The Manhattan Project didn’t just change warfare - it rewrote the relationship between science and power.
Today, AI is doing the same. The question is: have we learned anything?
Welcome to Plain Sight by Wyzr, where we bring hidden patterns into plain sight. Every week, we explore stories and ideas that help us understand why we are the way we are.
When Science Split the Atom (and the World)
Los Alamos, July 16th, 1945. 5:29 AM, Mountain War Time.
In the middle of a high desert plateau in New Mexico, a group of scientists huddle behind protective barriers twenty miles from Ground Zero. They’ve spent the past three years building something the world has never seen - a device that could unleash the fundamental forces that power the sun itself.
Robert Oppenheimer, the project’s scientific director, lies flat on the ground, his face pressed against the earth. Enrico Fermi fidgets with strips of paper - he plans to drop them when the blast wave hits to estimate the bomb’s power. Edward Teller smears sunscreen on his face and arms, then puts on welding goggles and sunglasses simultaneously.
At exactly 5:29:45 AM, the world changes.
A flash brighter than the sun turns night into day for over 200 miles in every direction. The heat is felt 20 miles away. Then comes the sound - a thunderous roar that rattles windows 120 miles distant.
Oppenheimer, watching the mushroom cloud rise 40,000 feet into the sky, recalls a line from the Bhagavad Gita:
“Now I am become Death, destroyer of worlds.”
The Trinity Test has succeeded. The atomic age has begun.
Three weeks later, Hiroshima and Nagasaki would be reduced to ash. Over 200,000 lives would be lost. The geopolitical order would be altered forever.
And remarkably, this wasn’t accomplished by generals or politicians. It was the work of physicists, mathematicians, and chemists - many of them European refugees who had fled fascism.
The Manhattan Project represented something unprecedented: the systematic transformation of abstract theory into world-changing technology. In less than four years, university researchers became the most powerful force on Earth.
The numbers are staggering. At its peak, the project employed over 130,000 people across 30 sites at a gargantuan cost of $28 billion in today’s dollars. Three entire cities were built from scratch.
But the real innovation was intellectual. For the first time, the world’s leading scientists were given unlimited resources, complete secrecy, and a single mission: turning E=mc² into an atomic bomb.
The result was an intellectual pressure cooker. Physicists worked alongside chemists, theorists collaborated with engineers. Hans Bethe, who led the theoretical division, called it “the most intense intellectual experience of my life.”
Yet many of these brilliant minds were simultaneously horrified by what they were creating. Leo Szilard wrote a petition signed by 70 scientists urging President Truman not to use the bomb. Niels Bohr spent months trying to convince Allied leaders to share atomic secrets with the Soviet Union to prevent an arms race.
This tension reveals something profound about transformative innovation. The scientists at Los Alamos discovered that breakthrough knowledge doesn’t remain neutral. Once certain thresholds are crossed, scientific discoveries become geopolitical weapons whether their creators intended it or not.
Consider what happened after 1945. Despite efforts to control atomic technology, within four years the Soviet Union had developed its own nuclear weapons. Within a decade, Britain and France joined the nuclear club. Today, nine countries possess nuclear weapons, and the knowledge exists in physics textbooks worldwide.
The genie never returns to the bottle.
Walk into any major AI research lab today - DeepMind, OpenAI, Anthropic, and you’ll find a remarkable parallel to Los Alamos circa 1943.
Once again, some of the world’s brightest minds are working to unlock a poorly understood but potentially transformative force. Once again, they’re doing so with excitement, ambition, and deep unease about the consequences.
Like the Manhattan Project physicists, today’s AI researchers are acutely aware of the dual-use nature of their work. The same systems that might revolutionize medical diagnosis could enable sophisticated disinformation campaigns. The algorithms that could democratize education might concentrate unprecedented power.
But AI also poses unsettling questions that echo the moral complexity of Los Alamos: Who gets to build it? Who gets to control it? Will it entrench existing power structures or democratize access to intelligence itself? Will it make us collectively smarter, or simply more replaceable?
Already, AI systems are automating creative work, influencing legal decisions, shaping political narratives, and disrupting entire job categories. Yet what’s more dangerous than what AI does might be what it knows - how it learns from our data, what patterns it reinforces, and what values it might eventually be trained to optimize for.
AI isn’t merely a tool like a hammer or computer. It’s a system of logic that increasingly reflects, and amplifies our collective choices, biases, and blind spots. Feed it biased data, and it perpetuates inequality. Train it on flawed reasoning, and it scales those flaws across millions of decisions.
And once again, the people building these systems are some of our brightest minds. Many are issuing public warnings even as they publish their next breakthrough paper. Geoffrey Hinton, the “godfather of AI,” left Google to speak more freely about AI risks. Yet like the Manhattan Project scientists watching their equations become weapons, they’re discovering that momentum in transformative technology is extraordinarily difficult to stop.
The Manhattan Project teaches us something crucial: the most dangerous moment isn’t when a technology is invented. It’s when it’s proven to work, and no one quite knows what to do next.
The central question is whether we can recreate the collaborative intensity of Los Alamos for constructive rather than destructive purposes. What would it look like to mobilize our brightest minds with the same urgency we once devoted to building weapons?
Consider the challenges that might benefit from Manhattan Project-level coordination: developing clean energy systems that could actually replace fossil fuels, creating AI systems that augment rather than replace human intelligence, designing economic models that distribute rather than concentrate prosperity.
After all, the same species that split the atom also composed symphonies, built cathedrals, and sent humans to the moon. The choice of what to do with our collective intelligence remains ours.
So here’s a question worth pondering: If you could assemble today’s equivalent of the Manhattan Project - with the same resources, urgency, and caliber of minds - what problem would you choose to solve?
And perhaps more importantly: Would you trust yourself, and the rest of us, with that kind of power?
What we’re reading this week
The Making of the Atomic Bomb by Richard Rhodes. This Pulitzer Prize-winning masterpiece weaves together the scientific discoveries, personal dramas, and political decisions that led to nuclear weapons. Rhodes shows how individual brilliance and coordinated effort can reshape the world—for better and for worse.
Hope you enjoyed this edition of Plain Sight. If you did, do share with your friends. Until next week.
Best,
Utkarsh
