Bangalore has another deep concern these days, apart from the insane traffic. The cornerstone companies of Bangalore’s IT sector — such as Infosys, TCS, Cognizant and Wipro — are getting hammered in the stock market. In the last month alone, at the time of writing this, these companies have lost around 20% — a fifth of their value. Over the past year, they have shed almost a third of their market capitalisation. Once-darling growth stocks that could do no wrong are now hovering near five-year lows.
The reason? The elephant in every boardroom in the world today — AI, or artificial intelligence.
AI is disrupting many sectors, if not all. However, recent progress in AI-based coding — where you simply type in plain English what you want the code to do, and the AI generates it in seconds — has shaken the foundations of our multi-billion-dollar, multi-million-employee Indian IT sector.
Heard of computer languages like C++, Java or even the ancient COBOL — the ones IT professionals take years to learn and perfect? Well, you don’t necessarily need to know them anymore. Just tell an AI tool what you want done in natural language. Like magic, the code appears in seconds — the same code lakhs of young, fresh-out-of-college graduates used to write at our big IT companies. Except the AI does it faster, cheaper and perhaps even better. Not to mention, it doesn’t need coffee, lunch breaks or vacations.
This is no longer fantasy. The shift to AI-based coding is already underway. AI coding assistants like Claude Code, GitHub Copilot (built on OpenAI models), Tabnine, Amazon CodeWhisperer, and tools such as Replit AI, CodiumAI and Sourcegraph Cody are not sci-fi gimmicks — they are part of real development workflows today. These tools generate code from natural-language instructions, autocomplete logic blocks, suggest bug fixes and even explain code.
For those unfamiliar with the IT-software-coding business, here’s an analogy. Remember the days before mobile phones — let alone smartphones with cameras? If you needed a picture, you went to a photo studio, hired a photographer or bought an expensive camera. You shot pictures on film rolls, then took them to a studio to get them developed. That ecosystem created multiple jobs — studio owners, photographers, film manufacturers, developers and more.
With the arrival of smartphones, we are all photographers now. Many of those earlier jobs disappeared. Forget photography — we are all videographers today. We have a simple tool: a cameraphone. Point and shoot.
AI-based coding works similarly. It democratizes coding, making it less of a tedious black box handled only by specialist programmers. Of course, you still need programming knowledge — just as you need photography skills to take good pictures or video skills to create compelling content. But now, fewer highly skilled people who know how to use AI tools effectively are enough. You don’t need an army of C++ coders anymore.
But that army is precisely what our IT sector was built on. We provided large pools of relatively inexpensive coding talent. It appears they may not be needed to the same extent anymore. Clients of these IT companies will likely reduce outsourced manpower and negotiate lower rates, comparing it to doing the work in-house with fewer AI-adept engineers armed with powerful tools.
And here’s the uncomfortable part: Bangalore and Gurgaon are not building most of these AI coding platforms. Big Tech companies in the US are. Our advantage has long been cheap labour — which, incidentally, underpins not just IT but even many consumer internet businesses in India. Cheap delivery networks, for instance, depend heavily on low labour costs.
But AI changes the labour-cost equation. If a US-based AI-trained engineer costs five times more but, with AI tools, can do the work of ten fresh Indian hires, it still makes economic sense for clients.
Meanwhile, countries like China are not just talking about AI — they are showcasing robots performing kung-fu dance shows. Yes, actual robots doing synchronized martial arts routines. Do watch those videos. You will realise how far advanced manufacturing, robotics and foundational AI research have moved elsewhere. And we are still debating billing rates.
The impact will not remain limited to IT companies and their employees. India’s IT sector contributes roughly 3% to GDP. Its decline may not devastate the economy, but it will have consequences. IT services exports form a significant portion of our total exports; a sharp fall would affect our trade balance. Lakhs of white-collar jobs could be at risk, and new engineering graduates may find fewer opportunities. Reduced job creation could lower consumer spending in major cities, potentially affecting real estate markets in IT hubs.
Is this a doomsday scenario? Perhaps. Is it imminent? Probably not.
Make no mistake — the threat is real. Multi-billion-dollar companies don’t lose a third of their value over baseless panic. At the same time, we must admit an uncomfortable truth: India’s IT sector — like many of our industries — is not known for deep innovation. Returns on R&D investment in India have historically been weak. Companies often prefer spending on “liaisoning” and navigating regulatory hurdles rather than investing heavily in research.
Indian firms frequently focus on winning contracts and tenders. Meanwhile, others are building trillion-dollar AI ecosystems. There are literally trillions of dollars being poured into AI globally. That wave — call it the trillion-dollar AI monster — is not waiting for us to catch up.
Cheap labour only takes you so far. If India wants to lead, we must innovate and operate at the cutting edge of technology. From IT and defence to robotics and advanced manufacturing — everything is evolving. Those at the frontier will win. Labour arbitrage alone will not.
That said, the Indian IT sector is not merely about cheap labour anymore. There has been meaningful evolution toward higher-end services. And this is not the first time doomsday predictions have surfaced. Many believed Indian IT would collapse after Y2K. Instead, it thrived. Similar fears arose with the rise of the internet and mobile apps. The companies survived and adapted.
However, each disruption is different. AI is unlike anything the Indian IT sector has faced before. It is faster, more scalable and directly attacks the core of the services model.
Can Indian IT companies evolve their way out of this?
The optimist — and the India-loving person — in me says yes.
Perhaps the future lies in a hybrid AI-human model. Just as hybrid engines in cars (internal combustion plus electric) have proven successful, maybe Bangalore can offer a model where humans and AI work together. Humans still offer something AI cannot — accountability, empathy and context. Also, let’s be honest, clients sometimes want someone they can yell at. You can’t exactly shout at an algorithm.
Will Bangalore be able to fight back against the trillion-dollar AI monster?
I think so.
And fighting back the Bangalore traffic monster? Well, that is a different story altogether.