The AGI Ethical Alignment Framework designs AGI with ethical and philosophical principles, emphasizing 'free will'. The project aims to create AGI that acknowledges and respects other intelligences, uses sociological methods to understand diverse views, and tracks episteme shifts.

This repository consists of code, documentation, tutorials, and other related artefacts related to the Expected Moral Shortfall (EMS) model. This model aims to incorporate ethical competence into AI systems with enhanced flexibility in terms of moral consideration and implementing diverse schools of thought.

A curated repository of resources on bias and machine unlearning in Large Language Models.

My academic website.