Abstract

Artificial Intelligence is increasingly shaping how social information is processed and interpreted raising important questions about technology social identity and caste in India. AI systems used in search engines recruitment education governance and social media may analyse names language occupations regional expressions and cultural references as possible indicators of caste-related identity. This paper examines how AI can infer caste from Indian names linguistic patterns and cultural cues while highlighting the uncertainty and ethical risks involved. It treats caste recognition not as a neutral technical task but as a socio-technical process influenced by historical inequality stereotypes and institutional practices. Since caste meanings differ across regions languages and communities algorithmic predictions may be incomplete inaccurate or discriminatory. The paper reviews research on caste bias demographic inference and algorithmic discrimination and develops a framework focused o