Artificial Intelligence is one of the most used words in India. It has become a buzzword, almost similar to Y2K in the late 1990s. Everyone knew that it was going to herald a once-in-a-generation change. But something unexpected happened towards the end of January that ended whatever complacency status quoists had.
On January 30, US technology company Anthropic released 11 open-source plugins for its AI workplace suite Claude Cowork. Unlike conventional chatbots, Cowork functions as an autonomous digital colleague: it reads files, drafts documents, reviews contracts, and executes multi-step workflows across legal, finance, sales, and marketing — with minimal human instruction. Days later, Anthropic released Claude Opus 4.6, a model capable of coordinating teams of AI agents for financial research and due diligence.
The new AI agents had almost eliminated the role of humans in technology, sending tremors across the global corporate world. Nearly $285-billion market capitalisation was erased globally. In India, the Nifty IT index fell 5.87 per cent — its steepest fall since March 2020 when India announced Covid-induced lockdown — wiping out nearly `2 lakh crore. TCS and Infosys each fell over 7 per cent on the day; Tech Mahindra lost over 5 per cent.
The core fear that triggered this meltdown: if one AI agent can do the work of teams, India’s headcount-based outsourcing model faces existential repricing. It threatens India’s domination in global IT and ITeS services and forces Indian businesses and government to relook at the country’s readiness to adapt and build local AI infrastructure.
Even before ChatGPT popularised artificial intelligence in the country, the Central government in 2024 approved the IndiaAI Mission — a `10,300 crore programme under the Union ministry of electronics and information technology (MeitY). The goal was audacious but clear: build India’s own AI infrastructure, train its own models, and reduce the country’s dependence on foreign AI systems controlled by American and Chinese firms.
About 18 months ago, India was debating whether it needed 1,000 GPUs (graphics processing units) for AI research. By 2025, however, the government had deployed over 38,000 GPUs — beating its original target of 10,000 — and offered them at a subsidised rate of just `65 per hour to Indian startups and researchers. This target climbed further to 50,000 GPUs by the end of the year.
From almost nothing, India had built one of the more accessible public compute pools in Asia. India became one of the four countries in terms of compute capacity, climbing up from the seventh spot two years ago, which is indeed an excellent performance.
India’s most ambitious goal was to develop its indigenous large language models (LLMs) — AI systems trained from scratch on Indian data, capable of reasoning in Indian languages, and deployed within Indian borders.
The government selected Bengaluru-based startup Sarvam AI in April 2025 to build India’s sovereign LLM. Running parallel, the government-led initiative BharatGen anchored at IIT Bombay built Param 2, a 17-billion-parameter multilingual model. Both projects are trained across all 22 scheduled Indian languages. The ambition is to eventually scale to a trillion parameters, serving use cases in agriculture, law, health, finance, and education.
Beyond the flagship projects, the IndiaAI Mission expanded to include 12 organisations building LLMs: from Fractal Analytics (developing India’s first large reasoning model) to Tech Mahindra (building a Hindi-first model for education), Gnani.ai (speech-to-speech AI handling 10 million voice interactions daily), and Soket AI (building open-source foundational models with permissive licences).
The Mission also established Centres of Excellence in healthcare, agriculture, sustainable cities, and education. It launched the IndiaAI Safety Institute, created a datasets platform, and began a skilling programme to retrain government officials and young professionals for an AI-enabled economy.
So, how does India fare globally in AI technology? While the success was far more than the time invested, it is still far from the frontier.
According to the Stanford AI Index, India is among the top four countries globally in AI skills, capabilities, and policies — a climb from seventh place just two years prior. India is the second-largest contributor to AI projects on GitHub. Enterprise AI adoption stands at 89 per cent, well above the global average of 69 per cent. A wave of foreign capital is coming to India, indicating that the country will be a consequential AI market, not a peripheral one.
However, in the frontier model development, India has not yet come up with a system that competes with GPT-4, Gemini, or China’s DeepSeek. While India’s LLMs are promising, culturally relevant, and cost-efficient, they are not yet world-class.
The challenge is the most acute. India has fewer than 300 skilled AI researchers, which is far fewer than the UK or France.
Though India produces more engineers than any other nation annually, the pipeline of deep AI researchers is thin.
India has enormous data from digital public infrastructure at the population scale (Aadhaar, UPI, ONDC), which can allow companies to build cost-efficient solutions for people. It spends just 0.7 per cent of GDP on research and development. Without sustained, patient investment in foundational research, India risks being confined to a consumer and adapter of AI and the creator. Companies also focus too much on paying dividends to shareholders rather than investing in cutting-edge R&D.
In an era when militaries are harnessing AI capabilities, India using foreign AI capabilities is akin to depending on a foreign military. India must, therefore, plug brain drain and race against time to achieve its goals.
Disclaimer : This story is auto aggregated by a computer programme and has not been created or edited by DOWNTHENEWS. Publisher: deccanchronicle.com
