Ahead of the digital asset tax scheduled for 2027, South Korea’s tax authority is turning to an artificial intelligence system to detect tax evaders.
According to the Korea Times, the National Tax Service (NTS) of South Korea has launched a tender to develop an AI-based platform designed to monitor digital asset transactions and identify potential tax evasion cases. The project marks another step toward implementing digital asset taxation, which has been repeatedly postponed and is now set to take effect in January 2027.
To make this tax system effective, the government has allocated approximately 3 billion won (around $2 million) to build a state-of-the-art technological platform.
The NTS plans to leverage artificial intelligence and machine learning algorithms to process massive volumes of cryptocurrency trading data. The goal is to identify unusual behavior, suspicious trading patterns, and potential attempts to evade tax obligations.
The tax mechanism includes a total levy of 22%, consisting of a 20% income tax and an additional 2% local tax. The tax will only apply to annual profits exceeding 2.5 million won, approximately $1,700.
By March 2026, the contractor responsible for developing the platform will be selected. The design phase is scheduled for April, followed by continuous testing throughout the year. A pilot program is planned for November 2026, with the full system expected to launch between November and December of the same year.
The platform will not only collect data but also employ advanced analytical capabilities to support targeted tax audits, uncover hidden income from delinquent taxpayers, and detect evasion strategies related to trading.
The NTS announced that analyzed data and lists of suspicious individuals will be shared with other government agencies, including the Korea Customs Service and the Bank of Korea.
Originally approved in 2020, the regulation’s implementation was postponed three times due to industry opposition and political disagreements over tax thresholds.
