India must counter AI censorship and exclusion with fairer sovereign data

By Nimmish Sany Frontier math puzzles, cancer cures, new superconductor materials — the list of breakthroughs enabled by large language models (LLMs) in recent months is long. Yet, ask one to explain Advaita Vedanta or Sunyavada , and it will likely expose a chink in its armour. Depending upon the data used to train it, the model will likely offer a Western, or perhaps a Chinese, interpretation of these ideas.