Morality in AI is often oversimplified. @davidjurgens.bsky.social and @shivanikumar.bsky.social kick off the "Human-Centred NLP" orals #ACL2025NLP with UniMoral, a huge dataset of moral scenario ratings in 6 languages! They find LLMs fail to simulated human moral decisions. bsky.app/profile/shiv...

Can AI grasp how humans across cultures reason through moral dilemmas? ✨Meet UniMoral-a unique multilingual dataset merging psychology & NLP to model moral reasoning as a pipeline. It enables LLMs to reason about decisions and their ethical implications across languages. Thread🧵(1/n)