Geopolitics & India’s rebirth in deep tech

It takes time for a nation to come together, especially when the State, the bureaucracy, the scholarly class and the business class have to work together. It took a hundred-plus years for the USA to make such a transition.

It took China three generations to reemerge from Japanese occupation and the Stalinist culture of Mao.

Only when the four elements figured out how to work together and place emphasis on science, manufacturing and intellectual property did the USA and China prosper. And this is finally happening in India. For 79 years, India was all about bureaucratic authority, institutional intellectualism, political obfuscation and corporate diplomacy. Now they have all come together for India’s semiconductor ecosystem to rise. Will it work? This article will tell you how things have come together, but you have to make up your own mind on whether they will work together.

If you listened to or watched or read the news in September 2026, everything was about Semicon India. A total of $13 billion was committed by the government alone to enable innovation in the hardware ecosystem.

World leaders such as Carl Pei, the founder of OnePlus and Nothing, took to ‘X’ and posted that India’s future was in manufacturing and even stated that manufacturing alone could make India rise. And rise for what reason, one may ask?

Of course, economics would tell you that manufacturing economies uplift the per capita income of a nation, which will enable people to move away from poverty, so that one day they may own a house and enjoy three square meals a day. World leaders continue to believe that India can become the next Japan, Korea and China if it embraces manufacturing. But what we must understand is that India needs manufacturing for its sovereignty, so that no powerful nation can dictate its future. We need not become a Japan or a Korea, and that’s where China was different. The former became entrenched in American economic and military superpower, while China navigated American supremacy to position itself as a superpower.

At the heart of all manufacturing is your semiconductor ecosystem. Control of the hardware ecosystem, either with deft geopolitical moves over the control of resources or by building the entire ecosystem inside a country — upstream and downstream — decides whether a nation can indeed become a superpower in the long run.

Without this ideal, India can be prevented from taking control of its destiny in the AI era. Mind you, the digital infrastructure, rails to data, for the internet is controlled by the USA and China. Geopolitics has realigned globalization, and at the centre of it is neo-nationalism, which is creeping up in the USA and around the world.

Countries that are leading the intellectual property race control the hardware and software necessary for global corporations to scale across the world while realigning the world order of globalization.

India only has control because it produces food for its population. If it has to move up the geopolitical value chain, in controlling semiconductors, it has to make self-reliance an absolute reality. To unfurl this argument, we must know the history of India’s manufacturing and whether Indian science can pull it off.

Independence from colonial extraction

Let us start with the geopolitics in 1947 and why Atmanirbharta (self-reliance) was a necessity back then, before we take a trip to the globalization of the nineties, thanks to IT services, and the journey back to self-reliance with the new world order.

Seventy-nine years ago, the founding fathers of India decided that the country would not be subject to colonial-style corporate extraction ever again.

The East India Company, along with other global trading corporations and the English Crown, had dismantled India down to the very last resource, thereby subjecting a majority of the people of India to living in squalor.

So, knowing its history, the Indian leaders blended a Soviet communist-style industrial outfit for its engineers while guided by a British/American-styled bureaucratic setup. These three employment sources, namely the governing class, the bureaucratic class and the scholarly class, were drawn from democratic elections, civil service exams and technical education.

This trio, the founding fathers believed, would rise above the ordinary and govern the nation towards self-reliance. Unfortunately, it did not go as planned because being commercially minded was considered a mistake. Entrepreneurs had to hide their show of wealth.

To make it worse, the separation of Centre and State created a massive rift in the ideals of governing India’s business outfits. As a matter of fact, India liked science, not the commercialization of it.

Indian businessmen needed the support of the politicians, or the governing classes, and the State’s bureaucratic machinery to further their industry. It was not a marriage made in heaven; it was a political marriage to ensure the continuity of these two classes while we continue to find out the identity of India.

Nevertheless, the Indian Central government continued to announce grandiose plans, which it does to this day, to make India self-reliant. Yet, as contradictions go, India was way ahead in its semiconductor journey until 1989, way ahead of other countries, when it threw it all away in a heartbeat.

For example, you would not believe that India had a semiconductor plant way before Taiwan. More on this as you read the story.

In the 1950s, India began to invest in heavy industries under the ownership of companies called Public Sector Units. These units were responsible for engineering large dams, power industries, mining, oil refining, aeronautics, defence equipment and much more.

If you got this far, you would be wondering what of the private industrial and merchant class, which was extremely influential before 1947.

Let us say this: the narratives, at the time, were against private entrepreneurs. They were only tolerated to an extent, and the Indian State was the proverbial “leviathan.”

The government had strangely discouraged entrepreneurship in manufacturing unless the said entrepreneur or business could cap revenues and production.

With all this high-handedness, the Bombay Club and the Madras Club continued to build. Even to this day, names such as the Tata Group, Britannia Industries, Mahindras, and Bajaj are part of business folklore because they survived an idealistically minded government at the time. Similarly, in southern India, it was the Murugappa Group, the TVS Group and the Amalgamations Group that continued to make headway.

Just to tell you how difficult it was for Indian businessmen:

The government passed the Industrial Policy Resolution of 1956 and the MRTP Act of 1969, which were institutionalized as the “License Raj,” where everything was capped to a license. Large companies faced exorbitant corporate tax rates (60 percent plus), and there was zero access to new technology from abroad. The good thing was that they could work with government-built scientific institutions to indigenize any Western technology.

Although these companies lived in a high-tariff regime that shut out international competition and had no access to international technology, they were able to indigenize processes with State institutions. These companies managed the State-level and Central-level politics to continue their hold on manufacturing.

It is odd to explain nations at the time: India believed that science was not meant for commercialization, while in America it was another story.

Let us take a detour to the USA.

Inventing chips — changing humanity

When India was trying to erase the worst of capitalism and punish its private industries for the mistakes committed by their Western forebears, the USA was becoming a superpower in science.

In 1947, the USA became the powerhouse of silicon inventions when companies like AT&T began to license transistor patents, thanks to government intervention to prevent a monopoly. This move created several breakthroughs in developing the first commercial transistors by a team of engineers from Texas Instruments.

From there on, there was no stopping innovation as the first integrated circuits were invented, which led to single-transistor chips being made for single-purpose use, such as space missions. Then by 1971, Intel launched the first single-chip microprocessor, with multiple transistors, which could be deployed for multiple applications with programming. This was a breakthrough for humanity: instead of having multiple integrated circuits operating different functions, you could now do it with one chip.

While the USA was on this journey for global dominance, India was on a trajectory of its own, which was to make products in India for Indians, creating manufacturers of consumer goods.

In the 1970s, the Indian government began to ease regulations on entrepreneurship and attracted entrepreneurs who could build for local markets on a small scale. Again, the irony is that these entrepreneurs could only remain small, and they were not allowed to grow beyond a certain revenue cap or were penalized for increasing revenues.

This sets the context as to why this journey mattered for Indian SMBs.

The interesting story of Indian SMBs, defence and space

Around 1971, while Intel had released its multi-transistor chip, the government of India decided to earmark 800 products to be specifically manufactured by small businesses.

Products such as food items, consumer goods, stationery, textiles, and light engineering suddenly saw an influx of new-age entrepreneurs. These entrepreneurs became the bedrock of the present Indian manufacturing story — some of them went on to build components for the automobile and aerospace industries as time went by. Interestingly, at the same time, all public works were also executed by private contractors who would later go on to possess the capabilities to build large data centres. But more on this in another story.

Even as Indian private manufacturing continued into the eighties with mechanical clunkiness, it is interesting to note that the government also created companies that worked in deep science.

It connected the intellectual might of the Indian Institute of Science, set up in 1909, to enable public sector units like ISRO, DRDO, BEL and defence.

These organizations were importing commercial off-the-shelf chips from American companies to build systems for communication — defence radio sets and radars — and early missile guidance systems.

Several of them were based on single-transistor modules and also utilized multi-transistor processor chips, which were indigenized for Indian avionics (MiG-21), atomic energy instrumentation, and silicon rectifiers in heavy industry. For a young nation, which was embracing elements of benevolent socialism and controlled capitalism, much to the chagrin of the Western economies, it managed to chug along with its science and manufacturing.

Its leaders manoeuvred global pressures with ideals; to make it specific, the non-aligned ideal was met with laughter from the West, which, when it mattered, curbed Indian imports of chips. India was, in essence, a nation dependent on the Western superpowers, so to speak.

In that gloom there was good news: there were technology transfers from France, the British, the Dutch, the USA and the Soviets. This momentum was with India in the eighties — not China and definitely not Taiwan — as the governments and the bureaucrats of the time pushed for a semiconductor fabrication factory to support Indian industries.

And thanks to another technology partnership with an American company, called American Microsystems, with whom government-employed Indian engineers went on to build 5-micron size processors, now the stage was set for the launch of the Indian Semiconductor Complex in 1984, a whole four years ahead of any Taiwanese company, especially the famed TSMC.

The world was watching if India could indeed become a Japan or a Korea even at that time.

The Central government, with the support of the Punjab government, set up the Semiconductor Industrial Complex, in Mohali, to make this complex the beating heart of India’s deep tech journey.

Unfortunately for India, this plant burned down to the ground in 1989, and it took eight years to become operational again. And just to tell you readers that in this short time frame, when the complex was not operational, India had changed its entire approach to business; it decided to go the services route and change its course forever with the belief that American-led globalization would last forever.

India opened its markets in 1991, and it became overtly capitalistic because the internet created new business models. India could now serve as America’s back office. It traded its intellectual sovereignty to grow its middle class based on a consumption-led economy, which at some point became an albatross around its neck because all core science-led assets belong to the West and China.

Sadly, over the next three decades, India had no indigenization of hardware other than for public sector companies in defence and space.

People and nations go where the money is, and with information technology, wealth could be built in under eight years, and the real estate boom in house ownership deterred all governments from supporting deep science.

Nonetheless, the government allowed the likes of Cadence, Synopsys, Intel, Texas Instruments, Freescale and Broadcom to set up design services in India, which really puts things in perspective as offshoring of business processes from America to India, with information technology, worked, and it generated wealth.

Capital then began to support ideas that could create wealth in five years, and therefore software became the calling card for Indian entrepreneurs. Hence, the hardware industry would coin the term “Patient Capital” to forever talk about a time when capital would be patient to back them.

The lack of such capital starved entrepreneurs in the deep science business, who floated IT services companies in order to stay relevant.

A majority of these entrepreneurs left to work in the USA and Europe, which continued to accommodate Indians with a mind for deep science. Even the Indian Semiconductor Complex was transferred to ISRO in order to keep it operational.

By 2014, things came full circle for India’s deep tech because the government woke up to oil bills and the import bill on all semiconductor equipment in the form of telecom equipment, power management systems, mobile phones, industrial systems, intelligent automotive modules, and much more.

Annually, the country was spending around $7 billion on semiconductors alone. India therefore had to modernize, and this time the government had to bring back the self-reliance narrative when the USA, Russia and China decided to unbalance the world order.

The move was imminent — the Indian government had to bring a breath of fresh air to the silent science community building away in silent rooms across the country.

Now, to do such a thing, we must remember that the government of India never abandoned its deep tech in the nineties, and that’s the sole reason it was able to roll back to self-reliance as a narrative. To come back to the present, we have to revisit the past once again.

The quiet years

In 1996, software was clearly the future, yet the government continued to support science. The Central government had launched the Technology Development Board, which provided interest-free loans and capital to entrepreneurs who pursued deep tech. From this scheme emerged two large businesses called Bharat Biotech and Shantha Biotechnics, who go about ever so quietly with their job. This has perhaps been the secret sauce of Indian science. It has been quiet in building its foundations.

The government continued its efforts well into the new millennium. In the year 2000, it launched the aptly titled “New Millennium Indian Technology Leadership Initiative” under the guidance of the Council of Scientific and Industrial Research to support deep tech businesses.

Unfortunately, things were not moving because, at the time, IIT professors were not allowed to commercialize their research, and there weren’t many commercial businesses ready to work with science. That said, this initiative benefited the government in building bioinformatics software and molecular therapies, which continue to support our pharma and biotechnology businesses.

The government realized it had to change, and in the period between 2009 and 2019, the Central government allowed professors of IITs to hold minority stakes in companies or take a three-year sabbatical to build a deep tech business.

In the same time frame, the government even set up the Biotechnology Industry Research Assistance Council and the Innovations for Defence Excellence as platforms that would support early-stage startups rooted in deep science.

By 2015, by positioning India as a startup nation, the government managed to influence private capital to look at India as a deep tech hub, also because of all the structural changes that had happened over the last eighty years.

Hopefully, patient capital and good government can now take India to its semiconductor success. It’s from this long context that we must explore where Indian innovation stands today, as showcased at Semicon India.

30 years on — the waking moment for deep tech

With the success of IT services, entrepreneurs like Narayana Murthy, Nandan Nilekani and Kris Gopalakrishnan began to celebrate Indian research and science. Although, over the last decade, the consumer-based startup ecosystem had blossomed — the likes of Flipkart and Paytm had become unicorns — it was no coincidence that a small portion of the VC community began to view all the changes happening in Indian science.

Many deep-science-loving entrepreneurs who had made money in IT services believed that Indian science could also go global. They became VCs in the process.

Coincidentally, interest was growing in academia; the case in point was IIT Madras, which alone has incubated over 500 deep tech startups.

Apart from the IITs, several deep tech funds cropped up — Ideaspring Capital, Ankur Capital, Speciale Invest and several others began funding science papers to go commercial. And with the success of deep tech startups like Ather, ideaForge, Agnikul Cosmos and Skyroot, Indian science was back on track.

To bring the reader back to the context — with the USA, China and Russia setting narratives of a new global order, India had to make moves to indigenize the hardware, once again, for its sovereignty and relevance in the digital world.

This was the primary reason why India wanted to build its entire chip-building stack — manufacturing, supply chain, design and final integration — within its boundaries. You don’t want to worry about having a fab or fabless argument because India wants to do both to support sovereignty of its capabilities in the AI, space and defence era while globalization is realigning itself to a new world order, where the USA is betting heavily on AI while it stops shipping back-office work to economies like India, and China is showcasing its scientific might.

Therefore, September 2026 is going to be a special month because India has made a definitive call on changing its stance on hardware.

Media reports say that Semicon India garnered over $11 billion in committed private capital, and along with it there was a signing of 56 MoUs and strategic collaborations across the global electronics value chain.

Then, $263 million, or approximately ₹2,170 crore, was deployed by the India Deep Tech Alliance (IDTA) across 56 specific deep-tech startups at the event.

Combined with the above was a massive committed government outlay of ₹1,27,500 crore ($13.5 billion).

So, while the government built highly successful, centralized, state-run ecosystems — in space and defence — out of sheer geopolitical necessity, it missed the semiconductor bus for the last 40 years.

Like we said earlier, the same geopolitics has brought the narrative back to hardware.

Now every major global semiconductor firm — including Intel, NVIDIA, Qualcomm and Texas Instruments — is betting on India once again.

Let’s not forget that the Indian government launched the ₹76,000 crore ($10 billion) India Semiconductor Mission (ISM) in late 2021 to subsidize fabs. This has expanded into the Semicon 2.0 framework, as mentioned earlier, raising total outlays toward ₹1,27,500 crore, or $13.5 billion. Hopefully, this move will secure sovereign control over foundational tech for India.

The Tata Group is constructing India’s first commercial mega-fab in Dholera, Gujarat, alongside an advanced packaging facility in Sanand, partner-backed by Taiwan’s PSMC and Micron Technology. L&T Semiconductor Technologies recently unveiled 40 localized products, showcasing India’s readiness to design and manufacture commercial power platforms like silicon carbide (SiC) MOSFETs natively.

In addition to all of the above, what is unique is India’s willingness to open up the space sector via its IN-SPACe program, which has led to over 200 active startups.

Indian innovation will finally rise; it’s only a matter of will that will make India a powerful player in the new world order. UpStreamVoice, for one, will not want to ever support the “license and inspector raj” era, which benefited very few entrepreneurs, and neither does it want to support the back-office economy.

The new India should be atmanirbhar in every way possible, and only by making Indian science commercially viable will this journey be fruitful.

This has been the missing trick for India: its businesses must figure out how the world can benefit from Indian science and technology. Now that the government and the corporate world have taken eighty years to come together, hopefully they do not create monopolies once again. That’s when India is in deep trouble once again, where science is used for extraction and exclusion rather than progress.

Remember the East India Company — let’s not get there ever. At least that’s what makes us at UpStreamVoice happy. All four working elements of making a nation — the bureaucracy, the government, the scholarly class and the corporate class — have come together to make deep tech work in India.