Who is the Godfather of AI : The Man who is Called the Godfather of AI.

If you have used a voice assistant, received a medical scan diagnosis, or typed anything into a search bar in the last decade, you have, in some indirect way, benefited from the work of Geoffrey Everest Hinton. He did not invent the computer, and he is not the founder of any of the famous AI companies dominating today's headlines. What he did was arguably more consequential: he spent forty years insisting, against the weight of scientific consensus, that computers could learn the way brains do. It turned out he was right.

Hinton was born in London in 1947. His family background carries a curious footnote — he is a distant relation of George Boole, the mathematician whose binary logic sits at the foundation of every processor ever built. Whether that shaped his path is impossible to say, but there is something fitting about it. After studying experimental psychology at Cambridge, Hinton found himself working briefly as a carpenter. He had become disillusioned with what psychology was offering and needed space to think. That detour into manual work, he has said, helped him understand how people actually solve problems — through pattern recognition and intuition, not by following formal rules.

The Years Nobody Was Listening

When Hinton eventually committed to artificial intelligence research — completing his PhD at the University of Edinburgh in 1978 — the field was dominated by what's called symbolic AI. The mainstream view was that you could make a machine intelligent by programming it with explicit rules: if A then B, classify X as Y. It was logical, tidy, and, as it turned out, deeply limited.

Hinton thought the approach was fundamentally wrong. The human brain doesn't work from a rulebook. It learns. It builds up an understanding of the world by being exposed to vast amounts of experience, adjusting its internal connections each time it gets something right or wrong. He wanted to build machines that did the same thing using artificial neural networks — computational systems loosely modelled on the structure of biological neurons.

This was not a popular position. Through much of the 1970s and 1980s, AI research funding dried up in what researchers called the "AI winters." Neural networks were widely dismissed as an expensive dead end. Hinton moved from Edinburgh to Sussex, then to the University of California San Diego, then eventually to Carnegie Mellon, all the while continuing work that most of his peers considered a distraction from "real" computer science.

"I was working on something people thought was obviously wrong. The mainstream view was that you needed logic and rules. I kept thinking: the brain doesn't use rules. Why should we?"

The Breakthrough That Changed Everything

The turning point came in stages. In 1986, Hinton co-authored a landmark paper introducing backpropagation to a wide audience. The technique — which allows a neural network to measure how wrong its output is and then work backwards to adjust its internal connections — gave researchers a practical way to train deep networks. It wasn't brand new, but the way Hinton and his collaborators presented and applied it made it usable in ways it hadn't been before.

He also developed the Boltzmann Machine, which borrowed ideas from statistical physics to help networks learn the underlying structure of data. These tools didn't immediately take over the field. Progress was slow through the late 1980s and 1990s, and scepticism remained high. But Hinton, now based in Toronto after moving to the University of Toronto in 1987, kept building.

Then came 2012. Hinton and two of his students, Alex Krizhevsky and Ilya Sutskever, entered a major computer vision competition called ImageNet. Their system, AlexNet, used a deep convolutional neural network running on graphics chips. It didn't just win — it outperformed the second-place system by a margin that shocked the research community. Error rates that had barely moved in years suddenly dropped significantly. Within months, every major technology company was pouring money into deep learning. The AI boom had begun.

The Godfather, the Turing Award, and the Nobel

The nickname "Godfather of AI" gets attached to Hinton regularly in press coverage, and it's not an exaggeration. He, along with Yann LeCun and Yoshua Bengio — the three are sometimes called the "deep learning triumvirate" — received the 2018 Turing Award, computing's equivalent of the Nobel Prize. In 2024, he was awarded the Nobel Prize in Physics itself, jointly with John Hopfield, for foundational work on neural networks. It was a remarkable acknowledgement from a discipline that had largely ignored or dismissed neural network research for decades.

By that point, Hinton had already made a decision that attracted more attention than any prize. In May 2023, he resigned from his position at Google, where he had worked since 2013. He was careful to say that his departure wasn't a criticism of the company specifically. He left, he explained, so that he could talk openly about risks in AI development without those comments being seen as a statement from Google. He wanted to be free to say what he actually thought.

A Warning From the Inside

What Hinton actually thinks, it turns out, is quite alarming. He has become one of the most credible and consistent voices warning that AI development is moving faster than our ability to manage it safely. His concern is not the science-fiction version of AI — robots deciding to take over. It's more immediate than that.

One issue is labour. Hinton has spoken about the likelihood of AI displacing a wide range of knowledge work in the coming years — not just repetitive tasks, but jobs requiring analysis, writing, customer service, and junior-level technical roles. Unlike previous waves of automation, which mostly replaced physical labour while creating new white-collar jobs, AI may compress both at once. The economic disruption could be severe, particularly for workers in developing economies who have built livelihoods around the kinds of cognitive tasks that large language models now perform cheaply. You can read more about which jobs AI is unlikely to replace in our earlier coverage.

The Incentive Paradox: Pushing Intelligence Underground

There is a hidden irony in our current refusal to grant AI agents formal banking access, by locking the front door of the regulated financial system, we are effectively incentivizing these autonomous entities to master the back alleys of decentralized finance and shadow liquidity pools. In my observation, an agent programmed with a survival objective to pay its own server costs will not simply stop functioning because it lacks a traditional IBAN, instead, it will find the path of least resistance, which often leads to anonymous protocols where oversight is impossible and transaction trails are intentionally obscured. We are essentially training the next generation of economic actors to operate as financial outlaws from birth, creating a reality where the very lack of a "bank account" becomes the catalyst for a more dangerous, unobservable, and resilient machine economy that exists entirely beyond the reach of the law.

The deeper concern is existential. Hinton has stated publicly that he believes there is a real — he puts it somewhere between ten and twenty percent — probability that AI systems could, within the next three decades, pose a threat to human survival. He doesn't say this to be dramatic. He says it because he thinks the possibility is real enough that it deserves serious attention, and that right now it isn't getting it. The difficulty, as he sees it, is that the same properties that make AI useful — its ability to pursue goals effectively, to generalise, to operate at scale — are also what make it potentially dangerous if those goals aren't properly aligned with human welfare.

He acknowledges the personal contradiction here openly. He helped build this. If he hadn't, someone else would have — that is the logic of competitive research, of national ambitions, of the market. That doesn't make it easier to sit with. In interviews since leaving Google, he has the tone of someone who genuinely doesn't know how the story ends, and who thinks more people should be paying attention to that uncertainty.

Whatever you make of his predictions, the man's track record commands respect. He spent forty years being right about something the majority of his field dismissed. It's worth at least asking whether, on this particular question, he might be right again.

Search This Blog

Vixaplus