PhD Scholarship in Culturally Inclusive AI Safety for Large Language Models

This PhD scholarship is for candidates interested in developing culturally inclusive AI by creating methods and datasets to align large language models with global social and legal norms. The project focuses on developing multilingual datasets and advanced methods to detect and mitigate social and legal norm violations in large language models (LLMs). Candidates should have an interest or experience in culturally inclusive AI safety, multilingual NLP, AI ethics, or computational social sciences, and must hold a relevant degree with strong skills in programming and machine learning. The scholarship explores AI safety for large language models, developing culturally inclusive methods to detect and mitigate social norm and legal violations globally.