Researchers at China's national police academy have developed an artificial intelligence framework capable of detecting illicit cryptocurrency transactions with nearly 90% accuracy, according to a recent study.
The system combines a memory module with large language models to identify pseudonymous cross-border trading patterns that typically evade traditional surveillance.
The development marks a significant escalation in the technical capabilities available to Chinese authorities for monitoring digital asset flows.
By automating the detection of money laundering and other illegal trading practices, the framework reduces the reliance on manual forensic analysis, which has historically been resource-intensive and slow.
This move aligns with a broader trend of increasing regulatory scrutiny on cryptocurrency markets in Asia.
Chinese authorities have previously published detailed technical reports outlining forensic methods for tracking, seizing, and freezing crypto holdings.