South Korea's Ministry of Science and ICT and the Ministry of Finance and Economy held the "2026 Public Institutions AI Innovation Performance Competition" on Oct. 28 at Nurikum Square in Sangam-dong, Seoul.

The event presented results from the "2026 Public Institutions AI Innovation Challenge", a public-sector programme of the "Nationwide AI Competition" promoted since March.

In the AI Innovation Challenge, held for the first time this year, 316 tasks were submitted by 105 institutions from June to August. After expert review, a total of 30 tasks were selected: 2 top prizes, 8 excellence awards, 10 merit awards and 10 AI innovation awards.

The top prize in the AI service pilot and proof-of-concept (PoC) category went to Korea Western Power's "K-gas turbine, the country's first, driven by AI and inspected by robots". The project features AI analysis of power-generation facility data and links digital twins with AI agents and robots. It used "K-AI full-stack" technology that integrates the application of a domestically made neural processing unit (NPU) and AI models. The ministry said about 2.1 billion won in annual economic benefits are expected.

In the AI use-case category, Incheon International Airport Corp won the top prize. It shifted an aviation safety management system based on visual checks and manpower to 24-hour AI prediction and detection, so it can identify risk factors in aircraft takeoff and landing in advance.

AI use cases closely linked to people's daily lives were also unveiled. Korea Electric Power Corp showcased a service that analyses power-use patterns of single-person households to detect abnormal signs and identify the risk of dying alone early. Korea Minting and Security Printing Corp demonstrated its "K-brand global authenticity verification service", which uses an AI-based digital watermark to check authenticity with smartphone photos alone.

Ryu Je-myung (류제명), second vice minister of the Ministry of Science and ICT, said the event was meaningful in that public institutions, as agents of AI innovation, showcased services people can feel and results in improving administrative efficiency. He said the ministry will support the development of proprietary AI models, expansion of AI computing infrastructure, and demonstration and wider rollout of domestically produced NPUs so that domestic AI technologies are used more broadly at public-sector sites.