Decolonial AI Alignment: Openness, Vi\'{s}e\d{s}a-Dharma, and Including Excluded Knowledges

arXiv:2309.05030v3 Announce Type: replace-cross
Abstract: Prior work has explicated the coloniality of artificial intelligence (AI) development and deployment through mechanisms such as extractivism, automation, sociological essentialism, surveillance, and containment. However, that work has not engaged much with alignment: teaching behaviors to a large language model (LLM) in line with desired values, and has not considered a mechanism that arises within that process: moral absolutism -- a part of the coloniality of knowledge. Colonialism has a history of altering the beliefs …

abstract ai alignment alignment artificial artificial intelligence arxiv automation cs.ai cs.cy deployment development however intelligence language language model large language large language model line llm prior stat.ml surveillance teaching through type values work