The Game Which Isn't Played: Why India's Tech Industry Built a Fortune and Lost the Future

A few weeks back, I had written an article on 'Equality in action: How India's DEI programs benefitted everyone'. This is to hit home the point equality is a flowing theme on the ground and really practised for the benefit of everyone when world is looking with suspicion and anxiously worried at 'We, Our' buzzwords. Coincidentally around the same time a Chinese AI assistant called Kimi released its K3 model and it was making waves with parameters comparable to other best frontier models of the world. So I checked Kimi to write an article on equality when China is my favorite subject when propagating equality. Here I am, using Chinese model released at right time to propagate equality which should augur extremely good for equality itself and bring a notion of honesty for the cause, saying India is indeed a land of equality. Trust me, the notion of equality puts so many at ease and peace. I revealed this story to hit home the point that Chinese development of frontier AI models augurs extremely good for the world and required for the world. Indeed, Kimi has written the article beautifully, on par with any other. But, where is the Indian push, effort to create something very relevant and meaningful for itself and for the world, even after working with screens and screens alone for the last thirty years. These timely creations in an area of strength are yet to be realized, alarmingly bad, even after many years of mastering the art. It is time, we have an iota of attitude to change the landscape or lay behind forever. The following is bitter and hard truth we need to recognize and alter it in our lifetimes. 

For three decades, Indian IT has been the crown jewel of the economy — a $250 billion behemoth that turned Bengaluru into a global back office. TCS, Infosys, Wipro, and HCLTech became household names not by inventing the future, but by maintaining someone else's present. They built the plumbing of the digital world without ever building the house. And now, as artificial intelligence rewrites the rules of technology, India finds itself with the largest IT workforce on the planet — and not a single globally competitive foundation model to its name.

This is not a funding problem. It is not a talent problem. It is a *comfort* problem. And it is time the nation stopped celebrating mediocrity dressed in nationalist garb.

### The Services Trap: Execution Without Invention

India's IT miracle was built on labor arbitrage — brilliant engineers at fraction-of-Western-costs, delivering reliability, not originality. The business model rewarded predictability: meet the client's spec, deliver on time, invoice by the hour. As one analysis put it, "Employees work on other companies' roadmaps, not their own... Success is measured by delivery speed and compliance, not originality." 

This created a generation of institutionalized engineers — technically superb, creatively stunted. Seven years inside an Infosys or Wipro, and a mind trained to execute specifications becomes ill-equipped to write them. The industry did not just fail to innovate; it actively disincentivized it. Why risk a product when services brought steady quarterly revenue? Why fund R&D when clients wanted faster turnarounds? Indian IT firms optimized for today, while Amazon spent years losing money to build moats for tomorrow. 

The result? India holds just **0.37% of global AI patents** compared to China's 70%.  That is not a gap. That is an abyss.

### The Scarcity Mindset: A Culture That Fears Failure

Beneath the business model lies something more insidious — a cultural architecture that punishes risk. India's middle class, forged in scarcity, taught its children to optimize for safety: get the stable job, save early, avoid unnecessary risk. "Innovation requires experimentation, loss, boldness, and the belief that failure is not the end — but part of the journey."  In most Indian IT firms, hierarchy runs deep, failure is frowned upon, and "innovation" is managed through rigid dashboards and "innovation councils" driven by compliance, not curiosity. 

Compare this with Silicon Valley, where failing startups are called "learning labs" and engineers jump between product ideas with freedom. In India, a failed venture brings social stigma. A stable job at a services giant brings family pride. We did not just export our best talent to Google and Microsoft — we exported our risk-takers, while keeping the risk-averse at home to man the support desks. 

### The Nationalist Delusion: Confusing Origin With Excellence

When DeepSeek shocked the world, India's response was predictable — not introspection, but a rush to claim parity. At the India AI Impact Summit 2026, Sarvam AI unveiled foundation models with nationalist fanfare. Union Minister Ashwini Vaishnaw called it proof that Indian engineers could do with "frugal resources" what others did with billions. Twitter lit up with patriotic fervor. Commentators called it India's DeepSeek moment.

But here is what the cheerleaders missed: **DeepSeek did not succeed because it was Chinese. It succeeded because it was technically excellent, radically open, and globally accessible at near-zero cost.** The "made in China" label was incidental to its adoption, not the reason for it. India's current discourse, however well-intentioned, is conflating national origin with product-market fit. 

We have seen this movie before. Koo was launched as India's answer to Twitter, backed by Tiger Global and Accel, endorsed by Union ministers, valued at $275 million — and dead by 2024 because it was a clone without meaningful innovation.  Krutrim, India's first AI unicorn, has laid off over 200 employees, lost senior executives, and recorded roughly 100,000 downloads while ChatGPT boasts 110 million Indian users. When Krutrim announced it would host Meta's Llama models on Indian servers, an AI founder publicly remarked that "an intern having a good PC with GPU can do it in their bedroom." 

This is the pattern India keeps repeating: launch with nationalist framing, ride the media wave, watch users drift back to the incumbent when quality gaps appear, then blame "funding winter" for what was actually a failure of substance. 

### The Education-Industry Chasm

The rot begins early. India's engineering colleges teach outdated curricula, rarely aligned with real-world problems or future tech trends. Collaboration between companies and professors is minimal. Research is underfunded and ignored.  While Stanford birthed Google and NVIDIA, and Harvard spawned Facebook, India's IITs — despite their brand value — have produced remarkably few globally consequential product companies. The obsession with the "IITian" brand has become a substitute for actual output. As one critic noted decades ago, "If an IITian starts a paan shop, the heading goes, 'The IITian left his cushy job to start a paan shop...' The media is only feeding into our own obsessions." 

We don't lack brains. We lack ecosystems that let brains build instead of bill.

### The "Frugal" Consolation Prize

India's AI narrative has retreated into a comforting myth: we don't need frontier models, we need "frugal AI" for Bharat. Sarvam AI and others are building competent Indic-language models, optimizing for voice latency in Hindi, reducing token costs for regional languages. This is valuable work. But let us be honest about what it is — **a consolation prize**. When you cannot compete on the frontier, you redefine the frontier as "frugal." When you cannot build GPT-4, you celebrate being "cost-effective."

There is dignity in serving local needs. There is no dignity in pretending that local optimization absolves you of global ambition. DeepSeek didn't win by being frugal. It won by being *better*. OpenAI didn't capture the world by cutting costs for non-English speakers. It captured the world by pushing the boundaries of what machines could do.

### What Went Wrong? Everything That Mattered.

India's IT industry did not fail because of a single policy or a bad year. It failed because:

- **It chose servitude over sovereignty.** We became the world's back office and mistook that for technological leadership.

- **It institutionalized risk aversion.** A culture that treats failure as sin cannot produce invention.

- **It confused brands with building.** IITs and IT companies became status symbols while actual product creation languished.

- **It celebrated the wrong metrics.** Revenue per employee, headcount growth, and quarterly margins became the gods, while R&D spend and patent generation were treated as expenses.

- **It exported its courage.** The bravest Indian engineers left. The cautious stayed, climbed hierarchies, and optimized spreadsheets.

### The Question Before the Nation

Here is the uncomfortable truth: India's techies sit in front of screens for twelve hours a day, but most are not creating — they are servicing. They are debugging someone else's code, attending someone else's standup, optimizing someone else's margin. The industry has built a comfortable cage, and an entire generation has learned to call the cage home.

The world does not need India's "frugal AI" as a charity case. The world needs India to stop making excuses. China built DeepSeek under sanctions. The US built OpenAI with billions. India built TCS with millions of engineers — and still has nothing globally indispensable to show for it.

This is not about hating the services industry. It employed millions, lifted families out of poverty, and changed India's global image. But gratitude for the past cannot become justification for stagnation in the present. No nation in history became great by optimizing someone else's roadmap forever.

The AI moment is a reckoning. The old model — body-shopping dressed as technology — is dying. Either India builds products that the world *needs*, or it settles for being the world's permanent intern. The choice is not economic. It is existential.

**Stop celebrating comfort. Start building worth.**