In recent years, digital currencies based on blockchain technology have brought about a boom in the research of encrypted digital currencies with the design concept of peer-to-peer trading and decentralization, but due to the lack of supervision and obvious speculative characteristics of the digital currency market, the sharp fluctuations in the market can easily trigger investor sentiment fluctuations, which in turn will lead to social instability and even financial system risks. In this paper, through the analysis of the financial risk source factors of blockchain digital currency, the evaluation index is established, the risk evaluation index system is constructed, and then the CART classification algorithm is used to analyze it and evaluate and test the model. The characteristic factor structure obtained by the CART algorithm in this paper better explains the financial risk characteristics of blockchain digital currency and gives relatively reliable identification results. The results show that the CART decision tree classification method is effective and has a high accuracy rate, which classifies the financial risks of blockchain digital currency, and the method has excellent adaptability and matchability for the classification of risk problems.