The annual presentation of the Nobel Prizesat a special ceremony in Stockholm in Sweden each December recognizes crucial achievements in science. 

This year’s prize in physics jointly recognizes John Hopfield and Geoffrey Hinton for their research into machine learning with artificial neural networks. 

When announced in October, the Nobel Prize committee observed humans possessed a unique learning ability that surpasses other species on the planet.

"We can recognize images and speech, and associate them with memories and past experiences — billions of neurons wired together give us unique cognitive abilities," said Ellen Moons, chair of the physics committee.

"Artificial neural networks are inspired by this."

From this inspiration, both Hopfield and Hinton pioneered the earliest developments of artificial neural networks, co-opting statistical and computational physics to build systems capable of storing and recreating information.

Climate science too has benefited from the ability of neural networks to expand modelling capabilities, while healthcare is increasingly implementing artificial intelligence (AI) technology to analyze and diagnose disease. 

In praising the work of Hopfield and Hinton, Moons also recognized the consequences of its misuse. 

"While machine learning has enormous benefits, its rapid development has also raised concerns about our future. Collectively, humans carry the responsibility for using this new technology in a safe and ethical way for the greatest benefit of humankind."

Hinton: 'Machine Learning will exceed people in intellectual abilities'

Now a Nobel Laureate, British-Canadian computer scientist and cognitive psychologist Geoffrey Hinton spoke to the press shortly after the winners were announced.

"I am flabbergasted, I had no idea this would happen, I am very surprised," Hinton said of his reaction to the news.

Nobel laureates are informed of their winning the prize as close to the announcement as possible. On occasion, they cannot be reached, such is the importance of keeping the identity of each prize's recipient a secret.

Hinton said with certainty that advancements in neural networks realized today will have a huge influence on humanity.

"This will be comparable with the industrial revolution. Machine learning will exceed people in intellectual abilities," he added.

While he listed the numerous applications, such as healthcare, AI assistants and increases in work productivity, he also pointed echoed Moons' remarks about the threat AI could pose if humans lose control of the technology. 

Hinton also admitted to using ChatGPT4 — a mainstream large language model — often. "I don't totally trust it, as sometimes it can hallucinate," he added.

AI, machine learning and deep learning — simply explained

Terms like machine learning, artificial intelligence and deep learning were used extensively in the Nobel Prize announcement.

Advancements in computer science has led to extensive research in these fields, said Royal Swedish Academy of Sciences Secretary-General, Hans Ellengren.

While AI as an umbrella term used to describe systems that emulate human intelligence, machine learning describes how systems are able to learn from data and improve predictive decision-making.

Statistical physics used by both recipients in their work throughout the 1980s and 1990s laid the foundations for modern AI. Hopfield and Hinton's early work established neural nets capable of retrieving information based on previous inputs.

Today, neural networks are the building blocks of deep learning models, just like neurons are the building blocks in the human nervous system. Just as neurons are tied together by synapses in the brain, artificial neural nets are made up of layers of nodes.

A simple neural network has only a few layers, but complex models must have more than three layers, which gives it the power to solve more complex problems.

'Surprising but justified': physicists tell DW

"Totally justified and totally courageous on the part of the committee. Because although Hopfield is a trained physicist, Hinton is not," theoretical physicist, Tilman Plehn from the University of Heidelberg told DW.

He refers to Hinton as the inventor of deep learning. "Hopfield laid the groundwork and Hinton made it usable. He is a visionary. In the 90s, nobody really wanted to think about this new field. But he didn't give up. He is the picture of an inter-disciplinary researcher," Plehn adds. 

“Physicists like myself use machine and deep learning all the time as a method to derive more power from data,” particle physicist Marumi Kado told DW.  He uses neural networks to interpret billions of pictures taken by a specialized camera of particle collisions, too miniscule for the human eye.

"Transparency in the development and application of AI methods is very important. Nevertheless, a political discussion about the potential dangers of AI is absolutely necessary and must accompany research in computer science, mathematics and physics," saidMichael Krämer, a theoretical physicist from University of Aachen.

'Godfather of AI': Who is Geoffrey Hinton?

Hailed as the "Godfather of AI" and a pioneer in that field, Geoffrey Hinton has previously expressed regret about his role in advancing AI, particularly regarding its potential future impacts. "If I hadn’t done it, somebody else would have," he told the New York Times last year. 

In 2017, the 76 year old co-founded the Vector Institute in Toronto and became its chief scientific advisor. A year later, along with Yoshua Bengio and Yann LeCun, Hinton received the prestigious Turing Award, often called the "Nobel Prize of Computing," for their groundbreaking work in deep learning. The trio, dubbed the "Godfathers of Deep Learning," continued to give public talks together.

In May 2023, Hinton resigned from his position at Google, where he had worked for over a decade so he could freely express his concerns about the risks associated with AI, including its potential misuse, job displacement, and existential threats from advanced systems.

He emphasized the need for collaboration among AI developers to establish safety guidelines and prevent harmful outcomes.

Edited by: Wesley Dockery, Matthew Agius