This paper investigates the relationship between law and eXplainable
Artificial Intelligence (XAI). While there is much discussion about the AI Act,
for which the trilogue of the European Parliament, Council and Commission
recently concluded, other areas of law seem underexplored. This paper focuses
on European (and in part German) law, although with international concepts and
regulations such as fiduciary plausibility checks, the General Data Protection
Regulation (GDPR), and product safety and liability. Based on XAI-taxonomies,
requirements for XAI-methods are derived from each of the legal bases,
resulting in the conclusion that each legal basis requires different XAI
properties and that the current state of the art does not fulfill these to full
satisfaction, especially regarding the correctness (sometimes called fidelity)
and confidence estimates of XAI-methods.

该研究探讨了法律与可解释人工智能（XAI）之间的关系，通过对欧洲法律、AI 法案以及 XAI 方法的要求等方面的研究，得出法律对不同的 XAI 属性有不同的要求，并指出现有技术水平尚不能充分满足这些要求，尤其是在 XAI 方法的正确性和置信度估计方面。

AI 决策应如何解释？从欧洲法律的角度来看的解释要求

How should AI decisions be explained? Requirements for Explanations from  the Perspective of European Law

We present preliminary results about Legistix, a tool we are developing to
automatically consolidate the French and European law. Legistix is based both
on regular expressions used in several compound grammars, similar to the
successive passes of a compiler, and on a new specialized language of
functional type, allowing to describe the changes applied to the texts. Instead
of creating manually a full consolidated version of a text at each modification
date, Legistix generates automatically programs from legal documents written in
natural language to automatically create the consolidated versions.

通过使用编译器的连续传递和一种新的专业的函数类型语言，我们正在开发 Legistix 工具来自动整合法国和欧洲法律，而不是手动创建文本的完整整合版本，从而自动生成程序来创建整合版本。