Cultural Innovation

ARTIFICIAL INTELLIGENCE (AI) has emerged as the defining frontier of global competitiveness, shaping economies, governance, and even cultural identity. As nations race to harness their potential, the question for India is no longer whether to participate but how to lead. The AI era cannot be built on imitation. India’s path must be uniquely its own—one that draws on its vast scale, linguistic diversity, and digital public infrastructure to build systems serving both economic growth and societal progress. This demands a new compact among government, academia, startups, and corporations—a collaborative effort to create a sovereign, inclusive, and trusted AI ecosystem rooted in public purpose. The rise of large language models (LLMs) such as GPT and Gemini has dazzled the world with their fluency and scale. Yet, behind the spectacle lies a sobering truth: these models consolidate power more than they democratise it. Only a handful of trillion-dollar corporations can afford to train them, leaving most nations outside the circle of influence. Their inner workings are opaque and inaccessible to public audit, and their foundations are overwhelmingly shaped by Western, English-language data. The result is a subtle but real flattening of worldviews— systems that struggle to understand low-resource languages or culturally diverse contexts. For countries like India, this imbalance risks perpetuating a new form of technological dependency. The current LLM wave, in many ways, reflects a race for dominance rather than inclusion. Venture capital follows scale, not necessarily social impact. But true innovation will not come from building ever-larger models; it will come from building smarter ones—systems that are efficient, contextual, and deeply aligned with human values. The real opportunity for India lies in developing AI that understands its people, languages, and institutions—in building technology that works for India, not merely in India. India’s vulnerability today stems from limited ownership across the AI value chain—data, compute, and knowledge. Although the country generates vast volumes of digital data through its public platforms, telecom networks, and online marketplaces, much of this data is processed and refined by foreign platforms that use it to train global models India has little access to. The situation is similar in compute infrastructure: control over high-performance chips and cloud capacity remains concentrated among a few global entities, effectively making India a renter of its own digital capabilities. Meanwhile, India’s best AI researchers and engineers often end up serving global interests, while domestic research institutions struggle with underfunding and fragmentation. What appears to be a technological imbalance is, in reality, a matter of strategic sovereignty. Yet, this asymmetry also offers India a chance to innovate differently. The LLM era is not the endpoint of AI—it is simply the beginning of a new phase. For India, the challenge is not to replicate trillion-dollar models but to leapfrog them—to pioneer new paradigms of AI that are efficient, inclusive, and socially grounded. We can build smaller, frugal, multilingual models that directly empower citizens, enterprises, and governments. The world may chase the largest models; India can lead by building the most meaningful ones—rooted in its people, its data, and its values. Few nations possess the foundational ingredients India already has. Our Digital Public Infrastructure—Aadhaar, UPI, ONDC, and the Account Aggregator framework—forms a globally acclaimed base for ethical and scalable digital innovation. Add to that India’s unmatched linguistic and cultural richness which provides a natural testbed for multilingual and multimodal AI. Our academic and startup ecosystems are vibrant and fast-evolving, while India’s large private enterprises bring the engineering muscle and execution capacity needed to scale innovation. If India plays this right, AI can become to the 2030s what DPI was to the 2010s—an enabling layer that fuses public purpose with private ingenuity. AI will not, and should not, be built by governments alone, nor can it be left solely to the market. The task before us is to institutionalise collaboration—to align incentives for public and private innovation around shared national goals. Corporations bring global R&D experience, compute infrastructure, and implementation expertise. Startups bring agility, domain focus, and risk appetite. Academia anchors long-term research and intellectual depth. Government and think-tanks like iSPIRT provide public architecture, interoperability standards, and governance frameworks. Together, they can co-create what might be called a ‘Public–Private AI Commons’—a federated ecosystem where infrastructure is open, innovation is competitive, and outcomes are collectively beneficial. To move from aspiration to capability, India must invest in deep and enduring R&D institutions. We need a National AI Research Grid connecting public labs, universities, and startups through shared datasets and compute infrastructure—a kind of “CERN for AI”. Alongside, a Compute Commons can ensure broad access to high-performance computing resources, while an Open Language Data Commons can ethically curate multilingual datasets for all Indian languages. Fellowship programmes should help researchers transition into entrepreneurial roles, and a Public-Interest Research Fund can seed mission-driven AI efforts in healthcare, agriculture, education, and law. Deep R&D in India must be open, multidisciplinary, and future-facing—protected from short-term pressures and corporate silos. The foundation for all of this already exists. India’s DPI provides a secure and interoperable digital fabric on which AI applications can be built for real-world impact. Imagine citizen-facing AI co-pilots that help people access welfare, healthcare, or financial services through natural-language conversations in their own dialects. Picture AI tools that assist policymakers in simulating outcomes, forecasting needs, and allocating resources more effectively. Consider enterprises innovating responsibly atop open DPI rails to reach new markets. The underlying principle is simple: AI as a public good—enabled by DPI, scaled by private innovation, and governed by shared responsibility. As AI weaves itself into every layer of the economy, India must also shape a techno-legal framework that safeguards trust while enabling innovation. This means ensuring algorithmic transparency in high-stakes systems, establishing interoperability standards for agents and APIs, guaranteeing data and model portability to avoid monopolistic lock-ins, and embedding ethical safeguards into the design phase rather than treating them as compliance afterthoughts. India should also maintain sovereign compute capabilities for critical sectors such as healthcare, defence, and public administration. The goal is not regulation for its own sake but governance that creates confidence—turning trust into a national competitive advantage. India’s AI journey, then, must be collaborative by design. Our objective is not to nationalise innovation but to nationally align it—harnessing the strengths of every actor in the ecosystem towards a shared vision of prosperity, inclusion, and trust. If the last decade was about connecting India through digital rails, the next will be about empowering India through intelligent ones. The world may pursue the biggest models. India can lead by creating the most human ones—AI that speaks every language, reflects every culture, and serves every citizen.

Cultural Innovation

Sign in to read full story