OpenAI has introduced IndQA, a benchmark designed to evaluate the performance of AI systems in various Indian languages. The benchmark assesses cultural understanding and reasoning abilities across 12 languages and 10 different domains, such as history, science, and entertainment. IndQA was developed with input from domain experts to ensure its relevance and accuracy. --- Why it matters: This matters for researchers working on multilingual AI models because it provides a standardized evaluation framework for assessing their performance in Indian languages, which are often underserved by current benchmarks. Source: https://openai.com/index/introducing-indqa