The Korea Seed and Variety Service announced on October 14, 2026, the release of a newly developed artificial intelligence software designed to predict optimal rice seed harvest times. Jointly created with the Korea Electronics Technology Institute's Jeollabuk-do Province branch, the program aims to address climate change challenges.
How the AI Program Works
When a farmer enters basic information such as the cultivation region and heading date, the artificial intelligence analyzes temperature data from 1980 to 2025 alongside current statistics. The system then calculates the optimal harvest time for each field and displays it in a graph.
A pilot test conducted this year across 9,565 plots using 24 government-supplied rice varieties demonstrated a margin of error of plus or minus three days.
Distribution and Future Plans
According to the Korea Seed and Variety Service, the software and its user manual are being distributed free of charge to government-distributed seed production sites, local governments, and private seed companies.
The agency plans to further improve prediction accuracy by incorporating agricultural satellite data alongside field harvest statistics. KSVS Director General Yang Joopil stated that the program will assist farms struggling with traditional estimation methods amid climate change.
Key facts and where they come from
- KSVS and the Korea Electronics Technology Institute jointly developed the AI software.
The Korea Seed and Variety Service (KSVS) on Oct. 14 said it jointly developed the high-tech program with the Jeollabuk-do Province branch of the Korea Electronics Technology Institute.
- A pilot test on 9,565 plots showed a margin of error of plus or minus three days.
showed the high accuracy of the program, with harvest times having a margin of error of plus or minus three days.
- The software analyzes temperature data spanning from 1980 to 2025.
recorded temperature data from 1980 to 2025, along with this year's statistics to calculate the optimal harvest time for each field in a graph.
