In a recent Mann ki Baat broadcast, Prime Minister Narendra Modi pulled off a linguistic feat that even P.V. Narasimha Rao, his predecessor who was fluent in 17 languages, could not achieve. Modi addressed his diverse audience in their native tongues—from chaste Telugu and Bengali to Odia and Punjabi.
But unlike Rao, who belonged to the pre-internet era, Modi had advanced tools at his disposal. He got Bengaluru startup Sarvam AI to recreate his voice, originally recorded in Hindi, in multiple Indian languages. Sarvam AI, selected by the Union government to build India’s first homegrown large language model (LLM), is building a model capable of reasoning, fluent in Indian languages and ready for scaling to the population level. The Mann ki Baat demo was just a preview.
This generative AI ‘trick’, however, is just a starting block when it comes to India’s deep tech ambitions. Deep tech refers to innovations from startups and companies driven by advanced science and engineering in fields like robotics, space, defence, clean energy, biotechnology and quantum computing. With India’s technology startup boom now largely driven by consumer app-driven ‘shallow tech’, uncomfortable questions are being raised about how future-ready the country is in the deep tech space.
Union Minister Piyush Goyal recently sparked debate by calling out India’s startup ecosystem. “We are focused on food delivery apps,” he said, “turning unemployed youth into cheap labour so that the rich can get their meals without moving out of their house!”
Goyal raised two valid points. One, Indian entrepreneurs are not focusing enough on deep tech and advanced foundational science solutions. Two, it is not just a matter of GDP growth or better lives. Today, in a world increasingly shaped by geo-strategy, a nation’s technological edge directly influences its global standing.
“We want to be the third largest economy,” said entrepreneur Saurabh Srivastava, a recipient of the Padma Shri award. “We want to be among the three major powers. But how do you have a say in what goes on in the world? You have to have something that people want. We don’t have a choice. The government recognises it.”
The great wall of China
Building deep tech muscle requires trial-and-error experimentation with advanced science, massive investment, long years of research, and multiple stages of trials, testing and approvals. But the results can be earth-shaking. A classic example is how the Chinese company DeepSeek upended the AI race with its cheap and indigenous model. The question is no longer whether India would attempt something similar with deep tech, but whether the country can.
Deep tech is a high-stakes field. In a recent study by the Australian Strategic Policy Institute (ASPI), China was ahead in 37 of 44 critical technologies in crucial fields like defence, space, robotics, energy, biotechnology, AI and quantum computing.
“Western democracies are losing the global technological competition [to develop] and control the world’s most important technologies, including those that don’t yet exist,” ASPI noted. “China has built the foundations to position itself as the world’s leading science and technology superpower, by establishing a sometimes stunning lead in high-impact research across the majority of critical and emerging technology domains.”
Countries are also building barriers by withholding materials critical to deep tech advancement—all in an effort to thwart rivals. From the US regulating access to high-power GPUs (graphics processing units, needed for advanced computing), to China restricting access to magnets and rare earth elements needed for manufacturing clean energy solutions like batteries, this trend has shifted into top gear.
“[The] geo-strategic issues coming into play… [are] showing everybody the importance of having certain technologies yourself. You can’t really depend on other countries. That means we need to focus on a lot of areas that are strategically important to us,” said C.S. Murali, chairman (entrepreneurship cell), Foundation for Science, Innovation and Development at Indian Institute of Science (IISc) in Bengaluru.
The elephant dances… slowly
This “can’t-do-without” scenario contrasts directly with India’s success on the software-as-a-service (SaaS) side. The nation has been super successful in coding quick-fixes for clients across the globe, or by spawning nifty mobile apps that serve its vast domestic market. But deep tech is on a whole different level.
“Unlike software companies where two people can get together in a garage [with] laptops and start building something, many deep tech companies are building physical products, which require, therefore, hardware and software. You need more money to acquire components, build prototypes and, of course, hire engineers,” said Srivastava.
Saurabh Srivastava
Getting funding, especially in the initial stages, can be tough. “It is one of the biggest bottlenecks,” said Sunil Gupta, CEO of QNu Labs, a quantum cybersecurity company in Bengaluru. “Deep tech ventures are asset-heavy, intellectual property-focused, and take years to mature.”
Add to that India’s abysmal track record in research & development. India’s R&D investment is very low—less than 0.7 per cent of GDP. China’s is 2.4 per cent and the US is at 3.5 per cent, with GDPs multiple times bigger than that of India.
“It’s an industry problem,” Srivastava said. “Globally [if one takes 3 per cent of GDP as the average R&D spending], 2 percentage points come from industry and 1 percentage point comes from the government. But in India, two-thirds come from the government. Our industry is not interested in R&D.”
India Inc is yet to get over the pre-liberalisation mindset of importing innovations. “They are used to a lazy model of success. It’s not that they don’t have the money. It’s a mindset problem,” Srivastava said. “Globally, if companies don’t invest in R&D, they die. In India, because we have had a protected market, you can afford to not invest in R&D, and do well.”
Add to that the acute lack of investor support faced by many deep tech startups. Big money chasing quick solutions power India’s ‘unicorn’ success in consumer apps. But deep-tech takes long gestation periods, running into many years in most cases, and the results can also be hard to predict.
“Unfortunately, even the so-called early-stage investors don’t really come in very early. They would like to see an almost-ready product,” said Srivastava.
Laina Emmanuel and Rimjhim Agarwal, founders of BrainsightAI that crunches neurological diagnosis data using AI, said they struggled in the initial years because of lack of backers.
“India is not well-known for taking on early-stage bets,” Emmanuel said. “We really had to struggle in the first few years spending our own seed money. We did tiny, tiny things. We actually got somebody on board whose only job was to write a lot of grant applications.”
In fact, when they finally got a government grant, it took almost two years for the money to be released. Contrast this with China, which has an ‘all in’ philosophy. “The state support in China for startups is enormous,” said Murali of IISc Bengaluru. “Once they decide to pursue something—be it AI or drones—they throw a lot of money and support. As a result, a very large number of entrepreneurs jump into the mix.”
The silver lining
Things are changing. The Union government has decided to finally approve a scheme worth Rs1 lakh crore to fund research, development and innovation in critical sectors like semiconductors, AI, green energy and biotech. The new scheme envisages long-term, low-interest loans to startups and other private players to invest in core research.
“To truly lead, India must invest aggressively in R&D, strengthen industry-academia linkages, and develop risk capital mechanisms tailored for deep tech,” said Kanchan Ray, chief technology officer at Nagarro, a Germany-based digital engineering company. “Government initiatives are in the right direction, but they must be complemented by a sustained focus on IP creation, talent development and infrastructure.”
India’s fledgling space-tech field could be a guiding light for the way forward. Space has for long been a government domain, but it did not stop entrepreneurs from setting up startups to test out satellite design and find solutions for space programmes. Their aim: cater to a global market. “The government of India never told private companies not to build satellites,” said Sanjay Nekkenti, who set up Dhruva Space in Hyderabad when India’s space business was fully government controlled. “I don’t think many gave a thought to the opportunity in building satellites for the global market.”
Today, IN-SPACe (the Indian National Space Promotion and Authorisation Centre), the commercial arm of India’s space department, is driving private sector participation in space technology. The result: revenues are expected to cross $44 billion in a few years.
Into the deep
India may lag China and the US, but it has fared decently. It is in the top five in many specialisations—super capacitors and advanced aircraft engines, to name just two.
Thankfully, investors are now more interested. Padmaja Ruparel, co-founder of India Angel Network, one of India’s leading early-stage investor groups, said her firm’s investing graph was changing. “Pre-pandemic, we used to get a lot of software products. While that has not gone down, there has been a dramatic change: we are getting a lot of AI and space-tech companies. We are getting [investment opportunities] in rare earth materials, biotech, semiconductor and manufacturing tech. I don’t think we would have seen any of these sectors five years ago. That is a big change.”
The increasing shift of global capability centres to India also augurs well for India’s deep tech ambitions. “Driven by access to top tech talent, cost advantages, and a thriving startup and innovation landscape, most Fortune 500 companies are shifting their R&D to India. But to sustain this momentum, we must go beyond scale and focus on deep IP creation, long-term R&D investment, and stronger academia-industry collaboration,” said Sanjeev Azad, chief technology officer and global innovation head at GlobalLogic, a digital engineering multinational.
The ball has certainly started rolling, but many stakeholders would need to step up to help India score. Especially industry needs to spend more on research and ensure funding to entrepreneurs with big ideas. “To accelerate this momentum, we [need to] align our education, policy and industry priorities,” said Srinivas Reddy, managing director (India) of US-based AI-solutions company EPAM. “India would then not just contribute to the next wave of tech transformation, but also redefine and shape it.”