Editor’s Note: Since 2023, Microsoft’s Office of Responsible AI has partnered with the Strategic Foresight Hub at the Stimson Center to convene a diverse group of experts from the Global South to evaluate the impacts of AI in emerging markets. Guided by the question of how AI-related risks and benefits might manifest in various social, cultural, economic, and environmental contexts, program participants identify technological and regulatory solutions that can help mitigate risks and maximize opportunities across the globe. Fellows also have the opportunity to publish at Stimson. In the RAI Case Studies, Fellows share insights about responsible AI governance from within their own thematic and geographic areas of expertise.
By Giulia Neaher, Managing Editor for RAI Case Studies
In India, the triad Samaj, Sarkar, Bazaar — meaning Society, Government, and Market — has long served as a guiding compass for inclusive development. It reflects a uniquely Indian model where innovation is not driven by any one actor in isolation, but emerges from the synergy between communities, public institutions, and markets working together for the public good. Today, this philosophy underpins India’s AI strategy — anchoring innovation in public purpose and ensuring that the gains of technological progress extend beyond economic metrics to tangible societal impact.
The Indian government is trying to champion this integrated approach through initiatives like the India AI Mission, building on the vision first articulated in the 2018 National Strategy for AI. Both emphasize the importance of aligning policy, innovation, and community needs to create a cohesive and inclusive AI ecosystem.
These efforts aim to create a unified AI ecosystem through public-private partnerships, the establishment of cutting-edge computing infrastructure, and the democratization of AI skills and education. By focusing on areas like responsible and ethical AI, local language innovation, and socially impactful applications, this collaborative model aims to ensure AI is both inclusive and responsible.
For years, India’s AI ecosystem was fragmented; government, industry, academia, and civil society often worked in silos, limiting the scale and impact of innovation. Despite global recognition in AI patents, research output, and startup growth between 2018 and 2020, the absence of a unified framework left critical gaps in accessibility, scalability, and ethical governance. This case study traces India’s shift toward a more integrated, multi-stakeholder model — rooted in the Samaj, Sarkar, Bazaar philosophy — and explores how this inclusive approach can offer a blueprint for equitable AI governance worldwide.
India’s AI Ecosystem: Progress and Opportunities
India is emerging as a global leader in AI, positioned uniquely between developed economies and the Global South. By leveraging its vast talent pool, vibrant tech ecosystem, and proactive policy push, India has transitioned from a fragmented AI landscape to a unified, multi-stakeholder model. This transformation has been achieved through policy action like the National AI Mission, increased investment in talent development and startups, and growing adoption amid recognition of AI’s vast economic potential.
The 2025 announcement of the intention to build an indigenous foundation model marks a turning point, moving India from consumer to creator of core AI infrastructure. The country is now among a select group of nations investing in sovereign AI capabilities built for public purpose.
In the meantime, India has made significant strides in AI adoption, with notable achievements that position it as a key global player:
- Talent Leadership: India leads globally in AI skill penetration. The 2024 Stanford AI Index ranks India first, meaning the country’s AI workforce is 2.8 times more skilled in AI than the global average.
- Economic Impact: AI is projected to contribute between $450–500 billion to India’s GDP by 2025, aligning with the government’s vision of a $5 trillion economy. This growth is expected to be driven by sectors such as consumer goods & retail, agriculture, banking & insurance, as well as healthcare.
- Growing AI Adoption: 92% of Indian employees report regularly utilizing AI in their daily work, significantly surpassing the global average of 72%. Additionally, India’s 2024 AI adoption index score is 2.47 on a 4-point scale, with 87% of companies in the middle stages of AI maturity — classified as Enthusiast and Expert adopters.
- Startup Ecosystem: India’s AI startup ecosystem is growing rapidly. The number of generative AI startups in India increased 3.6 times from mid-2023 to mid-2024. These startups are focusing on sectors like healthcare, agriculture, and fintech, contributing to India’s position as a global hub for AI innovation.
These developments showcase India’s potential to not only adopt but also innovate in AI, driven by a synergy of public-private initiatives, academic contributions, and community engagement.
India’s early AI journey, however, was marked by fragmentation. Government, industry, academia, and civil society often operated in silos, limiting the scalability of innovation. Despite notable achievements — ranking 8th globally in AI patent filings and 3rd in academic publications by 2020 — these efforts rarely translated into large-scale, inclusive applications. The absence of shared infrastructure and coordinated governance frameworks left critical gaps in accessibility, scalability, and ethical deployment.
India’s transition to a more integrated AI ecosystem is now yielding tangible results, as listed below:
- Unified Ecosystem & Compute Access: The 2024 IndiaAI Mission, backed by ₹10,371 crore ($1.25 billion), laid the foundation for a comprehensive, public-purpose AI infrastructure. By mid-2025, India had scaled its AI compute capacity to over 34,000 high-end GPUs, accessible at subsidized rates of ₹67/hour (~$0.78), through the IndiaAI Compute Portal. This has unlocked training access for startups, students, and researchers across the country, while powering sovereign model development and lab networks in smaller cities.
- Public-Private Synergy: Collaborations like Microsoft’s work with RailTel and Apollo Hospitals showcase how private sector capabilities are intentionally being aligned with national priorities to create AI-first sectors.
- Sectoral Deployment:
- Agriculture: Predictive analytics and water management tools are continuously improving yields.
- Healthcare: AI enhances remote diagnostics and hospital workflows.
- Governance: Tools like Bhashini are making public services multilingual and accessible.
- Localized Innovation: Inclusive initiatives, such as AI-driven tools for the visually impaired, reflect India’s ability to build context-aware solutions.
- Generative AI Leadership: Indian enterprises are among the fastest adopters of generative AI globally, with over 33% of projected global market focus by 2027 concentrated in this space.
Together, these developments signal a shift from isolated innovation to a collaborative, mission-driven ecosystem. Aligned with the Samaj, Sarkar, Bazaar ethos, India is not just advancing technologically — but trying to do so in a way that is inclusive, ethical, and globally relevant.
Case Studies: Inclusive AI Solutions in Action
India’s AI transformation is thereby not just driven by technology, but by how it is applied — grounded in public needs, local languages, and equitable access. The following initiatives showcase how AI is being deployed to solve real-world challenges, from linguistic inclusion and open data to sovereign infrastructure and regional innovation.
Multilingual AI and Digital Inclusion Initiatives
India’s AI vision emphasizes linguistic diversity. The Bhashini program, launched in 2022 and expanded through 2025, anchors the country’s efforts to build multilingual AI that ensures language is no longer a barrier to accessing technology.
Bhashini provides open-source speech-to-text, text-to-speech, and translation APIs in 22+ Indian languages. These tools are used by developers to build inclusive services — ranging from voice-based e-governance tools to regional language support for e-commerce platforms.
The platform has enabled over 300 AI models in Indic languages. It powers real-time translation for speeches, supports state-level apps like Rajasthan’s “Pehchan” registration tool, and integrates into services like ONDC’s multilingual shopping assistant. These efforts ensure citizens across linguistic backgrounds can access digital services
AI Kosh – A National Platform for Inclusive Datasets
India’s ambition to democratize AI development hinges, of course, on access to high-quality, inclusive, and ethically sourced data. The launch of AI Kosh in March 2025 represents a foundational step in that direction.
AI Kosh is a centralized, secure repository that hosts over 360 curated, non-personal datasets and pre-trained AI models. These include datasets on Indian languages, health records, satellite imagery, pollution data, census indicators, and agricultural trends, compiled from public institutions and ministries. The platform also features development tools, quality metrics, and an AI sandbox for experimentation.
AI Kosh addresses one of the most overlooked challenges in AI: dataset bias and lack of context for global applications. It provides inclusive, anonymized datasets for startups, researchers, and academic institutions to build AI solutions tailored to Indian needs — especially in areas like healthcare, agriculture, education, and regional language translation.
Researchers and communities can now contribute open datasets, especially in regional languages and public health, while the Ministry of Electronics & IT and NDAP curate, manage, and fund the repository. On the other hand, startups can use AI Kosh datasets to develop locally relevant models and applications.
AI4Bharat: Open-Source AI for India
An initiative by the Indian Institute of Technology (IIT) Madras, AI4Bharat focuses on creating open-source AI tools to solve challenges unique to India, particularly in low-resource settings. AI4Bharat develops natural language processing (NLP) models for underserved Indian languages, with applications in telemedicine, smart governance, and education. Its open-source datasets empower developers to create localized solutions.
Tribal and rural populations now benefit from AI-powered governance and healthcare tools. Developers across India leverage AI4Bharat’s resources to innovate in regional contexts, and the initiative is expanding datasets for tribal dialects, ensuring inclusivity for historically marginalized groups.
Building India’s Sovereign Foundation Model
To reduce dependence on foreign AI models and assert technological sovereignty, the Government of India selected Sarvam AI in early 2025 to build the country’s first large-scale, indigenous foundation model. Trained entirely in India using 4,096 government-backed NVIDIA H100 GPUs, the 70-billion-parameter model is designed for fluency in Indian languages, advanced reasoning, and deployment on low-resource devices.
This marks a pivotal step in India’s AI journey by creating a population-scale model aligned with its linguistic, cultural, and regulatory context. In May 2025, the initiative expanded with three more teams: Soket AI Labs (120B open-source Indic LLM under Project EKA), Gan.ai, and Gnani.ai, each tasked with developing specialized models for language, speech, and real-time interaction.
Supported by a ₹1,500 crore ($174.5 million) innovation fund and nearly 190 proposals from across academia and industry, this multi-pronged push reflects India’s ambition to rank among the global top five in each AI domain. Sarvam AI thereby stands at the forefront of this mission — delivering foundational models that not only perform, but speak to India’s values, languages, and people.
Policy, Governance, and Ethical AI: Mapping a Responsible Ecosystem
India’s Responsible AI ecosystem relies on the active cultivation of a multidisciplinary network of policy, social, and technological initiatives, both within and outside of traditional regulation. By dually cultivating a sustainable, interrelated AI market and informed, ethical regulatory schema, India works to build a long-lasting and durable AI ecosystem.
Systemic Factors
The following factors (policy, public-private partnerships, community, scalability, ethics, and talent) work together to create an interdisciplinary, holistic AI ecosystem for India. Each systemic factor is further elaborated below with concrete examples of how it is being implemented or observed in the Indian context.
Policy Frameworks and Government Leadership
Proactive Policy Development: Initiatives like the IndiaAI Mission and Bhashini reflect India’s commitment to national AI strategies that address linguistic inclusion, ethical governance, and cross-sector collaboration.
Lesson: Clear, actionable policies foster innovation while aligning public, private, and community interests.
Public-Private Partnerships
Stakeholder Synergy: Collaborations — like Microsoft with Apollo Hospitals — demonstrate how government leadership paired with private sector expertise enables AI-first sectors.
Lesson: Clear, actionable policies foster innovation while aligning public, private, and community interests.
Startup Ecosystem Development: Programs such as Bhashini offer open datasets that empower startups to build regional language solutions.
Lesson: Supporting grassroots innovation accelerates inclusive AI development.
Community-Centric Design
Feedback Loops: Initiatives involving user groups — such as visually impaired communities — ensure real needs shape AI design.
Lesson: User-centered design improves relevance and adoption.
Linguistic and Cultural Inclusion: Bhashini underscores the need to serve India’s linguistic diversity.
Lesson: AI must adapt to local languages and cultural contexts to be truly inclusive.
Scalability Through Open Data and Infrastructure
Open-Source Models: Projects like AI4Bharat democratize innovation by making high-quality models and datasets publicly accessible.
Lesson: Open tools expand equitable access to AI development.
Shared Infrastructure: Cloud platforms and compute access (e.g., Microsoft’s AI infrastructure, IndiaAI Cloud) lower entry barriers.
Lesson: Shared infrastructure enables participation beyond elite and established institutions.
Ethical AI and Accountability
Bias Mitigation: Regular audits — mandated in the IndiaAI Mission — reduce algorithmic discrimination.
Lesson: Accountability frameworks are essential for fairness.
Transparency and Explainability: Tools like SUPACE (Supreme Court AI) promote clarity in decision-making systems by assisting with low-stakes legal tasks like background research and case summarization.
Lesson: Transparency fosters trust and legitimacy in AI use.
Talent Development and Capacity Building
AI Skills Training: Microsoft’s goal to train 10 million people in AI by 2030 highlights the scale of investment needed.
Lesson: Skill-building is foundational to sustained AI growth.
Academic Collaborations: Institutions like IIT Madras and the Wadhwani Institute bridge research with real-world applications.
Lesson: Academia-industry partnerships accelerate innovation and talent pipelines.
Policy and Regulatory Shifts
India’s regulatory scheme complements this interdisciplinary ecosystem by adopting a “light-touch, innovation-friendly” regulatory approach grounded in ethics and oversight.
Key Initiatives:
- Digital Personal Data Protection Act (2023): Sets foundational privacy, consent, and data rights rules for AI systems.
- Safe & Trusted AI Program (2024): Supports research in AI risk, fairness, and transparency tools.
- AI Safety Institute (2025): Establishes technical and legal standards for explainability and accountability.
Ethics in Practice:
Public sector models are vetted for fairness, open-source models are monitored for misuse, and projects like AI Kosha and Sarvam AI embed cultural alignment and transparency by design.
India’s model shows how AI progress can be balanced with equity, safety, and public purpose — offering a blueprint for inclusive and ethical AI governance worldwide.
Challenges and Opportunities Ahead
On the one hand, the success of India’s AI ecosystem faces significant challenges. These include limited digital literacy and connectivity, despite inclusion efforts such as E-Parvai; algorithmic bias and ethical concerns, which can lead to social discrimination if not properly addressed; a significant talent gap that slows startup innovation; and fragmented collaboration among government, academia, and industry that impedes strategic progress. Additional complications will likely arise from the need for improved energy infrastructure and massive amounts of energy to power AI data centers, raising the question of how to increase energy consumption sustainably. And, last but not least, striking the balance between foreign investment and domestic data governance will continue to be a challenge that emerging AI partnerships will need to balance. Addressing these issues requires targeted expansion of literacy programs and internet access, regular ethical audits, scaled-up skilling initiatives, and formalization of the Samaj, Sarkar, Bazaar frameworkfor ecosystem-wide collaboration.
However, India’s AI ecosystem is also positioned for significant growth. Strategic opportunities including robust public–private partnerships, exemplified by initiatives such as MARVEL and Microsoft’s healthcare collaborations, can enable scalable adoption in vital sectors. Meanwhile, opportunities for inclusive and societally beneficial advancement abound through programs like Bhashini and AI4Bharat that develop culturally and linguistically relevant AI solutions; the integration of generative AI in education and health to overcome literacy barriers and improve services; and the application of AI for sustainability, deploying tools for water management, waste reduction, and climate-resilient agriculture as advanced by the Wadhwani Institute.
India’s success will hinge on sustained collaboration, ethical governance, and strategic investment. Prioritizing connectivity, skilling, and inclusive innovation will not only advance domestic goals but also position India as a global leader in responsible AI.
