AI Disrupts India's $2 Trillion IT Outsourcing Industry

On June 3, India's IT sector crashed.

TCS dropped 9%, Infosys fell 4.3%, and Wipro declined 3.7%. The Indian IT index slid 5.8% in a single day, marking its largest one-day drop in four months.

Not earnings surprises, not sudden policy changes—only two words have caused market panic: AI.

Over the past year, the world’s attention has been focused on the United States and China—with OpenAI, Anthropic, and DeepSeek taking turns in the spotlight. But the first target of AI’s blade was India.

If AI can truly write code, conduct testing, write documentation, and handle customer service, what will happen to the countries in the world most dependent on "selling programmers"?

India is the answer.

Several industry insiders told PenDao: This transformation suddenly accelerated in the second half of 2025, marked by the emergence of Agentic AI. AI can now handle 70% to 80% of the work required to build a SaaS software.

Kouding Intelligence, an AI programming company that has completed three funding rounds, told Pencil Road: "The impact of AI programming on the IT outsourcing economy is not just an influence—it could be a 'killing'."

An industry that has sustained India for 30 years.

Many people don’t realize that India’s most profitable industry isn’t manufacturing or the internet—it’s IT outsourcing.

According to data from the National Association of Software and Services Companies (NASSCOM), India’s technology industry generated total revenues of approximately $282.6 billion (about ¥2 trillion) in fiscal year 2025, with IT services revenue reaching around $137 billion, accounting for nearly half of the entire industry.

More importantly, exports. In the fiscal year 2025, India's tech industry generated $224 billion in export revenue, accounting for nearly 80% of total revenue.

In simple terms, one of India's most profitable businesses is working for companies in the United States and Europe.

Luan Tian, CEO of Kedi Chuhai, cut to the chase: "Software outsourcing is essentially 'selling people'—charged by headcount or hours, much like the construction industry."

Over the past 30 years, India has nearly rewritten its national destiny through this industry. U.S. companies needed software developed, European banks needed systems maintained, and global Fortune 500 firms needed digital transformation. What did they do? They outsourced the work to India, which then organized thousands of engineers to take on these tasks.

This created a classic pattern: more customers, more projects, more engineers, higher revenue.

Today, India's leading IT service companies are representatives of this model.

In fiscal year 2025, TCS achieved annual revenue exceeding $30 billion with nearly 600,000 employees; Infosys recorded annual revenue of approximately $19.3 billion with over 320,000 employees; Wipro achieved annual revenue of approximately $10.5 billion with over 230,000 employees.

These three companies alone employ over 1.15 million people.

Revenue and Employee Size of India's Big Three IT Companies (Fiscal Year 2025) Source: NASSCOM

More importantly, the growth logic of these companies has been highly consistent over the long term: hire more engineers, take on more projects, and generate more revenue. For decades, a key metric used by capital markets to evaluate Indian IT companies was not even their AI capabilities, but their number of employees.

According to Reuters data, India's IT industry has reached a scale of approximately $283 billion. Many international investors even refer to India as the "back office of the world."

But this logic is changing. A key factor related to AI: the IT outsourcing business is being disrupted by AI-powered programming.

Su Wen said: "New technologies rarely kill directly in existing markets. More commonly, they exclude you entirely from emerging markets—like the player who used to stand under the basketball hoop to catch rebounds. Does that position even exist anymore? No matter how well you catch, it doesn’t matter."

Therefore, capital markets have begun to worry about one thing.

In February this year, India's IT sector lost $22.5 billion (approximately RMB 160 billion) in market value within a week. At the time, many thought the market reaction was excessive. But by June, panic resurfaced.

On June 3, India's IT index plunged 5.8% in a single day, marking its largest drop in four months; TCS, India's largest software exporter, fell 9%, Infosys dropped 4.3%, and Wipro declined 3.7%.

TCS Headquarters Building | Source: Forbes India

More notably, this is not an isolated incident. As of early June, India’s IT index has fallen 22% since 2026, and declined 26% throughout the entire year of 2025. In other words, this once-leading sector that drove India’s economic growth has become one of the worst-performing industries for two consecutive years.

India's IT index has plunged for two consecutive years. Source: Economic Times

The reason is simple: more and more institutional investors are realizing that AI is replacing exactly India’s core industries—such as coding, software testing, operations support, documentation, and customer service. These roles once required large teams of engineers, but an increasing number of companies are now turning to AI to handle them.

Even more alarming: the capital market is not worried about "all programmers losing their jobs," but rather that India's most profitable business model will fail.

The old logic was: a client wants a project, so an Indian company sends 100 people to earn the revenue from 100 workers. In the future, it may become: a client wants a project, AI completes 80% of the work, and only 20 people are needed.

Kouding Intelligence, an AI programming company that has completed three funding rounds, provided a set of data to Pencil Road: tasks that previously required a development team of 100 people can now be accomplished by just 2-3 individuals; the development cost for an e-commerce website, which used to range from hundreds of thousands to millions of dollars, can now be reduced to just $6–8.

Even more alarming is the average transaction value. "The average transaction value for software development companies could decline by 70% to 90," Su Wen told PenDao.

Brokerage reports show that the overall net profit margin of the software outsourcing industry has dropped from nearly 10% to approximately 0.1% (not limited to India). This means that AI is severely compressing the profit margins of Indian IT companies.

The global perspective is even more striking.

According to Mordor Intelligence, the global IT outsourcing market size is projected to reach approximately $618 billion in 2025. Of this, about 40% to 60%, or roughly $250 billion to $450 billion—equivalent to approximately RMB 3 trillion—is at risk of being directly replaced or significantly price-competed by AI.

Global IT Outsourcing Market AI Replacement Risk (2025) Source: Mordor Intelligence

For a $280 billion industry, this is a nuclear-level alert.

A more dangerous signal has appeared.

A drop in stock price alone isn't significant. What truly matters is hiring. More direct changes have already appeared in the headcount of leading companies.

Su Wen's assessment is more aggressive: "Reducing engineering staff by a factor of 20 is the bare minimum."

Senior programmer Mei, with over a decade of experience at major tech companies, believes: "The future trend is a 10:1 compression. A project team of two to three thousand people may eventually need only two to three hundred."

In the fiscal year 2025, TCS, India's largest IT services company, had approximately 607,000 employees, a decrease of about 13,000 from the previous fiscal year. Infosys had approximately 324,000 employees, a decrease of about 15,000 year-over-year.

This is a rare phenomenon in India's IT industry over the past several years. For the last three decades, the number of employees at these companies has almost always increased. Growth was the norm; contraction was the exception. Today, this 30-year upward trend is turning downward.

ANSR's founder and CEO, Lalit Ahuja, stated directly: "There is a cautious sentiment in the market, and companies are reducing hiring."

The hiring market across India’s tech industry is sharply contracting. As of June 2026, active tech job openings in India have fallen to 93,000—the lowest level in 28 months. Job openings for tech roles requiring less than two years of experience plunged 44% year-over-year—nearly half of all entry-level positions have disappeared.

Su Wen explained the underlying logic: "Developments with a complexity score below 4 can be fully replaced. What once required a team of 100 people now only needs 2 to 3."

In past growth cycles, project expansion often meant increased hiring; now, revenue growth is increasingly decoupling from employee growth.

The biggest concern for India’s tech companies used to be: not enough people. Today, they’re starting to wonder: are there too many?

When leading companies like TCS and Infosys simultaneously begin "downsizing," the direction of an era has changed.

Why is India more dangerous than others?

Because India, hit by AI, is not an edge industry but a core industry.

Here’s a simple example: If AI impacts an e-commerce company, the effect is limited. If AI impacts the advertising industry, the effect is still limited. But India is different—IT services are one of India’s most important export industries. India’s IT sector has generated total revenues exceeding $315 billion, accounting for more than 7% of India’s GDP and employing over 6 million people. Behind these 6 million jobs are 6 million families and the livelihoods of tens of millions of people.

More importantly, this is not an isolated industry. IT outsourcing supports India’s training sector, real estate market (office and residential buildings in Bangalore and Hyderabad), service industry, and education sector. Each IT job generates at least three to five related jobs. This means that the impact of AI on the IT industry could ultimately affect the employment ecosystem of 20 to 30 million people in India.

Meanwhile, India itself faces a more pressing practical issue: according to Reuters data, the urban youth unemployment rate remains as high as 13.6%.

Many young people are already searching for jobs. The unemployment rate for Indian university graduates has surged to 29.1%, with 40% of graduates under 25 unable to find employment. Over 1.5 million computer science graduates enter the job market each year, but only 42.6% meet industry employability standards.

Today, AI is once again putting pressure on employment rates. Employment pressure, skill mismatches, and AI substitution are not three separate issues—they form a mutually reinforcing death spiral.

Su Wen spoke harshly: "The new market has passed you by. You didn't lose to competitors—you were eliminated by technology."

Everest Group analysts bluntly stated: "AI will no longer require L1 and L2 engineers." Yet these L1 and L2 engineers form the foundation of India's IT industry, serving as the first step for 1.5 million computer graduates each year and the gateway to a better life for millions of families.

Of course, India could also become the biggest winner.

The story isn't over. Because India has another set of data.

According to the latest BCG report, "AI at Work 2026," India has become one of the most active countries in AI adoption, ranking first globally in AI usage among employees and managers.

Another survey by ADP, titled "People at Work 2026," covering 34 countries, found that 80% of Indian employees use AI tools multiple times per week, and 41% use AI daily—compared to global averages of 50% and 20%, respectively.

In other words, globally, about 1 in every 5 people uses AI daily; in India, about 2 in every 5 people use AI daily.

India vs. Global AI Adoption Rates Source: BCG/ADP 2026

Not only are employees using it, but enterprises are also deploying it at scale. At the end of May, Microsoft disclosed data showing that TCS, Infosys, and Wipro each had deployed over 100,000 Microsoft 365 Copilot licenses, totaling more than 300,000 seats across the three companies—recognized by Microsoft as one of the largest enterprise AI implementations globally.

In other words, India is experiencing two developments simultaneously: AI is disrupting traditional outsourcing jobs, while also penetrating enterprises at an unprecedented pace.

This is also why the head of Microsoft India recently stated publicly that India has become one of the fastest-growing AI markets in the world.

What might be the new paradigm for IT outsourcing in India?

In fact, while capital markets are still concerned that India’s IT outsourcing model is being disrupted by AI, India’s largest tech companies have already begun seeking new ways to generate revenue: no longer selling engineers, but selling AI-powered productivity.

The most notable example is TCS. In the first quarter of 2026, TCS disclosed that its annualized order volume for AI-related business had reached $2.3 billion, up from $1.8 billion in the previous quarter, representing a quarterly growth of approximately 28%.

Meanwhile, TCS secured new orders of $12 billion in a single quarter, remaining at a historical high. This highlights an interesting trend: customers haven’t stopped spending—they’ve simply changed how they spend.

In the past, customers purchased 100 programmers. Today, customers purchase AI solutions, agent systems, and automation capabilities.

Many believe India’s biggest opportunity in the future is to follow the U.S. and create a Cursor. However, a prevailing view is that India’s true opportunity lies in becoming the world’s largest AI implementation hub.

On large models, the U.S. has already seen the emergence of monopolistic companies: OpenAI, Anthropic, Google, Meta, and others. But models are just the beginning; the real complexity lies in deployment.

By 2026, India had over 2,100 Global Capability Centers (GCCs) serving multinational corporations such as Microsoft, JPMorgan Chase, Goldman Sachs, Walmart, and Pfizer, generating approximately $100 billion in revenue and directly employing over 2.36 million people.

These global capability centers essentially handle software development, system integration, enterprise digitalization, data management, and IT operations—tasks that require substantial engineering execution capabilities, which represent India’s deepest accumulated advantage over the past 30 years.

According to IDC, global enterprise AI spending is projected to exceed $630 billion by 2028. India’s greatest opportunity may not be competing in the model market, but rather in the deployment market.

If over the past 30 years, India exported engineers to the world, then over the next 10 years, India may export to the world: agent deployment capabilities, AI operations capabilities, and AI productivity.

Perhaps this is India's true new paradigm.

This article is from the WeChat public account "Pencil News" (ID: pencilnews), author: Ai Yu, editor: Wang Fang.