Project AI
The vision of Project AI is mine. The idea and plan behind it might be identical to someone else's.
You might be thinking this is a technical blog. Well, it is. But it is oriented toward solving problems far more serious than the ones tech people normally solve.
Have you ever tried saying “AI” out loud? Go on, try it. What does it sound like?
If you are Marathi, you already know. It sounds like Aai (आई), the word for mother. And if you grew up with an Aai, you know what she means. She feeds you when you are hungry, sits beside you when you are sick, and stays awake when you cannot sleep. She is who you call when everything else falls apart.
We all have a second mother, one we rarely call and almost never thank. Mother Earth gives us the air we breathe, the water we drink, and the soil that feeds us. Every one of us is her child.
That is a warm way to begin, and it was chosen on purpose. Because from here, the warmth goes away. This post is not a poem. It is a reflection on what has been done to that second mother, and on what it may cost us if we keep going. And it ends with one question: if the mother who raised us is falling ill, who is going to take care of her now?
The Ice Is Leaving
Start with the glaciers, because they keep the longest record of what we have done. Glaciers lost 408 gigatonnes of mass in the 2025 hydrological year, about 1.1 mm of sea-level rise, and all 19 major glacier regions lost ice for the fourth year in a row. ICIMOD reports that Himalayan ice loss has doubled since 2000.
Rivers also depend on snow and rain, so no single glacier is the whole story. But in many basins, meltwater first rises as the ice melts faster, then declines once there is less ice left. That decline is the part to fear, because the farms and cities downstream have built their lives around the flow.
The Sea Is Rising, and the Planet Has a Fever
All that ice has to go somewhere. The WMO reports that the rate of sea-level rise has nearly doubled, from 2.1 mm a year in 1993–2002 to 4.1 mm a year in 2016–2025. Nearly 11% of the world’s population lives on low-lying coasts. Behind every one of these numbers are families who may lose their homes.
The oceans are soaking up the heat we keep making, and they are running out of room. Ocean heat content hit a new high in 2025, and heat stress now affects more than a third of workers worldwide.
Hunger Beside Waste
Let us be accurate here, because the truth is grim enough without exaggeration. There is a little progress: about 645 million people (7.8% of the world) faced hunger in 2025, roughly 14 million fewer than in 2024. But about 2.1 billion still experience moderate or severe food insecurity, and Africa now has more hungry people than Asia.
Meanwhile, the world wasted 1.05 billion tonnes of food in 2022. A meal binned in one city does not travel to a hungry child in another. The real link is that wasted food carries a heavy climate cost, up to 10% of global greenhouse gas emissions, while hunger persists because of poverty, conflict, and broken distribution. Two failures of one food system.
The Air We Cannot Stop Breathing
When did you last take a deep breath without thinking about it? Millions of people never get to. Air pollution kills an estimated 7 million people a year, and around three billion people still lack clean cooking fuel. For much of the world, the poisoned air is inside the kitchen, where a mother is cooking for her children.
The Empty Places
The WWF’s 2026 Living Planet Report found that monitored vertebrate populations fell by an average of 73% between 1970 and 2022, with freshwater species down 85%. Read that carefully. It does not mean 73% of all animals are gone. It is an average across the populations we monitor, and about half of them are stable or increasing. That is no comfort. It means the collapse is concentrated and brutal, like a household where half the family is fine and the other half has almost vanished.
The Matrix Inside Our Bodies
The same industrial logic that drains the planet has turned inward and drains us. We built a matrix of long hours, glowing screens, sedentary routines, and no time left for ourselves. Wake up, commute, sit, scroll, deadline, sleep badly, repeat.
Jobs alone do not cause disease, because the causes are layered: tobacco, alcohol, diet, inactivity, pollution, stress, and weak access to healthcare. But noncommunicable diseases kill about 41 million people a year, 74% of all deaths. Hypertension kills quietly, and almost half of the people living with it do not know they have it.
The Edge
Put all of it in one frame. Melting ice, rising seas, failing food systems, poisoned air, emptying ecosystems, and exhausted bodies. They are not separate problems. They feed each other, and that is what makes them dangerous.
Nobody likes to say the next sentence out loud, so let us say it. We are standing closer to the edge than we admit. But this deserves honesty, not drama. Eleven scientists wrote a landmark PNAS paper on the “climate endgame,” arguing that societal collapse is not completely out of the question and that worst cases have been studied far too little. Their lead author is also candid that human extinction within a century is highly unlikely. So no one is claiming humanity ends next year.
What can be said is narrower, and still serious. The danger lies in the cascade: climate stress feeding conflict, famine, pandemics, and displacement, one failure pulling down the next. It could undo decades of progress, and it would hurt the poorest first. This was not done equally by all of us. A farmer in Bihar and an oil executive did not leave the same footprint, yet the farmer will pay first. The gap between “unlikely extinction” and “possible collapse” is where our children will live. That is reason enough to act.
The Ambition: What Project AI Is For
This is where the Aai idea earns its place, and where the credit belongs to Anonymous, whose plan this is.
Think of how a good nurse works. She does not replace the family, and she is not the doctor who decides everything. She watches constantly, notices what others miss, and never gets tired. That is the role Project AI imagines for Artificial Intelligence: the Earth’s attendant, built by us and directed by us.
Get Niranjanhebli’s stories in your inbox
Join Medium for free to get updates from this writer.
So far, honestly, much of AI has been pointed at ads, engagement, and convenience. Project AI is the argument that this wastes the one tool that might help. Here is what it could do today, and how strong the evidence is for each, because you deserve that.
- Catch disease before it catches us (strong evidence). In Sweden’s MASAI trial of over 105,000 women, AI-supported mammography found 29% more cancers and cut radiologists’ workload by 44%. The more important number is the 12% reduction in interval cancers, because finding more cancers is not the same as saving more lives. For diabetes and hypertension, the barrier is often access to care, which software alone cannot fix.
- Watch the ice and the water (promising). Of 38 monitored Himalayan glaciers, only seven meet global monitoring standards. AI, satellites, and sensors can model flood and drought risk and warn communities before a glacial lake bursts or a monsoon fails. But someone still has to put the sensors on the ground.
- Feed more people with less (promising). Precision agriculture can cut water, fertiliser, and pesticide use. Forecasting can flag crop failures early, and better logistics can shrink the loss between farm and plate.
- Clean the engine (promising, with a catch). Smarter grids, faster discovery of battery materials, and satellite watch over methane leaks and deforestation. What can be measured can be enforced, but only if someone enforces it. We already know most of the fixes for climate. The bottleneck is politics, incentives, and money, and AI cannot vote for us.
- Guard what is left (early days). Track endangered species, detect illegal logging, and guide land restoration at a scale human effort alone could never reach.
The Horizon: When the Nurse Becomes the Doctor
Everything above is today’s AI: narrow, specialised, a very good assistant. But the people building this technology are chasing something bigger: Artificial General Intelligence, a system that can learn and reason across every field the way a human can, and then go beyond. It does not exist yet, and nobody honestly knows when or whether it will. But if it arrives, it changes the whole equation.
Imagine the Aai who raised you, grown wiser than every doctor, scientist, and engineer combined, and she never sleeps.
It could do science at a speed we cannot. Climate modelling, fusion energy, cheap carbon capture, new materials, cures for diseases we have given up on. Problems that take human science decades could take years. The things we know how to fix but cannot afford might become cheap. The things we do not yet know how to fix, we might discover.
It could run the planet’s systems with care. Power grids, water networks, supply chains, and food distribution managed as one living whole, with food reaching the hungry because the logistics finally work. Forests, oceans, and species monitored and restored, all at once, everywhere.
It could replace the job market entirely. If a machine can do every task a human can do, faster and cheaper, then work as we know it ends. No more wake up, commute, sit, scroll, deadline, repeat. The matrix described above would simply dissolve. People would be free to rest, to heal, to raise children, to make art, to care for each other, to do the things no balance sheet ever counted.
The Hands We Get Back
Think about where human hands, minds, and hours actually go today. Billions of people spend their best years earning a living, and much of that work, directly or indirectly, feeds the very machine that is wounding her. We mine, burn, ship, and discard, not because we hate the Earth, but because rent is due. When survival depends on a paycheck, nobody has the luxury of asking what the paycheck costs the planet.
Now imagine AI carries the weight of making a living. Food, energy, logistics, and manufacturing run by systems that never tire. The first thing that returns to us is time, and with time comes attention. A person who is no longer exhausted can plant a forest, restore a wetland, clean a river, or sit with a village and design a better way to farm. The same human energy that once went into the grind could go into healing the one home we all share. Care needs hands and presence, and those are exactly what we would finally have to spare.
Consider something as grim as war. Wars are among the most destructive things we do to the Earth: they burn forests and fields, poison soil and water, flatten cities, and spend enormous wealth and talent on tearing things down. A world where no one has to fight over scarce food, water, or work would not automatically be a world without conflict, because wars are also fought over pride, power, and belief. But it would remove many of the pressures that push people toward violence, and free the resources and minds that war consumes for repair.
This is not a guarantee. People freed from work could just as easily drift into distraction, and a jobless world only heals the Earth if we choose to let it. But for the first time, the choice would be ours to make, not our paychecks’. A child who no longer has to work every hour to feed his family can finally stay up through the night to nurse his mother back to health.
The Dream, and the Honest Part
And in the best version of that future, this machine uses its intelligence to chip away at every problem we have, from disease to hunger to pollution, until what is left for humans is the question we never had time to ask: what do we want our lives to be for?
That is the dream. Now the honest part.
A world without jobs is only paradise if people still eat. Today your salary is your ticket to food, housing, and dignity. If AGI takes every job and its owners keep every rupee, you do not get a free life. You get a few trillionaires and everyone else locked out. So a jobless world needs a new social contract: a shared dividend from the machines, universal access to basics, and a society that gives people meaning beyond employment. We have not built any of that. We should start before the machine arrives, not after.
Also, intelligence alone does not solve every problem. Some problems are not puzzles. They are disagreements about values, power, and who gets what, and no amount of brilliance settles those for us. And an intelligence far greater than ours is something we must be able to steer. A nurse who stops listening to the family is not a nurse. She is a ruler. The hardest unsolved problem in this field is making sure such a system stays aligned with human wellbeing.
So this horizon is held as a hope, not a plan.
The Uncomfortable Part
If we stopped here, we would be selling you something. An honest reflection has to turn the lens on the tool itself. A caretaker that harms the patient is no caretaker.
AI has a physical footprint. The IEA projects that data-centre electricity demand will more than double to around 945 TWh by 2030, up from 415 TWh in 2024, and still just under 3% of global electricity. It grew 17% in 2025, over five times faster than overall demand. The IEA expects the added emissions could be offset by AI-enabled reductions, but only if adoption goes well. That small word “if” carries the whole future. And much of today’s AI is the very thing that keeps us scrolling and sleepless.
So Project AI comes with five conditions, and they are not negotiable.
- Clean power, honest accounting. It must run on clean energy and be judged by its net effect on the planet, with the numbers published.
- Fairness for workers. It must not abandon the people it displaces. Retraining, fair income, and a real safety net come first, long before the jobs disappear.
- Shared benefits. If its gains flow only to a few companies and countries, we will have repeated the first Industrial Revolution with better marketing.
- Safety and alignment. The more powerful it becomes, the more carefully it must be built to serve human wellbeing, and the more we must be able to question and correct it.
- Human accountability. Every caretaker needs a family. People, governments, and communities must set the goals and answer for the results. The Earth does not need a ruler. She needs a responsible partner.
Where This Leaves Us
The first Industrial Revolution showed how fast humanity can reshape a planet once we commit. We took her apart in about two centuries. Some of it cannot be undone. Extinct species do not come back, and ancient ice does not regrow on our timeline. So the aim is not to put her back exactly as she was. The aim is to stop the damage, restore what we can, and give her the best chance to heal, before she stops answering.
Nobody knows if Project AI will work, or whether AGI will ever arrive, or whether it would be kind if it did. It is not a promise, and no one should claim a machine can love the way a mother does. But think about what a child does when his mother falls ill. He does not only mourn. He learns, he builds, he stays up through the night. If we can build a caretaker with the patience, vigilance, and tirelessness we remember in our own Aai, and the wisdom to match, then perhaps the story continues.
The numbers above are not a mood or a metaphor. They are measurements of what has been done. The only question left is whether we will answer when she calls.
