India’s long-term AI strategy is increasingly focused on building its own foundational models, tailored to local languages, economic needs, and technological constraints. Experts at the Bengaluru Tech Summit emphasized that foreign-built systems cannot fully address India’s cultural and industrial diversity. They warned that if India relies too heavily on global models, the technology may widen existing inequalities instead of promoting inclusive growth.
A major concern is accessibility. Leaders from India’s AI ecosystem stressed that the country must ensure AI reaches every segment of society, not just wealthy urban users or large enterprises. If only a small fraction of the population benefits, the emerging “AI divide” could become even more severe than the earlier digital divide, leaving millions without equal opportunities in education, healthcare, or employment.
Home-grown models are also essential for practical reasons. AI systems trained in foreign contexts often struggle with India’s unique linguistic landscape, noisy real-world environments, and resource-constrained settings. Indian entrepreneurs highlighted the need for AI that can handle multilingual speech, regional dialects, and industrial applications that require scientific reasoning.
