India is the Johnny-come-lately in the global artificial intelligence (AI) race and has promised a bold initiative to develop its own foundational AI model to rival the likes of ChatGPT and DeepSeek R1. This announcement, made on Thursday by India’s IT Minister, Ashwini Vaishnaw, marks a key moment in India’s AI journey, as the country aims at building AI systems that reflect its unique needs, culture, and languages, while being a formidable player on the world stage.
The timing of this move is noteworthy, coming in the wake of DeepSeek, a Chinese AI company, gaining attention after its model surpassed ChatGPT as the top-ranked free app on Apple’s App Store. This development has stirred conversations about AI dominance, which has so far been largely monopolized by American companies, particularly Silicon Valley-based OpenAI. Vaishnaw expressed confidence in the country’s ability to compete globally.
He stated that India’s foundational model would be “world-class” and positioned to rival the best AI systems in existence. As part of this ambitious plan, India has announced a new initiative, featuring a common compute facility powered by 18,693 Graphics Processing Units (GPUs). These GPUs, sourced from empanelled bidders like Jio Platforms, CMS Computers, Tata Communications, E2E Networks, and Yotta Data Services, will be made available to startups and researchers. The facility is expected to be one of the most affordable in the world, with the government covering 40 percent of the costs to keep the rate well below global benchmarks.
Vaishnaw promised that the price per GPU hour will be significantly less than one dollar, a rate that is likely to democratize access to advanced AI resources. In addition to this, the government has called for proposals to develop AI models that are tailored to India’s specific context. This includes training models on Indian datasets, which will reflect the country’s diverse languages and cultures while removing biases. These foundational models are critical in generative Al, as they form the bedrock upon which a variety of AI applications are built.
Vaishnaw also emphasized the importance of algorithmic efficiency in achieving these goals. According to him, India’s approach to AI would allow for the creation of foundational models at a fraction of the cost and time taken by leading global competitors. He remains optimistic that within the next few months, India will unveil its own world-class AI model, with some developers expected to complete their projects within 4 to 6 months, while others may take up to 10 months.
The Indian government is not only focusing on the creation of foundational models but also on the broader AI ecosystem. Under the IndiaAI Mission, the government has allocated Rs 10,372 crore to strengthen AI infrastructure, with a focus on making 10,000 GPUs available for AI compute needs. This is part of a larger plan to use AI to address critical population-scale challenges across sectors like healthcare, education, agriculture, logistics, and weather forecasting. As part of the mission, 18 applications have already been selected for the first round of funding, with themes spanning agriculture, learning disabilities, and climate change.
Another critical aspect of India’s AI blueprint is AI safety. With global concerns over the ethical implications of AI, India is putting a premium on creating trustworthy and safe Al models. As part of the IndiaAI Mission, the government has established Al safety institutions that will follow a collaborative, huband-spoke model. These institutions will partner with multiple entities to develop the tools, frameworks, and processes necessary for ensuring Al safety.
Some of the projects approved under this initiative include research into machine unlearning, AI bias mitigation, explainable AI frameworks, privacy-enhancing strategies, and AI algorithm auditing tools. The government’s commitment to a safe and trusted AI ecosystem also comes in the face of potential challenges, such as export curbs on AI chips and GPUs imposed by the United States.
However, Vaishnaw reassured that India is regarded as a trusted country in the global tech space. He highlighted India’s respect for intellectual property rights and its history of signing important Memoranda of Understanding (MoUs) related to technology. Vaishnaw expressed confidence that India’s AI mission will continue to thrive due to the trust that the country enjoys globally.
