At the Computational Ethics Lab at New York University, we examine the ethical reasoning capabilities of large language models and generative AI systems through systematic analysis of their responses to moral dilemmas and ethical frameworks. Our interdisciplinary research combines computer science, philosophy, psychology, and cognitive science to understand how AI systems process ethical questions and make moral judgments.

Our lab investigates how large language models approach complex ethical reasoning by analyzing their responses to classical moral dilemmas, their ethical self-descriptions, and their alignment with established frameworks in moral psychology and philosophy. We're particularly interested in:

• Comparing ethical reasoning across different AI models

• Analyzing how AI systems utilize consequentialist and deontological frameworks

• Mapping AI ethical reasoning to moral foundations theory and developmental stages

• Developing methodologies for ethical benchmarking of AI systems

• Understanding the implications of AI moral reasoning for human-AI collaboration

We continue to investigate multiple dimensions of AI ethics, including cultural and religious influences on AI reasoning, the impact of fine-tuning on ethical frameworks, and comparisons between human and AI moral judgments. Our lab welcomes collaborations with researchers across disciplines interested in exploring the evolving landscape of computational ethics.