Entities currently developing large language models (LLM) backed by the IndiaAI Mission have been instructed to make bias mitigation a clear priority before the artificial intelligence (AI) models are deployed. Given India’s unique diversity, and society, care should be taken that government-supported foundational AI models should not throw up insensitive results when confronted with challenging prompts, officials at the ministry of electronics and information technology (MeitY) told ET.
“Sensitive connotations linked with caste, gender, food…
