Sketchnote · Click to open full-size in new tab
The Room Felt the Weight
There is a particular kind of silence that falls over a room when someone says something genuinely frightening. Not the horror-movie kind of fright, but the kind that comes from hearing a truth you've been vaguely aware of but haven't let yourself fully confront. That silence fell over a conference hall in Mussoorie on April 8, 2026, when Debjani Ghosh looked out at a roomful of India's most senior bureaucrats — IAS officers in the 17th round of mid-career training at LBSNAA — and said something that cut straight through all the polite conference-speak.
"In your roles today as senior leaders in government," she told them, "if you don't understand this, then the game is already lost. Because it is the policies that you will frame today that will determine what happens to us ten years from now."
She had opened not with a grand declaration, but with a question. She'd looked at the schedule — three talks on AI already that day — and asked the room what their biggest takeaway was. One has to live with it, someone said. Jobs will go, said another. We are too far behind compared to the US and China, said a third. Debjani nodded at each of them. And then she said: what I'm about to tell you is going to scare you even more. Because I'm going to describe the next ten years.
If you don't understand this, then the game is already lost. Because it is the policies that you will frame today that will determine what happens to us ten years from now. — Debjani Ghosh, to IAS officers at LBSNAA
The Pattern Reader
Debjani Ghosh is, by profession and disposition, a reader of patterns. She spent decades in industry — most famously as President of IESA and as a senior executive at companies including Intel — before being asked to come to NITI Aayog to build something the Indian government had never quite had before: an organization whose sole job was to see the future and then tell everyone about it.
The NITI Frontier Tech Hub publishes a tech scan every six months — a comprehensive analysis of what's going to happen in the next five to ten years — and sends it to the Prime Minister's Office and key decision-makers. Debjani was sharing a compressed version of the latest edition with the IAS officers. "A shorter version," she noted with characteristic dry humor, "in one hour."
The concept at the heart of her work is deceptively simple: technology doesn't arrive as bolts from the blue. It leaves trails. Thousands of small patterns, building on each other, pointing in a direction — if only you know how to read them. "Technology works through patterns," she explained. "If you look at the evolution of technology, it leaves behind... like we human beings leave behind footprints, technology leaves behind thousands and thousands of patterns."
Technology pathways are not shaped by isolated breakthroughs, but by the patterns that connect them. Those who learn to recognize these patterns early can better anticipate what comes next. — Slide 3, NITI Frontier Tech Scan
To illustrate, she traced the arc of AI through three pivotal moments. The first was Deep Blue beating Kasparov in chess in 1997 — a pattern, she said, that machines understood mathematics. The second was more jarring: in 2016, DeepMind's AlphaGo beat Lee Sedol at the game of Go.
"Chess you could still understand — it is a lot of maths into chess. Go is a hardcore strategy game. You can't play it just with numbers." The AlphaGo victory, she said, was her "aha moment" for AI — not ChatGPT. "I really woke up to it and said this is serious." The third pattern was the Transformer paper from Google, which showed how generative AI could work.
Those who had tracked these patterns weren't surprised by ChatGPT in November 2022. "The only surprise was it didn't happen earlier."
Eight Mega Patterns
Based on 10,000+ patterns analyzed over recent years, the Frontier Tech Hub identified eight mega-forces shaping the next decade. Debjani walked through them with the precision of someone who has spent years trying to make the complex legible to people who need to act on it.
The first and most fundamental: intelligence will flow through everything. Not as a standalone tool, but as a layer embedded into every workflow, every sector, every decision. "It's about diffusion of intelligence into everything. That's the job of AI."
The second: technologies are converging, not operating in silos. The biggest disruptions won't come from AI alone, or from biology alone — they'll come from the collision between them. The third: technology is going physical. Software is no longer eating the world; physical AI — robots, embodied agents, autonomous machines — is. The fourth: whoever controls the interconnected tech supply chain from end to end will control geopolitical leverage. "It's no more about saying, 'I have great engineers who can design XYZ.' That's not going to be enough anymore."
Competitive and geopolitical advantage will flow to those who command or secure trusted access to multiple layers of interdependent technology stacks. — Debjani Ghosh, NITI Aayog
Pattern five: success requires full-stack leadership — across energy, compute, models, and applications. Pattern six: it's no longer about research papers, but about outcomes — what changes in the economy, what improves in social development. Pattern seven, quietly terrifying: access to clean energy will be the greatest roadblock for all of technology, not just AI. And the eighth — the one that hung in the air longest: how do we redefine the role of human beings in an AI-augmented world?
The Stakes: What AI Could Add to the Global Economy by 2035
The Intelligence Revolution: It's Not About ChatGPT
The first of the four recommended focus areas for India was the Intelligence Revolution. But Debjani was careful to correct a dangerous misconception. India's relationship with AI at the enterprise level, she pointed out, was badly off track.
"India is the number one market for ChatGPTs," she said, and you could almost hear the grimace. "They are not the magic. The magic is how is AI transforming banking? How is AI transforming healthcare? How is AI transforming logistics and supply chain? The magic of AI is in workflow integration." The NITI report she referenced shows India's 18 priority arenas contributing $1.4–1.9 trillion in incremental GDP by 2035 — if the workflow integration actually happens.
The real frontier of AI deployment, she argued, was in reimagining government itself. Not using AI to understand what has already happened — most Indian government AI use cases still live here — but to predict what will happen. She described how governments are now building digital twins to stress-test policies before they're released. China built digital twins of most of its cities; Singapore built one of its entire power grid. "They can simulate, predict what will go wrong, and ensure that their maintenance is predictive rather than reactive."
We have to start thinking about technologies beyond conversational AI. I think that's the first challenge for enterprise and government in India. — Debjani Ghosh
When the Lights Go Out — and That's a Good Thing
To explain the second revolution — Physical AI — Debjani showed a video. On the left: a humanoid robot in 2023, jerky and uncertain, like a toddler on ice. On the right: the same robot in 2025, fluid, purposeful, walking through a home. Two years. The difference was staggering. "Now you know the meaning of exponential, right?"
She painted the journey in three stages: first, agentic AI managing complete workflows. To illustrate, she described coming home and telling her AI to prepare for making Palak Paneer — it would find the recipe, check her pantry, find the delivery slot that fits her schedule, add items to cart. "What I haven't yet given it approval for is to pay. I'm very scared of that." The room laughed. But the point landed: this AI isn't a chatbot. It's a digital employee.
Second stage: physical humanoid robots. Agibot holds 31.9% of the global humanoid installation market; its Chinese rival Unitree holds 26.5%. The top four companies in humanoid robotics are Chinese.
Third — and here she paused for effect — dark factories.
With lights dimmed and no workers in sight, this car factory in China uses hundreds of robots to turn out dozens of electric vehicles an hour, 24/7. This is a dark factory, an area of a plant so automated that in theory the lights could be completely shut off. — Debjani quoting from a visit by Indian leaders to China
The questions she asked the room were rhetorical but piercing. How many human beings manage the car factory? "Around 50." The phone factory? "Around 100." Productivity through the roof. And then, the uncomfortable truth: "India can't be the only one to say we don't want that future. You have to get the whole world to come and say we don't want that future."
The job displacement angle was honest and uncomfortable. "New jobs that are getting created unfortunately are not like the old jobs. Data entry operators employed millions. The new jobs don't do that. The new jobs are all hyper-specialized." She had searched every institute in India for someone teaching haptic engineering. "Finally I was able to find one. But they're not teaching it, but they do have a lab."
There are 1.5–2 million IT jobs at risk in India from AI by 2031. But there's also potential for 2 million new jobs — if the country builds the right skills fast enough. The NITI roadmap on AI jobs lays out how.
The Revolution Nobody Is Talking About
At around the 45-minute mark, Debjani said something that made at least some of the IAS officers sit forward.
"Biotechnology is the next mega general-purpose technology," she said. "This is going to be way bigger than AI."
And then she repeated it, because she could tell it hadn't quite landed the first time. "This is one place where India actually has a strong ecosystem already, thanks to the vaccine work that we had done."
The numbers she cited were arresting. Genome sequencing cost $3 billion in 2003 — the cost of the entire Human Genome Project. Today it costs around $600. The CRISPR gene-editing platform has moved from labs into clinical approvals, with the FDA having approved 48 cell and gene therapies by late 2025. Most recently, the first customized CRISPR therapy was used to treat a baby born with a severe liver genetic disorder — and was succeeding.
But Debjani wanted to break the perception that biotechnology is just about medicine. "Everyone used to think of biotech as health. Biotech is way bigger than health." Aircraft manufacturers are experimenting with self-healing materials grown from biology. Roads with self-healing cement are in pilot. McKinsey estimates that biology could eventually produce up to 60% of physical materials in the global economy.
If you can re-engineer molecules, you can re-engineer anything. This is going to be way bigger than AI. The race is on between China and US in terms of just how much they're spending and the size of the bioeconomy. — Debjani Ghosh
She laid out India's opportunity clearly: a current bioeconomy of $165.7 billion (4.25% of GDP), targeting $550 billion by 2035 and $1.2 trillion by 2047. The ABLE Association's projections underpin the government's thinking. The global bioeconomy is expected to exceed $16 trillion by 2035 — around 10% of global GDP. China's $500–600 billion bioeconomy is targeting $2 trillion by 2040 through gene editing and industrial enzymes.
And then came the warning that gave the room pause. "The future of warfare is biowarfare. It's about DNA mutation. If you start reading the scientific papers you will realize this is already happening. This is not science fiction anymore." How do you govern technology that can be used to cure hemophilia in tribal populations — or, with the same tools, engineer a pathogen? "We need to think from now."
The Four Revolutions India Must Master
The four revolutions are interdependent — you cannot lead in one without investing in all
The Sovereignty Problem
The third revolution — compute — brought Debjani to a subject she clearly found both technically fascinating and geopolitically alarming: the semiconductor supply chain. She traced it for the room, step by step.
EDA software — US dominance. Lithography — a single Dutch company, ASML. Wafer fab — US and Japan. Semiconductor materials — Japan and South Korea. Critical minerals — China, complete dominance. Foundries — Taiwan. Packaging — Asia, US, Europe. "This is a very difficult supply chain to crack."
India has critical minerals — but not the ability to convert them. India has engineering talent — but not the foundries. "Sovereignty has to come with trusted alliances." The slide she showed mapping who makes the humanoid robot — labelled "Key Challenge for New Entrants" — was a dense diagram of components from head to toe, nearly all dominated by Chinese or US firms.
The compute revolution, however, does offer India one genuine leapfrog opportunity: edge computing. The move from centralized data centers to intelligence at the device level is still in its early stages. "In the near future, you will have intelligence loaded on this watch. It will process the data on this watch." India has the design know-how, and the new semiconductor architectures — neuromorphic chips, quantum computing — haven't yet been cornered by any nation. "This is India's leapfrog moment."
The Energy Equation
The fourth revolution was energy, and Debjani delivered it with barely contained excitement. India's recent nuclear energy legislation had, she said, been genuinely positive news.
"Nuclear has to be the way forward. When it comes to clean, it's the cleanest. Most reliable energy source in terms of producing power, 93% of the time." She walked through the economics: high initial costs, but long asset life, cheap per kilowatt-hour over a 60-year lifespan. "The India news was so positive, because I think if we can scale it, we can rewrite the rules of the game."
The problem: five countries control 71% of global nuclear generation capacity. And uranium, the fuel, has its own hard-wired supply chain. "If we are dependent on uranium, we again have a very hard-wired supply chain." The nuclear renaissance — global nuclear generation hit a new record in 2025 — creates both an opportunity and a strategic constraint.
Resilience Over Security
Debjani had a confession to make. "I stopped using the word cybersecurity."
The reason was philosophical, and it resonated. Security implies prevention. But in a world where AI-powered attacks on power grids, hyper-personalized disinformation campaigns, quantum encryption collapse, and bioweapons are all emerging simultaneously — prevention is no longer achievable. "You know bad usage is going to happen. What I believe in is resilience. You know it's going to happen. How do you recover from it?"
The biggest risk for India, she argued, wasn't cyber attacks — it was operational. The lack of standards. "How many of you use your own processes and standards to procure technology?" Every hand went up. "When we do that, we are killing interoperability. And when you kill interoperability, you increase vulnerabilities 100x."
I don't believe in security anymore. You know bad usage is going to happen. What I believe in is resilience. Are you building resilient digital economies? — Debjani Ghosh
The Questions That Mattered
The Q&A that followed was where the room's latent anxiety surfaced fully. One officer raised the pattern of India's historical technology failures — GMOs, 3D printing, IoT, crypto — and asked whether this time would be different. Debjani's answer was characteristically direct: "I don't spend a lot of time on historical patterns because unfortunately we can't learn much from history in today's world. We can learn what mistakes not to make." She reminded them of what India had done: the national ID, India Stack, DPI. "It's our willingness to replicate what Nandan did in DPI across the other sectors."
Another officer asked the classic developing-nation dilemma: wait and watch to see which technologies win, or invest early with limited resources? Her answer was unambiguous. "If you wait and watch today, you can pretty much guarantee we will stay consumers." This decade, she said, is moving exponentially. By the time you decide to act, the game is over. But with finite budgets, the prescription was equally clear: "My recommendation would be pick five things, but go all in."
It doesn't matter which model it uses. Why do we want a sovereign model? Sovereignty should be at the data layer. — Debjani Ghosh
The sharpest moment came last, when an officer asked why India hadn't invested enough. Debjani paused for a beat, and then said something that will stay with anyone who was in the room.
"I don't want to get into studying history, sorry. Honestly I didn't spend time thinking about what we didn't do and why we didn't... I am a student of the future. The only thing I'll look at history for is are there any learnings on what not to do. That's all I'll look at. And then, start defining the future."
And then, speaking as a citizen rather than a technocrat: "Stop using history as an excuse!"
