volume, efficiency gains) determine differentiated evaluation scores, with lowest-performing institutions facing improvement orders and budget reductions. An inter-institutional AI best practice sharing platform is operated, and establishment of Chief AI Officer (CAIO) positions is mandated for each institution.
03
Phase 3 (2027): Predictive Administration and AI Decision Support
Going beyond simple automation, AI-based predictive administration is fully deployed across policy outcome forecasting, budget efficiency analysis, and citizen petition pattern prediction. MOEF introduces an "AI Fiscal Analysis System" to begin AI-supported national fiscal management. Welfare, tax, and employment agencies are required to provide AI-personalized services. A public AI audit framework is also established to ensure predictive administration reliability.
04
Phase 4 (2028): Self-Sufficiency and Global Model Export
Korea's public AI transformation model is packaged as "Public AI ODA" for export to developing nations in ASEAN, the Middle East, and South Asia. Target countries with high digital transformation demand — Bangladesh, Vietnam, Indonesia — receive administrative AI system construction support, civil servant AI training cooperation, and AI governance institutional design assistance. This positions Korea's new mechanism for expanding global influence.

AI Public Sector Budget Investment Structure and Key Projects

The total scale of MOEF's three-year (2025-2027) AI public institution investment plan is KRW 2.3 trillion. This is not a single-ministry budget but a pan-ministry AI budget pool coordinated by MOEF, consolidating AI-related budgets from key ministries including the Interior Ministry, Science Ministry, Welfare Ministry, and Employment Ministry. The budget is allocated across four domains: infrastructure, services, human resources, and global cooperation.

AI Infrastructure (KRW 720B)
G-AI PlatformKRW 250B
Public Data HubKRW 180B
Security & GovernanceKRW 120B
AI Computing InfraKRW 170B
AI Public Services (KRW 960B)
Welfare & Health AIKRW 320B
Tax & Fiscal AIKRW 210B
Employment & Education AIKRW 240B
Public Safety AIKRW 190B
AI Workforce & Systems (KRW 380B)
Civil Servant AI TrainingKRW 150B
CAIO DevelopmentKRW 60B
AI Regulation & AuditKRW 90B
AI Ethics & Legal FrameworkKRW 80B
Global Cooperation (KRW 250B)
AI ODA ProgramsKRW 120B
International AI StandardsKRW 50B
Public-Private ConsortiumsKRW 50B
Bangladesh & ASEANKRW 30B
MOEF Priority 127 AI Projects by Sector
SectorProjects2025 BudgetCore ObjectiveExpected Impact
Welfare & Health28KRW 320BAI Welfare Gap Prediction90% Reduction in Welfare Blind Spots
Tax & Customs22KRW 210BAI Tax Evasion DetectionKRW 2T Additional Revenue
Employment & Jobs18KRW 240BAI Job Matching & Training15%p Employment Rate Improvement
Education & Training14KRW 110BAI Personalized Learning30% Education Gap Reduction
Public Safety & Disaster19KRW 190BAI Crime Prediction & Emergency Response50% Response Time Reduction
Citizen Services & Admin26KRW 150BAI Automated Petition Processing40% Processing Time Reduction
Korea AI and Semiconductor Investment Opportunity AnalysisExplore the private investment opportunities linked to MOEF\'s AI policy roadmap and Bangladesh investor perspectives

Detailed Analysis of Key AI Projects by Sector

The most notable projects in MOEF's AI policy roadmap are the "Welfare AI Integrated Platform," "NTS AI Tax Audit System," and "Employment AI Personalized Service." These three projects are expected to become flagship examples of Korea's public AI transformation in terms of budget scale, social impact, and technological sophistication.

01
Welfare AI Integrated Platform: Zero Welfare Blind Spots
Jointly promoted by the Ministry of Health and Welfare and MOEF, this platform integrates health insurance, national pension, basic livelihood security, and disability welfare data to enable AI to proactively identify vulnerable populations in welfare blind spots. Currently requiring an average of 6 months for identification, AI enables real-time prediction. After 2025 pilots, nationwide expansion is planned for 2026 with a KRW 320 billion budget.
02
NTS AI Tax Audit System: Tax Evasion Detection Innovation
The National Tax Service launches its "AI Tax Audit Support System (Tax AI)" in the first half of 2025. AI integrates and analyzes financial transaction data, real estate registrations, and corporate accounting ledgers to automatically calculate tax evasion risk scores and prioritize audit targets. MOEF targets over KRW 2 trillion in additional annual tax revenue. Bangladesh's National Board of Revenue (NBR) has shown strong interest in this system and is exploring technology cooperation possibilities.
03
Employment AI Personalized Service: Resolving Job Mismatches
The Ministry of Employment and Labor's "AI Employment Matching Platform" uses AI to analyze job seekers' competencies, careers, and preferences to recommend optimal job placements and design customized vocational training programs. Converting the existing WorkNet's simple search functionality to AI-driven recommendations, the goal is to raise the employment success rate from the current 32% to 47% or higher.
04
Digital Citizen Services AI Automation: 24/7 Administrative Services
The Ministry of the Interior and Safety fully launches its "AI Citizen Service Automation Platform" in July 2025. AI chatbots and automated processing engines are deployed across existing channels such as Government24 and Minwon24, targeting 60% of all citizen requests processed by AI. AI translation is applied to foreign-language requests (including English, Vietnamese, and Bangla) to significantly improve administrative access for multicultural families including Bangladeshi foreign workers.

Bangladesh AI Cooperation: Implications and Opportunities

Korea's public AI policy roadmap offers Bangladesh implications across three dimensions. First, ODA channels open for Bangladeshi government agencies to directly adopt and cooperate on Korea's public AI models. Second, expanded Korean public institution AI demand creates export opportunities for Bangladeshi IT services and data firms. Third, opportunities increase for Korean AI companies to develop business and cooperation in the Bangladesh market.

Korea-Bangladesh AI Cooperation: Potential Areas and Progress
Cooperation AreaKorean SolutionBangladesh DemandCooperation ModeTimeline
Tax & Customs AINTS AI Tax Evasion DetectionNBR Revenue Expansion GoalODA + Tech Transfer2026-2027
Welfare AIWelfare Blind Spot DetectionBSSNP Beneficiary ManagementPilot Cooperation2026
Employment AIAI Job MatchingYouth Unemployment ResolutionPlatform Export2027
Citizen Service AI24/7 AI Citizen ServicesDigital Government ServicesJoint Development2026-2028
Disaster Safety AIDisaster Prediction AIFlood & Cyclone ResponseDisaster Prevention ODA2025-2026
Education AIAI Personalized LearningEducation DigitalizationEdTech Export2027-2028

Of the KRW 120 billion in AI ODA budget allocated by MOEF for 2025-2027, approximately 20% is earmarked for the South Asian region, meaning roughly KRW 30 billion is directed toward digital cooperation with Bangladesh and other South Asian countries. Bangladesh's Planning Commission and ICT Division are already in discussions with Korea's KOICA on AI administrative capacity-building cooperation, with an MOU expected in the first half of 2026.

Opportunities for Bangladesh Government Agencies
NBR Tax AIKorean Tech Transfer Discussions
Welfare Blind Spot DetectionBSSNP AI Pilot Introduction
ICT DivisionG-AI Platform Application Discussions
BIDA Investment BoardAI Investment Review Automation
Opportunities for Bangladesh Private Sector
IT Service CompaniesKorean Public AI Development Subcontracting
Data PreprocessingBangladesh Data Provision
AI Education StartupsKorean AI Curriculum Adoption
Fintech CompaniesKorean Tax AI Cooperation Model
Korea-Bangladesh Joint Projects
Disaster AI ODAJoint Flood Prediction Research
AI Talent ExchangeCivil Servant AI Training Cooperation
Public AI StandardsBilateral AI Governance Alignment
Smart CityDhaka AI Administration Pilot
2025 Economic Policy Comprehensive Analysis: MOEF Policy Direction and Bangladesh ImpactReview the broader MOEF 2025 economic policy direction and how AI policy affects Bangladesh trade and investment

Public AI Governance and Risk Management Framework

For MOEF's AI policy roadmap to succeed, governance and risk management are as important as technology adoption itself. Errors, biases, and security vulnerabilities in public AI services can directly impact citizens' lives, and AI malfunctions in sensitive areas like welfare, taxation, and employment can cause serious social repercussions. Accordingly, MOEF is building a public AI audit and ethics framework in parallel with AI deployment.

01
Public AI Audit Framework: Mandatory Independent Verification
High-risk AI systems (welfare benefit determinations, tax audit target selection, disaster alerts) are required to undergo independent algorithmic audits at least once annually. Audit results are reported to the National Assembly and made public. Upon discovery of bias or errors, immediate service suspension and improvement processes are activated. The Korea Internet & Security Agency (KISA) is designated as the public AI audit certification body.
02
AI Ethics Guidelines: Seven Principles for Public AI
MOEF establishes the "Seven Principles of Public AI Ethics" applicable to all public AI services: (1) Respect for Human Dignity, (2) Non-Discrimination and Fairness, (3) Transparency and Explainability, (4) Personal Data Protection, (5) Safety and Reliability, (6) Citizen Participation and Oversight, (7) Accountability. All public AI systems must pass pre-deployment compliance reviews. Compatibility with the EU AI Act is simultaneously considered.
03
AI Cybersecurity: Protecting Public AI Infrastructure
Public AI platforms face new cyber threats including hacking, data leaks, and AI model poisoning. The National Intelligence Service and KISA jointly establish "Public AI Cybersecurity Guidelines" and mandate year-round penetration testing (Red Team exercises) for the G-AI platform. A dedicated security budget of KRW 90 billion is allocated.
04
Citizen Participation and AI Acceptance Enhancement
To increase public acceptance of AI services, an "AI Public Hearing" system is introduced. Mandatory online and offline hearings are required before major AI service launches, and citizen panels participate in AI service design through a "Participatory AI" pilot. An "AI Decision Appeal" system is also introduced to allow citizens to challenge AI-made decisions, protecting citizens' rights.

MOEF's AI public institution strategy and 2025 policy roadmap represent the most concrete and fiscally backed blueprint for Korea's leap to becoming an "AI-Leading Government." Backed by KRW 2.3 trillion in budget, 127 priority projects, and management evaluation enforcement mechanisms, this strategy will fundamentally transform Korea's public sector by 2028. Bangladesh should actively pursue its role in this process as a technology cooperation beneficiary, a business partner for Korean AI companies, and even a co-design partner for public AI governance. Building proactive partnerships through ODA cooperation channels will be the key variable determining the pace of Bangladesh's digital transformation.

AI PolicyMOEFPublic AIDigital GovernmentAI RoadmapPublic Institution AIBangladesh AI Cooperation
MOEF AI Public Institution Strategy: 2025 AI Policy Roadmap | Dhaka Trade Portal