中国公安研究人员开发AI模型追踪加密货币洗钱，交易识别准确率接近90%

Odaily Planet Daily News: Researchers at the People's Public Security University of China have developed an AI framework for detecting illegal cryptocurrency transactions, with an overall recognition accuracy of nearly 90%. The model combines memory modules with large language models (LLMs) to analyze cryptocurrency transaction data, identifying anonymized and cross-border illegal transactions. The researchers stated that the framework can provide technical support for regulatory authorities in combating cryptocurrency money laundering and economic crime.

The study was published in the Chinese peer-reviewed journal Intelligence Journal. Reports indicate that as financial crimes involving cryptocurrencies draw increasing regulatory attention, Chinese procuratorial authorities have prosecuted 3,259 cases of money laundering involving virtual currencies and underground banks in 2025. (South China Morning Post)