The world’s most populous country has been shorted by AI.

The first country to be shorted by AI has appeared.

The once-booming Indian stock market is now facing an unexpected collapse. On July 8, the India Nifty IT Index closed at 23,568 points—down 49% from 18 months ago. The combined market value wiped out by ten major IT companies exceeded 1.9 trillion rupees, equal to 60% of India’s fiscal budget for FY2024.

Even stranger is that every major plunge in the Indian stock market corresponds to a major AI move in Silicon Valley:

On February 4, Anthropic rolled out enterprise-grade AI tools; the Indian Nifty IT Index plunged nearly 6% that day, posting its largest single-day drop since March 2020;

On May 12, OpenAI announced it would spend over $4 billion to set up a “front-deployment engineer” team; the index fell another 3.7%, dropping to a three-year low;

Worse than the stock market is the demise of India’s “IT golden age”:

The Times of India reported a detail: a young man from a rural area in Bihar borrowed money to finish an engineering degree. After receiving a TCS offer letter, he waited for a year—only to eventually receive an email rescinding the offer.

A more brutal case occurred in Bengaluru. This March, a 32-year-old software engineer Reddy hanged himself in his apartment. Soon after, his wife, who worked at IBM, also jumped from the 17th floor.

Police investigations found that Reddy, who had previously earned an annual salary of 8 million rupees in the United States (about RMB 570k), stayed unable to find a satisfactory job for a full year after AI caused his unemployment.

A country that built the middle-class dream by writing code is being反噬 by its own most successful industry. This isn’t because Indians aren’t working hard—it’s because the times have changed.

One

Over the past three decades, India’s IT industry has been a textbook example of grassroots come-up.

During the 1999 “Y2K” crisis, Western and European companies desperately needed cheap software engineers to check old code. Indians seized the opportunity with three advantages: good English, low wages, and willingness to stay up late. Afterwards, Indian IT outsourcing giants such as Tata Consultancy Services and Infosys emerged.

Today, IT outsourcing has become an unquestionable pillar industry for India: annual service exports averaging $150 billion, 4 million high-paying jobs, and entire upstream and downstream ecosystems that affect the livelihoods of tens of millions.

For young people in India, a computer degree is the ticket to climb out of the slums. “Study IT, enter a big tech company” has become the standard path to becoming middle class.

For example, as reported by The Economic Times:

Raj Vikramaditya grew up in a slum in Mumbai. He didn’t get into the IITs, and only attended an ordinary engineering college. But he started writing code from his first year. After graduating, he joined Google, earning a salary of 300k rupees per year—while the average salary of graduates from his school at the time was only 40k rupees.

But with AI, everything changed.

Last year, India’s largest private employer, TCS, announced the biggest layoffs in its history—cutting out more than 12k positions. In the first nine months of FY2025-26, TCS alone already reduced headcount by over 25k net.

The most direct reason for the layoffs is falling profits. Last year, Wipro’s large orders doubled, yet total revenue fell 2.3%; Infosys’s net profit has been declining consecutively. A report by the Indian IT Industry Association said the whole industry “is undergoing structural adjustment, with scale and speed beyond expectations.”

The IT hiring market has essentially frozen. The top five Indian IT companies combined only added 17 employees net. Meanwhile, India needs to produce 1.5 million computer-science graduates every year, and unemployment among young people aged 15 to 25 has surged to 40%.

What’s more, UnearthInsight estimates that over the next two to three years, 400k to 500k IT practitioners in India face layoff risk—70% are core staff with 4 to 12 years of experience.

Peak expansion in Tata Consultancy Services and Infosys employee headcount, 2015–2026. Source: Nikkei Asia

With mortgages not fully paid and kids still needing university, India’s older software engineers have had their world collapse.

The chill from layoffs is moving backward along the same path that wealth once spread forward. In Bengaluru and Hyderabad, real estate, car sales, food service, education and training all cooled across the board. In Q1 2026, year-on-year residential sales in India’s major cities fell 13%.

Fund manager Saurabh Mukherjea spotted an even more dangerous signal: large numbers of people anticipating layoffs are rushing to apply for personal loans and mortgages.

This chain reaction triggered by AI may just be beginning.

Two

Why did India’s decades of accumulated “household assets” get overturned overnight?

The root cause is that the essence of India’s IT outsourcing—“pricing by person-month”—is being cut cleanly in half by AI.

In the past, Indians sold programmers to Western clients by headcount and by hours worked, profiting from the spread. The cost of a Bangalore engineer for one month was less than one-fifth of Silicon Valley, making it a bargain for the giants.

But AI has overturned three assumptions at once.

First, do we really need so many junior programmers doing repetitive work? An AI programming tool can complete in a few hours what used to take a whole team several days. When code review, compliance tracking, and test cases can all be handed to AI, the logic of “bulk hiring” no longer holds.

Second, does the “cost advantage” of labor really exist? AI’s marginal cost is almost just electricity; a subscription of dozens of dollars per month can replace a junior engineer’s work for a week. Even if Indian engineers are cheaper, you still have to pay wages, social insurance, and rent office space.

Third, is technical outsourcing still necessary? The essence of outsourcing is “send the work to cheaper places,” but now AI can do the work right next to the customer—no need to fly over half the planet, no need to wait out time zones, no need to communicate requirements across the Pacific.

What’s even more damaging is that before, Indian programmers went to Europe and the U.S. to find jobs—but now that road is also effectively sealed shut.

In September 2025, Trump raised the H-1B visa fee from $5,000 to $100k; in April 2026, the U.S. Congress proposed a bill to pause issuance of new H-1B visas for up to three years.

In addition, countries such as the UK, Canada, and Australia have also tightened visa rules for India.

The New York Times commented: “India built the world’s back-office, but AI is making it shrink.”

This is a subtle cycle: back then, India’s IT replaced Western programmers through a cost advantage; now, what ends it is an even cheaper alternative.

Three

In fact, India is not without ideas for transitioning to AI.

At the start of 2026, New Delhi hosted the largest AI summit in Asia. Open AI’s Altman and Anthropic’s Amodei attended in person, giving plenty of face.

At the summit, Modi claimed that India would be built into a “global AI innovation hub.” The Indian government approved the “India AI Mission” plan, with total investment of 103 billion rupees (about $1.25 billion).

But India’s AI ambitions hit the ground and got “burnt” a bit—like the kind that happens at 50 degrees.

In May, the Indian city of Delhi hit a high of 52.9℃ at one point. In multiple places across Rajasthan, temperatures exceeded 45℃. Data compiled by the International Meteorological Organization showed that among the 100 hottest cities globally, India accounted for 95 of them.

After all, AI data centers are monsters that consume huge amounts of electricity and water and have extremely high cooling requirements. According to a report by India’s Energy and Water Research Institute, at present, more than half of India’s data centers already face temperatures above 35℃ for over 90 days each year.

And with climate warming accelerating, by 2040, nearly 90% of India’s data centers will be threatened by heatwaves.

India’s recent heatwave situation. Photo credit: Zhi Zong Wang

Beyond being choked by its own climate, the bigger issue is India’s poor infrastructure level: water networks and power grids are almost unable to hold up.

On May 21, India’s electricity peak demand set records for four straight days. In rural areas, daily power outages averaged over 10 hours, and even in IT centers like Chennai, night outages of 40 minutes to 1 hour have become normal.

In 2025, India’s data centers used 150 billion liters of water, and the figure is expected to double by 2030. At the same time, India is among the most water-stressed countries in the world—with 18% of the world’s population but only 4% of the world’s freshwater resources.

In addition, India’s AI is extremely dependent on foreign large models, with almost zero innovation at the foundational architecture level.

According to data from India’s Competition Commission, 83% of Indian AI startups focus on application development and mainly use foreign foundation models from companies such as OpenAI and Anthropic; 67% of Indian enterprises remain only at the AI application layer.

Today, while India can produce nearly 20% of global data, it only owns about 3% of global data center capacity. Dependence on Silicon Valley giants is gradually turning Indian firms into “colonies” for data training.

At the AI summit in New Delhi, Modi raised both hands and smiled brightly. But next to him, Altman and Amodei’s expressions looked a bit stiff.

Maybe they saw it long ago: a country lacking independent innovation, unable to even secure basic water and power, doesn’t have what it takes to carry the future of AI.

Four

This large-scale reckoning happening in India reveals an even deeper problem:

The card of the demographic dividend can’t be played anymore in the AI era.

As the world’s most populous country, India’s median age is only 28 years old, and in the next 20 years, tens of millions of young people will enter the labor market every year. But the latest data shows unemployment among young people aged 15 to 25 in India has already surged to 40%.

Too many people is also a burden. What’s more, India still carries heavy institutional constraints.

A report by “World Inequality Lab” shows that India ranks near the top globally in income inequality: the top 10% of income earners take 58% of the country’s national income. Most of them belong to high castes—i.e., the Brahmin group.

In Mumbai, a road divides two worlds—on one side, villa areas with swimming pools; on the other, slums where hundreds share a single water tap. Enrollment rates in universities for high-caste families are close to 100%, and two-thirds can study abroad; while across the country, there are still nearly 300 million illiterate people, and adult literacy is under 50%.

IT outsourcing is one of the few routes in India—one that is not completely dependent on caste and background, and only requires being able to write code—to climb upward. But now, this only path is narrowing.

At bottom, the demographic dividend is a paycheck with an expiration date. What truly determines how far a country can go is upgrading in technology, industry, and social governance—not staying forever in the business of “selling heads.”

India is making up for this lesson in the most painful way.

Source of this article: Zhigu Trends trend

Risk warning and disclaimer terms

        The market is risky; investing requires caution. This article does not constitute personal investment advice, and it does not consider any specific investment objectives, financial conditions, or needs of individual users. Users should consider whether any opinions, viewpoints, or conclusions in this article are consistent with their specific circumstances. Investing based on this is at your own risk.