Dates: 2024
Principal Investigator: Dr. Malak Baslyman
Description:  Eliciting Ethical requirements are essential for creating responsible AI systems, especially as AI becomes more integral to critical decision-making. Neglecting ethics early on can result in biased AI outcomes. This study addresses these concerns by identifying ethics goals, aligning them with measurable criteria, and analyzing current AI ethics regulations. Using Large Language Models (LLMs), the project proposes a method to identify ethical issues in existing requirements and categorize violations.  Additionally, it suggests engaging stakeholders in conversations to elicit ethical requirements, considering both the AI system's domain and the regulatory landscape. By easing the challenge of collecting ethical requirements and ensuring diverse perspectives are considered, this approach aims to foster a more comprehensive and holistic elicitation process.