从“世界模型”看人工智能在法律场景的实际应用  

"World Models"and the Applications of AI in the Legal Scenario

作　　者：刘庄 Liu Zhuang

机构地区：[1]香港大学法律学院,中国香港

出　　处：《中国应用法学》2024年第2期99-107,共9页China Journal of Applied Jurisprudence

摘　　要：在当前和未来很长一段时间内,数据分析和传统机器学习方法的应用,能够较好解决法律研究和实践中面临的问题,应该受到更多关注和重视;而ChatGPT和Sora等生成式人工智能,存在技术和市场两方面的困境,较法律领域的实际应用尚存在不小距离。法律实践者要重视数字思维,推动利用大数据和机器学习方法辅助决策;也要审慎对待大模型和生成式智能技术,警惕盲目投入导致资源浪费。面对层出不穷的人工智能模型以及法律界的“人工智能热”,本文建议学界和实务界务必密切观察人工智能对法律实践的影响;也建议法律人仔细找准问题、瞄准成熟技术,谋定而后动。In current time and foreseeable future,the application of data analysis and traditional machine learning methods can effectively address the challenges faced in legal research and practice,thus deserving more attention and recognition.However,there are both technological and market-related challenges associated with generative artificial intelligence models like ChatGPT and Sora,which still pose a considerable gap in practical applications within the legal field.Legal practitioners should prioritize digital thinking and promote the use of big data and machine learning methods to assist decision-making.At the same time,caution must be exercised in dealing with large-scale models and generative intelligent technologies to avoid blind investment and resource wastage.Faced with the continuous emergence of artificial intelligence and the"AI fever"in the legal industry,this paper recommends that academia and practitioners closely observe the impact of artificial intelligence on legal practice.Furthermore,legal professionals should carefully identify problems,target mature technologies,and plan before taking action.

关 键 词：世界模型 生成式人工智能 机器学习 数据分析 法律应用 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] D916[自动化与计算机技术—控制科学与工程]