China's AI-Powered Bitcoin Laundering Tool Hits 90% Accuracy

Chinese researchers have developed an AI-powered tool that can flag illicit Bitcoin transactions with near 90% accuracy. The framework, created by the People's Public Security University of China, combines a memory module and a large language model to detect laundering patterns in pseudonymous cross-border transactions.

The study, published in the Journal of Intelligence, aims to provide regulators with a repeatable way to trace illicit cryptocurrency transactions. According to Dr. Sun Jingchao, the corresponding author, the design allows for retention of transaction context across multiple hops, which is crucial for identifying laundering networks that move funds through several wallets before cashing out.

The timing of this development coincides with China's broader enforcement push against virtual currency-related crimes. In 2025, Chinese prosecutors indicted over 3,259 people for money laundering tied to virtual currencies and underground banks.