大模型生成的合同看似像样，实则漏洞百出，难保法律效力。
Contractual Deepfakes: Can Large Language Models Generate Contracts?
- 用统计模式生成合同文本，缺乏真实语义理解与法律推理能力。
- 生成合同常出现条款矛盾或不适用于具体交易场景。
- 适合关注AI法律应用局限性的从业者、法务人员阅读。
尽管大型语言模型具备惊人的文本生成能力，但它们并不理解词语含义，缺乏上下文感知和逻辑推理能力，其输出仅为统计上常见词序的近似。然而，合同起草常被视为可被该技术辅助的典型法律任务。本文旨在终结这一不切实际的设想。预测词语不同于在具体交易情境中使用语言，复现常见合同措辞也不同于法律推理。大模型虽能生成表面合理、通用的合同文档，但在冷静审视下，这些文件可能成为条款冲突的无用组合，或虽具法律效力却完全不适合特定交易。本文质疑了大模型威胁法律行业存续的简单假设。
原文摘要 · Abstract (English)
Notwithstanding their unprecedented ability to generate text, LLMs do not understand the meaning of words, have no sense of context and cannot reason. Their output constitutes an approximation of statistically dominant word patterns. And yet, the drafting of contracts is often presented as a typical legal task that could be facilitated by this technology. This paper seeks to put an end to such unreasonable ideas. Predicting words differs from using language in the circumstances of specific transactions and reconstituting common contractual phrases differs from reasoning about the law. LLMs seem to be able to generate generic and superficially plausible contractual documents. In the cold light of day, such documents may turn out to be useless assemblages of inconsistent provisions or contracts that are enforceable but unsuitable for a given transaction. This paper casts a shadow on the simplistic assumption that LLMs threaten the continued viability of the legal industry.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。