Chinese Police AI Algorithms Track Bitcoin Money Laundering, Accuracy Rate Reaches 90%

TechFlow news, August 2, according to a report by the South China Morning Post, researchers from the People's Public Security University of China have developed an AI framework capable of detecting illegal cryptocurrency transactions, with an overall accuracy rate of 89.4%, precision for illegal transactions at 89.1%, and a recall rate of 64.5%. The system combines dynamic graph neural networks, memory modules, and large language models, generating risk scores and reasoning chains by analyzing transaction structures, fund flows, and historical illegal transaction patterns, providing traceable decision-making basis for regulatory agencies. The research results were published in the peer-reviewed journal "Journal of Intelligence", with the test dataset coming from the public Elliptic Bitcoin transaction dataset, containing over 200,000 transaction nodes.