Chinese AI Model for Crypto Money Laundering Detection Achieves 90% Accuracy

Researchers at China's Public Security University developed an AI framework with 90% accuracy to detect illicit crypto transactions, aiding regulatory crackdowns as prosecutions hit 3,259 in 2025.

Woofun AI reports that researchers at the People's Public Security University of China have engineered an artificial intelligence framework designed to identify illegal cryptocurrency transactions. By integrating memory modules with large language models, the system analyzes transaction data to uncover anonymized and cross-border illicit activities with an overall recognition accuracy approaching 90%. The study, published in the peer-reviewed journal Intelligence Journal, indicates the framework offers technical support for regulatory authorities combating money laundering and economic crime.

Concurrently, data shows Chinese procuratorial authorities prosecuted 3,259 cases involving virtual currency money laundering and underground banks in 2025, reflecting intensified enforcement efforts.