The main goal of this project is to create a new software artefact: a custom Generative Pre-trained Transformer (GPT) for developers to discuss and solve ethical issues through AI engineering. This conversational agent will provide developers with practical application on (1) how to comply with legal frameworks which regard AI systems (like the EU AI Act~\cite{aiact} and GDPR~\cite{gdpr}) and (2) present alternate ethical perspectives to allow developers to understand and incorporate alternate moral positions. In this paper, we provide motivation for the need of such an agent, detail our idea and demonstrate a use case. The use of such a tool can allow practitioners to engineer AI solutions which meet legal requirements and satisfy diverse ethical perspectives.

翻译：本项目的主要目标是创建一个新的软件制品：一个定制的生成式预训练Transformer（GPT），供开发者通过AI工程讨论和解决伦理问题。该对话代理将为开发者提供以下方面的实践应用：（1）如何遵守与AI系统相关的法律框架（如欧盟AI法案~\cite{aiact}和通用数据保护条例~\cite{gdpr}），以及（2）呈现替代性伦理视角，使开发者能够理解并纳入不同的道德立场。本文阐述了开发此类代理的必要性，详细介绍了我们的构想，并展示了一个应用案例。使用此类工具可使从业者设计出既符合法律要求又满足多元伦理视角的AI解决方案。