A new generation of artificial intelligence called ChatGPT, which quickly attracted 100 million users worldwide, has become a hot topic of discussion. When it comes to artificial intelligence, the release of AlphaGo a few years ago was equally impressive. In the highly anticipated "human-computer battle", AlphaGo used deep learning algorithms to neatly defeat world champion Go expert Lee Sek seok of South Korea. "AlphaDog" became a household name and even made it to the annual Chinese hot word list. Compared to AlphaGo's outstanding skills in specific fields such as Go, ChatGPT seems to be more grounded, truly achieving "flying into the homes of ordinary people". It can engage in dialogue through learning and understanding human language, interact with contextual content, chat and communicate like humans, and even complete tasks such as writing emails, video scripts, copywriting, translation, and writing papers, which is widely sought after by people. With the global popularity of ChatGPT, driven by favorable market factors, domestic and foreign technology giants have quickly followed suit and invested a large amount of funds to develop similar artificial intelligence. With the help of this wave of enthusiasm, it can be foreseen that the daily work of various industries, including the judicial field, will be closely integrated with artificial intelligence in the future. In fact, many years ago, artificial intelligence had already made some attempts in the judicial field and was fruitful. Canadian author Gladwell described an artificial intelligence experiment in his book "The Stranger Effect". The project was jointly conducted by Harvard University economist Murray Nathan and bail experts from the University of Chicago. The collaborative team collected records of 550000 defendants who underwent interrogation hearings in New York City from 2008 to 2013, and found that the number of New York City judges who granted bail was approximately 400000. The project team establishes an artificial intelligence system that inputs the same inquiries provided by prosecutors to judges in these interrogation cases (including the age and criminal records of the defendants). Through statistics, analysis, and evaluation, the artificial intelligence system processes these cases and lists the 400000 bail personnel ultimately determined by the system. The final outcome of this "human-machine confrontation" in the judicial field is also quite uneven. Compared to the bail decision made by New York City judges, the artificial intelligence system selected 400000 bail personnel, and the repeat crime rate during the waiting period for trial was nearly 20% lower than the former. Moreover, the information on both sides is not symmetrical. When New York City judges make bail decisions, they have three sources of consultation, including basic information about the defendant, testimony from district prosecutors and lawyers, and face-to-face questioning of the defendant in court. The artificial intelligence system only has the defendant's age and criminal record, but the bail decision made is actually more reasonable. In the book, Gladwell attributes the excellent performance of artificial intelligence in bail decisions to its ability to use a large amount of data processing, to some extent avoiding the misjudgment caused by natural individuals in interpersonal communication due to the existence of "preset positions" and "transparency principles" (assuming that the communication object can speak and act consistently). In fact, the controversy caused by the advantages and disadvantages of artificial intelligence essentially reflects the positioning of the "human-machine" subject. On December 8, 2022, the Supreme People's Court issued the "Notice on