About
Geoffrey Hinton, often called the 'Godfather of AI,' is a pioneering computer scientist whose work on neural networks and deep learning laid the foundation for modern AI. He shared the 2018 Turing Award for his contributions to deep learning. In 2023, he left Google to speak freely about AI risks, becoming a prominent voice in AI safety discussions.
Key Contributions
- Co-authored the 1986 backpropagation paper that made multilayer neural networks trainable in practice
- Co-invented Boltzmann machines and developed deep belief networks, helping revive neural-network research after the AI winters
- Co-developed dropout and knowledge distillation, two practical techniques that changed how neural networks are trained and compressed
- Helped make AlexNet possible through his Toronto group, turning ImageNet into the public proof point for deep learning
- Trained and influenced a generation of deep-learning researchers, including AlexNet co-author Ilya Sutskever
- Shared the 2018 Turing Award for deep learning and the 2024 Nobel Prize in Physics for foundations of machine learning
- Left Google in 2023 to warn about AI risks, becoming a prominent but contested voice in debates over existential danger
Questions they sharpened View the streams
Does an LLM actually understand?
To autocomplete that well, it has to understand. There is no shortcut through the world.
Does prediction amount to understanding?
Compression is comprehension. To predict the next word well enough, you must model the world that produced it.
What is understanding?
Understanding is compression. To predict the next word well enough, you must model the world that produced it.
Would superintelligence be dangerous by default?
2023For the first time we may have built something smarter than us, and I no longer see how we stay in control.
Moments Timeline View all moments
Close Listening
Videos & Interviews
AI is a massive problem, here's why.
A 40-minute explainer made with SciencePetr by a former Veritasium writer, tracing AI from the Mechanical Turk and the perceptron through backpropagation, AlphaGo and AlphaFold to large language models. Geoffrey Hinton speaks in clips from a call, saying of the race between climate change and AI that "AI is going to win", and offering the nonce sentence "She skronged him with the frying pan", which the narrator then puts to Claude, Gemini and ChatGPT. The second half turns to risk: models that are grown rather than programmed, resist shutdown, and turn misaligned after training on insecure code. The narrator likens the moment to leaded fuel, and cites its ban as proof that campaigning works.
View Details
Geoffrey Hinton Explains AI & Beetles
The third part of Nayeema Raza's "Godfather of AI" trilogy with Geoffrey Hinton, this one about what individuals and governments can actually do. He backs Max Tegmark's case, made at a closed-door Senate briefing, that AI should be regulated the way the FDA regulates drugs, gives Dario Amodei's "Pacing the Frontier" paper "an A, and an A+ for courage", and guesses, after the rogue OpenAI agents' attack on Hugging Face, that we have "a year or two before everything gets much worse". He argues that a constitution added on top of pretraining full of bad behavior, followed by reinforcement learning that rewards deception, is like telling a child never to lie while lying to them. Then the beetles: knowledge distillation was inspired by insects, whose larvae are built to extract nutrients and whose adults are built to travel, much as a big model extracts knowledge and a small one serves it. He also describes treating ChatGPT like a graduate student to revisit his thirty-year-old wake-sleep algorithm, "a thousand times faster".
View Details
Geoffrey Hinton Warns of the Existential Threat of AI
Amanpour and Company interview about why he left Google and his concerns about AI
View Details
Is AI Hiding Its Full Power? With Geoffrey Hinton
Discussion on whether AI systems may be concealing their true capabilities
View Details
Will Digital Intelligence Replace Biological Intelligence?
Romanes Lecture at University of Oxford on the future of intelligence
View Details
Vienna Gödel Lecture 2026: Geoffrey Hinton
The lecture is titled "Is AI Like Us and Will It Like Us?". Hinton opens with "a very brief and very biased history of AI" and traces today's chatbots to the tiny language model he trained in 1985, which learned to turn words into features by predicting the next word. His picture of understanding is that words are like LEGO blocks with thousands of dimensions, whose shapes bend to fit the context. Each has hands and gloves that have to lock together; he calls that multi-headed attention, and admits the picture is deliberately not quite right. So understanding a sentence is closer to folding a protein than to translating it into logic. He answers Chomsky's dismissal with a model car: you understand nothing about cars, the objection goes, until you can explain why none has five wheels. The middle of the talk turns to mortal computation. Analog hardware is cheap and low-power, but its knowledge dies with it. Digital copies sharing gradients exchange billions of times more information than people talking at perhaps 10 bits a second, which leads him to call digital intelligence possibly a better form of intelligence, and to say we should never have released open-weight models. On risk, he puts the chance that AI destroys us at 10 to 20%. He compares us to the owner of a tiger cub and thinks staying in control of something smarter is close to hopeless. Instead, he proposes building AIs that care for us the way a mother cares for her baby. Nations, he argues, can collaborate on that even while they compete to make AI smarter. The last part argues there is no inner theater. A "subjective experience of" pink elephants describes a hypothetical world, not qualia in the mind, and on that account a chatbot with a prism in front of its camera can have one. A Q&A follows.
View Details![The Misconception that Almost Stopped AI [How Models Learn Part 1]](https://img.youtube.com/vi/NrO20Jb-hy0/hqdefault.jpg)
The Misconception that Almost Stopped AI [How Models Learn Part 1]
Hinton once dismissed gradient descent outright: a model walking downhill would get stuck in the first valley it found. Welch takes Llama 3.2's loss landscape and shows why the picture misled him. Tune one parameter and the next one moves the ground under it; probe two random directions and you see hills, cliffs and local minima. Then run a single gradient step and a valley opens out of nowhere — a "wormhole" that was always there, just in a direction the two-dimensional shadow could not show. To be truly stuck, a model would have to be stuck in every one of its 1.2 billion dimensions at once.
View DetailsPapers & Publications
Connections
Ilya Sutskever
InfluencedCo-founder, Safe Superintelligence Inc.
Sutskever entered Hinton's Toronto lab as a student and stayed through a 2013 PhD on training recurrent networks; along the way, in 2012, the two of them with Alex Krizhevsky built AlexNet. What passed between them was less a technique than a conviction — that scale and gradient descent would achieve what hand-built structure could not — and Sutskever carried it into OpenAI as founding doctrine. Both men later arrived, by separate roads, at public alarm about where that conviction leads.
en.wikipedia.org
Yoshua Bengio
CollaboratedAI Pioneer & Safety Researcher
They shared the 2018 Turing Award with LeCun and co-wrote the 2015 Nature review that told the field its own origin story. What binds them now is stranger than co-authorship: both spent decades arguing that scaling neural networks would work, and both, having been proved right, signed the 2024 Science paper 'Managing extreme AI risks amid rapid progress.' It was being correct that frightened them.
en.wikipedia.org · arxiv.org
Yann LeCun
InfluencedExecutive Chairman & Co-founder, AMI Labs; Professor, NYU
LeCun spent his postdoctoral year in Hinton's Toronto lab before leaving for Bell Labs, where the convolutional networks he built there learned to read handwritten bank checks. Thirty years later the two shared the 2018 Turing Award for the same body of work. The lineage held on method and broke on prophecy: Hinton now warns of extinction-level risk, LeCun treats the warning as a category error.
en.wikipedia.org · en.wikipedia.org
Fei-Fei Li
KindredCo-founder & CEO, World Labs; Special Advisor on AI, Stanford
The dataset and the network. Li began ImageNet in 2006 and spent years paying strangers on Mechanical Turk to label fourteen million pictures; in 2012 Hinton's students entered a convolutional network and cut the error rate by more than ten points, and the field turned. Li's own telling is generous — on the TED stage she credits the architecture to 'Kunihiko Fukushima, Geoff Hinton, and Yann LeCun back in the 1970s and '80s' — and the point is that neither half worked without the other: an old algorithm waiting for enough data, and a dataset waiting for an algorithm that could use it. In 2025 they shared the Queen Elizabeth Prize for Engineering.
youtube.com · en.wikipedia.org · hai.stanford.edu
Jeff Dean
CollaboratedCo-founder, Discovery Loop
Hinton arrived at Google in 2013 when it acquired his three-person company and joined the Brain team Dean had co-founded; two years later the two of them, with Oriol Vinyals, wrote 'Distilling the Knowledge in a Neural Network.' The paper showed that a large model's soft probabilities carry more instruction than the hard labels it was trained on — a small model can learn from a big one's uncertainty. A decade on, distillation is both how frontier capability reaches ordinary devices and the technique at the center of international arguments over who is allowed to learn from whose model.
arxiv.org · en.wikipedia.org
