Let me be direct with you. I am a developer who has spent 13 years in the Indian tech industry — in Mumbai, Hyderabad, and two years in Abu Dhabi. I have watched the AI shift happen from inside it, not from the outside. I have seen colleagues worry about their jobs, watched companies change how they hire, and used AI tools daily in my own work.

The question "is AI replacing jobs in India?" deserves a straight answer, not a hedged corporate statement or a panic-inducing headline. So here is what is actually happening — sector by sector, with real numbers, and with an honest assessment of what it means for you personally.

The Scale of What India Has Built — And What Is at Stake

To understand what AI disruption means for India, you first need to understand what India built over the last 30 years. The IT and BPO sector employs 5.4 million people directly and contributes 7.5% of India's GDP. For context — that is not just jobs. Those 5.4 million people buy houses, send children to private schools, take flights, drive cars, and pay for services across the economy. They are what economists call the "aspirational middle class."

Global equity research firm Bernstein put it plainly in an open letter to Prime Minister Modi earlier this year: 10 to 15 million Indians working in IT services and BPO have anchored India's aspirational middle class — buying homes, taking flights, driving consumption. When AI disrupts this sector, the effects ripple far beyond technology companies.

This is why the AI and jobs question in India is not just a technology story. It is an economic and social story. The stakes are real.

What the Fresh Data Actually Shows — April 2026

Let me give you the most current numbers, because the situation is moving fast.

IT Sector Hiring Has Already Fallen

This is not a prediction — it has already happened. NASSCOM reported that workforce growth in India's tech sector slowed to 2.3% in FY26, even as the industry continued to expand. Think about that for a moment. The industry is growing in revenue terms but reducing people.

India's largest IT firm, Tata Consultancy Services, which laid off 12,000 last July, has plans to hire just 25,000 fresh graduates this year compared to an average of 40,000 new hires over the last three years. For the last five years, gross hiring of IT firms averaged around 230,000, but in the financial year ending in March 2026, they added around 170,000.

That is a drop of 60,000 annual hires — roughly a 26% reduction — in a single year. The companies are not going bankrupt. They are doing more with fewer people, powered by AI tools.

Entry-level IT roles have been cut by 20 to 25% as AI reshapes hiring. And Cognizant has rolled out 'Project Leap' — a new AI transformation program that involves not only workforce reskilling but also job cuts.

From my own experience at Cognizant between 2019 and 2023 — I watched the shift from volume hiring to selective hiring begin. The company started using AI tools internally for code review, documentation, and testing before most people in the industry were paying attention. What is happening now is that shift accelerating dramatically.

The BPO Sector — A More Urgent Problem

If the IT sector is experiencing a slow squeeze, the BPO sector is experiencing something faster and more acute. By 2030, BPO employment could drop from four million to fewer than one million — that is the projection from analysts tracking the sector.

The reason is straightforward. AI-powered voice agents are evolving to handle customer queries with human-like empathy, tonal variation, and conversational nuances. The core value proposition of the Indian BPO industry — English-speaking agents at lower cost than Western markets — is being directly undercut by AI that costs a fraction of even Indian wages and works 24 hours a day without sick leave or turnover.

This is hitting real people in Hyderabad, Bangalore, Pune, and Chennai right now. These are not statistics. They are 4 million people who built stable middle-class lives in a sector that is contracting.

The harder truth: many displaced workers have moved into gig economy roles such as driving or delivery, sometimes earning more than their former BPO salaries. But gig work does not provide the stability, benefits, or career progression that BPO work historically offered.

The White-Collar Jobs That India Built Its Economy On

This is the part that surprises people. Previous waves of automation hit blue-collar and manufacturing jobs. AI is different. A World Bank report highlighted that "unlike previous waves of automation, AI has the potential to displace a range of non-routine, white-collar service sector jobs."

India's Economic Survey 2025-26 notes that over 60% of formal sector jobs are "susceptible to automation" by 2030, particularly in the IT and BPO sectors. That is a remarkable number. Three in five formal sector jobs in India involve some meaningful automation risk over the next five years.

Vinod Khosla — co-founder of Sun Microsystems and one of Silicon Valley's most credible voices on AI — said at the India AI Impact Summit 2026: "IT and BPO services will disappear, almost certainly within the next five years. In the next five years, AI will be better than most humans at most things." That is a stark statement from someone who has been consistently right about technology trends for 40 years.

Whether you agree with Khosla's timeline or not, the direction is clear. The question is pace, not direction.

Which Specific Jobs Are Most at Risk — Honest Assessment

Let me go sector by sector with honesty rather than reassurance.

Junior and Mid-Level Software Development

This is where I can speak most personally. I have used ChatGPT for Unity C# coding for two years. Claude for architecture and complex problem-solving. GitHub Copilot for autocomplete. What used to take me 3 hours for a complete feature implementation now takes 45 minutes with AI assistance. If one developer with AI tools can do what previously required 1.3 developers, companies need fewer developers for the same output.

Companies are increasingly moving away from large-scale fresher hiring towards more specialised roles. Even some mid-level roles are being affected, with routine coding and maintenance tasks increasingly automated or augmented by AI systems.

The jobs most at risk in software development:

• Junior developers doing basic CRUD operations — AI writes this code reliably now
• Manual QA testers — automated testing tools handle most regression testing
• Documentation writers — AI generates documentation from code
• Basic data analysts — ChatGPT and Claude can analyse datasets when given the right prompts
• Junior IT support — AI chatbots handle tier-1 queries effectively

The jobs that are becoming more valuable in software:

• System architects — designing how systems fit together requires judgement AI cannot replicate
• AI integration engineers — the people who connect AI tools into existing systems
• Domain specialists who code — a healthcare developer who understands clinical workflows is far more valuable than AI alone
• Security engineers — AI creates new attack surfaces that need human security thinking

BPO and Customer Service

The most immediate and largest-scale disruption. Voice AI in Indian and South Asian languages has improved dramatically. Sarvam AI, a Bangalore-based startup, is specifically building multilingual AI for this market. When an AI can handle a customer query in Hindi or Telugu with high accuracy for a few paise per interaction versus hundreds of rupees per hour for a human agent — the business case for human agents becomes very difficult to sustain.

The jobs at highest risk:

• Tier-1 call centre agents — handling standard queries, complaint logging, basic troubleshooting
• Chat support agents — AI chatbots handle this category better than any other
• Data entry operators — fully automatable with current OCR and AI tools
• Email response agents — AI drafts and sends routine email responses at scale

What survives in BPO:

• Complex complaint resolution — situations requiring empathy, judgement, and escalation authority
• High-value customer relationships — premium banking, wealth management, enterprise accounts
• AI supervision roles — monitoring, quality-checking, and correcting AI systems
• Training data generation — creating the data that trains the AI systems

Content and Creative Work

This is more nuanced than the other sectors. Generic, formulaic content — product descriptions, basic SEO articles, standard reports — is being produced by AI at scale for a fraction of human cost. I see this in the site I run: AI-generated content floods search results and the quality is often acceptable for basic purposes.

What AI cannot replace in content:

• Original reporting and journalism requiring source relationships and on-ground presence
• Creative strategy and brand voice — the thinking behind the content
• Complex technical writing requiring deep domain expertise
• Content that requires genuine lived experience — the kind that reads like a real person actually did something

If you are a content writer producing generic articles from briefs, your work is under pressure. If you are a writer with genuine domain expertise, a distinctive voice, and real-world experience — you are more valuable than ever because you are what separates authentic content from AI filler.

The Other Side — Jobs AI Is Creating in India

The disruption story is real. So is the opportunity story. Let me be equally honest about both.

Prompt Engineering — India Is Already Winning This

Knowing how to give AI effective instructions — prompt engineering — has become a genuine and lucrative skill. Indian professionals on Upwork and Fiverr are among the top earners globally in this category. The combination of English proficiency, technical background, and cost advantage means India is well-positioned for this work.

A skilled prompt engineer working for international clients can earn ₹8–25 lakhs annually from remote work. That is competitive with mid-level IT salaries at major companies, from home, without the commute or office politics.

Our AI Prompts Library has over 200 prompts across writing, coding, marketing, and business that you can use to understand what effective prompting looks like — and start developing this skill today.

AI Training Data — A Massive and Growing Market

Every AI model needs training data — labelled images, annotated text, human feedback on AI outputs. India has become a major global provider of this work. Companies like Scale AI, Appen, and Indian firms like iMerit employ thousands of people for data annotation, content moderation, and AI feedback roles.

This work pays less than traditional IT but requires less technical skill — making it an accessible transition for workers from at-risk roles. And the volume of work is growing as AI systems multiply.

AI Implementation Consulting

Every Indian company wants to adopt AI but most do not know how. The person who understands both the AI tools and the business context — who can walk into a textile manufacturer in Surat or a hospital in Chennai and explain what AI can and cannot do for their specific situation — is in extraordinary demand.

A joint analysis by NASSCOM and Deloitte estimates that India's AI talent pool could reach around 1.25 million by 2027, although demand may outpace supply, creating a significant skills gap. That skills gap is an opportunity for people who move now.

New Roles That Did Not Exist Before

What the Indian Government Is Doing — And Whether It Is Enough

The Indian government has not ignored this. The India AI Mission launched in 2024 with a ₹10,372 crore budget — one of the largest government AI investments in Asia. The mission covers three areas: building compute infrastructure, creating Indian-language AI datasets, and skilling the workforce.

The stated goal is training 5 million Indians in AI skills by 2027 through IITs, IIMs, and online platforms like NPTEL. India's IT minister Ashwini Vaishnaw acknowledged at a summit earlier this year that disruption to jobs in the tech sector was a "real challenge" — which is unusual candour from a government minister.

States are acting too. Telangana — where I am based in Hyderabad — has been particularly active. T-Hub, one of India's largest startup incubators, is now focused heavily on AI startups. The Telangana government's AI skilling initiatives are among the most aggressive in the country.

Is it enough? Honestly — probably not fast enough. The speed of AI deployment in companies is outpacing the speed of government skilling programmes. The workers who will benefit from India AI Mission training in 2027 are the same workers who may have already lost their jobs by 2026. Government programmes work on political timelines; AI adoption works on market timelines. They are not running at the same speed.

This is not a criticism of the intent — the intent is right. It is a realistic assessment of the pace mismatch.

The Honest Sector-by-Sector Risk Table

What Should You Do Right Now — Practical Steps

I get asked this question constantly. My answer has been the same for two years and the urgency has only increased. Here is what I tell people in the Indian tech industry specifically.

Step 1: Use AI Tools in Your Current Job Starting Today

Do not wait for your company to mandate it. Start using ChatGPT, Claude, or Gemini for your actual daily work this week. If you are a developer — use Copilot for code completion. If you are a writer — use AI for first drafts. If you are in finance — use AI for data analysis and report generation.

The goal is not to replace yourself. The goal is to become the person who does more with AI than your colleagues do without it. That is the person companies want to keep and promote.

Step 2: Identify the AI-Resistant Parts of Your Current Role

Every job has parts that AI handles well and parts that it cannot. Your job security lies in the second category. Take 30 minutes and write down:

• What parts of my job require judgement calls that have no clear right answer?
• What parts require client or colleague relationships?
• What parts require understanding of my specific company or industry context?
• What parts would be wrong if an AI made a mistake with no human review?

Those are your most valuable activities. Do more of them. Let AI handle the routine parts so you have more time for the high-value parts.

Step 3: Build One AI-Adjacent Skill This Quarter

Do not try to become an AI/ML engineer overnight. That is not realistic for most people and not necessary. Instead, pick one skill that makes you more effective with AI in your current domain:

• If you are in IT: Learn to use Claude Code or GitHub Copilot effectively. One weekend of focused practice is enough to see significant productivity gains.
• If you are in BPO: Learn to build simple AI chatbot flows using no-code tools like Botpress or Voiceflow. The skill of configuring AI systems is what the companies automating your role need.
• If you are in content: Learn to use AI as a research and structuring tool while your genuine perspective and voice provide what AI cannot.
• If you are in finance: Learn to use AI for data analysis and report generation — specifically, learn the prompts that make financial analysis faster and more accurate.

Step 4: Think About the Next 3 Years, Not Just Right Now

The workers who will be in the best position in 2028 are the ones making decisions today. If your current role has a clear automation trajectory — entry-level BPO, routine coding, data entry — the time to plan a transition is before you are forced to make one urgently.

This is not alarmist. It is the same advice any experienced career person would give about any technological shift. The workers who transitioned from landline telephone operators to internet roles in the 1990s who did it gradually on their own terms were in a far better position than those who waited until the call centre shut down.

My Personal Take — What I Tell People in Hyderabad

I am based in Hyderabad. I talk to developers, designers, content people, and young graduates regularly. Here is what I actually tell them when they ask me whether to be worried:

If you are in your first three years of a career and your role is primarily routine tasks — be concerned and act now. The market for pure routine-task roles is shrinking faster than most people expect. Not panic, not despair — but clear-eyed action. Start learning AI tools. Start building skills adjacent to AI. Start having conversations with people in roles that are growing, not shrinking.

If you have 5+ years of experience and genuine domain expertise — you are in a much stronger position than the headlines suggest. AI augments experienced professionals dramatically. The developer with 8 years of expertise using AI tools does not produce 1.1x output. They produce 2-3x output. That makes them more valuable, not less.

If you are a fresh graduate choosing a career path in 2026 — choose domains where AI is a tool, not a threat. AI/ML, cybersecurity, healthcare technology, infrastructure, and physical industries (construction, manufacturing, logistics) all need people who understand technology but cannot be entirely replaced by it.

The question I keep coming back to in my own work: am I using AI to do more of what I already do, or am I learning to do genuinely new things that AI enables? The first gives you job security. The second gives you an advantage.

Frequently Asked Questions