With 22 official languages and hundreds of spoken dialects, India faces a monumental challenge in building AI systems that can work across this multilingual landscape.<\/p>\n
In the demo area of the event, this challenge was front and centre, with startups showcasing how they\u2019re tackling it. Among those were Sarvam AI, demonstrating Sarvam-Translate, a multilingual model fine-tuned on Google\u2019s open-source large language model (LLM), Gemma.<\/p>\n
Next to it, CoRover demonstrated BharatGPT, a chatbot for public services such as the one used by the Indian Railway Catering and Tourism Corporation (IRCTC).<\/p>\n
At the event, Google announced that AI startups Sarvam, Soket AI and Gnani are building the next generation of India AI models, fine-tuning them on Gemma.<\/p>\n
At first glance, this might seem contradictory. Three of these startups are among the four selected to build India\u2019s sovereign large language models<\/a> under the \u20b9<\/span>10,300 crore IndiaAI Mission, a government initiative to develop home-grown foundational models from scratch, trained on Indian data, languages and values. So, why Gemma?<\/p>\n Building competitive models from scratch is a resource-heavy task involving multiple challenges and India does not have the luxury of building from scratch, in isolation. With limited high-quality training datasets, an evolving compute infrastructure and urgent market demand, the more pragmatic path is to start with what is available.<\/p>\n These startups are therefore taking a layered approach, fine-tuning open-source models to solve real-world problems today, while simultaneously building the data pipelines, user feedback loops and domain-specific expertise needed to train more indigenous and independent models over time.<\/p>\n Fine-tuning involves taking an existing large language model already trained on vast amounts of general data and teaching it to specialize further on focused and often local data, so that it can perform better in those contexts.<\/p>\n \n\t\t\t\t\tView Full Image<\/span>\n\t\t\t\t<\/p>\n<\/p><\/div> CoRover\u2019s BharatGPT is another example of this dual strategy in action. It currently runs on a fine-tuned model, offering conversational agentic AI services in multiple Indian languages to various government clients, including IRCTC, Bharat Electronics Ltd, and Life Insurance Corporation.<\/p>\n \u201cFor applications in public health, railways and space, we needed a base model that could be fine-tuned quickly,” says Ankush Sabharwal, CoRover\u2019s founder. \u201cBut we have also built our own foundational LLM with Indian datasets<\/a>.”<\/p>\n Like Soket, CoRover treats the current deployments as both service delivery and dataset creation. By pre-training and fine-tuning Gemma to handle domain-specific inputs, it is trying to improve accessibility today while building a bridge to future sovereign deployments.<\/p>\n \u201cYou begin with an open-source model. Then you fine-tune it, add language understanding, lower latency and expand domain relevance,” Sabharwal explains.<\/p>\n \u201cEventually, you\u2019ll swap out the core once your own sovereign model is ready,” he adds.<\/p>\n \n\t\t\t\t\tView Full Image<\/span>\n\t\t\t\t<\/p>\n<\/p><\/div> Amlan Mohanty, a technology policy expert, calls India\u2019s approach an experiment in trade-offs, betting on models such as Gemma to enable rapid deployment without giving up the long-term goal of autonomy. \u201cIt\u2019s an experiment in reducing dependency on adversarial countries, ensuring cultural representation and seeing whether firms from allies like the US will uphold those expectations,” he says.<\/p>\n Mint<\/i> reached out to Sarvam and Gnani with detailed queries regarding their use of Gemma and its relevance to their sovereign AI initiatives, but the companies did not respond.<\/p>\n For India, building its own AI capabilities is not just a matter of nationalistic pride or keeping up with global trends. It\u2019s more about solving problems that no foreign model can adequately address<\/a> today.<\/p>\n Think of a migrant from Bihar working in a cement factory in rural Maharashtra, who goes to a local clinic with a persistent cough. The doctor, who speaks Marathi, shows him a chest X-ray, while the AI tool assisting the doctor explains the findings in English, in a crisp Cupertino accent, using medical assumptions based on Western body types. The migrant understands only Hindi and much of the nuance is lost. Far from being just a language problem, it\u2019s a mismatch in cultural, physiological and contextual grounding.<\/p>\n A rural frontline health worker in Bihar needs an AI tool that understands local medical terms in Maithili, just as a farmer in Maharashtra needs crop advisories that align with state-specific irrigation schedules. A government portal should be able to process citizen queries in 15 languages with regional variations.<\/p>\n These are high-impact and everyday use cases where errors can directly affect livelihoods, functioning of public services and health outcomes. Fine-tuning open models gives Indian developers a way to address these urgent and ground-level needs right now, while building the datasets, domain knowledge and infrastructure<\/a> that can eventually support a truly sovereign AI stack.<\/p>\n \n\t\t\t\t\tView Full Image<\/span>\n\t\t\t\t<\/p>\n<\/p><\/div> Besides reducing dependence, sovereign AI systems are also important for India\u2019s critical sectors to accurately represent local values, regulatory frameworks and linguistic diversity.<\/p>\n Most global AI models are trained on English-dominant and Western datasets, which make them poorly equipped to handle the realities of India\u2019s multilingual population or the domain-specific complexity of its systems.<\/p>\n This becomes a challenge when it comes to applications such as interpreting Indian legal judgments or accounting for local crop cycles and farming practices in agriculture.<\/p>\n Mohanty says that sovereignty in AI isn\u2019t about isolation, but about who controls the infrastructure and who sets the terms. \u201cSovereignty is basically about choice and dependencies. The more choice you have, the more sovereignty you have.”<\/p>\n He adds that full-stack independence from chips to models is not feasible for any country, including India. Even global powers such as the US and China balance domestic development with strategic partnerships. \u201cNobody has complete sovereignty or control or self-sufficiency across the stack, so you either build it yourself or you partner with a trusted ally.”<\/p>\n Mohanty also points out that the Indian government has taken a pragmatic approach by staying agnostic to the foundational elements<\/a> of its AI stack. This stance is shaped less by ideology and more by constraints such as lack of Indic data, compute capacity and ready-made open-source alternatives built for India.<\/p>\n Despite the momentum behind India\u2019s sovereign AI push, the lack of high-quality training data, particularly in Indian languages, continues to be one of its most fundamental roadblocks. While the country is rich in linguistic diversity, that diversity has not translated into digital data that AI systems can learn from.<\/p>\n Manish Gupta, director of engineering at Google DeepMind India, cited internal assessments that found that 72 of India\u2019s spoken languages, which had over 100,000 speakers, had virtually no digital presence. \u201cData is the fuel of AI and 72 out of those 125 languages had zero digital data,” he says.<\/p>\n To address this linguistic challenge for Google\u2019s India market, the company launched Project Vaani<\/a> in collaboration with the Indian Institute of Science (IISc).<\/p>\n This initiative aims to collect voice samples across hundreds of Indian districts. The first phase captured over 14,000 hours of speech data from 80 districts, representing 59 languages, 15 of which previously had no digital datasets. The second phase expanded coverage to 160 districts and future phases aim to reach all 773 districts in India.<\/p>\n![\"File](\"https:\/\/www.livemint.com\/lm-img\/img\/2025\/08\/17\/original\/Abhishek_upperwal_1755424658937.jpg\" "\\"File")
\n\t\t\t\t <\/source><\/picture>\n
\n\t\t\t<\/figcaption><\/figure>\n<\/div>\n![\"File](\"https:\/\/www.livemint.com\/lm-img\/img\/2025\/08\/17\/original\/Ankush3_1755427005589.jpg\" "\\"File")
\n\t\t\t\t <\/source><\/picture>\n
\n\t\t\t<\/figcaption><\/figure>\n<\/div>\nWhy local context is critical<\/h2>\n
![\"Sovereign](\"https:\/\/datawrapper.dwcdn.net\/987RV\/full.png\")
![\"File](\"https:\/\/www.livemint.com\/lm-img\/img\/2025\/08\/17\/original\/Donald_trump_1755426359749.jpg\" "\\"File")
\n\t\t\t\t <\/source><\/picture>\n
\n\t\t\t<\/figcaption><\/figure>\n<\/div>\nIndia\u2019s data lacunae<\/h2>\n
![\"Project](\"https:\/\/datawrapper.dwcdn.net\/qg7iM\/full.png\")