Nobel Prize winner worries about AI ‘rising’

“He is considered one of the most important figures in the history of artificial intelligence (AI) – a visionary leader who helped shape the future of AI.” This is the positive assessment that Gemini, formerly known as Bard – an AI chatbot developed by Google, gave British computer scientist Geoffrey Hinton. Dubbed the “godfather” of AI, he and scientist John J. Hopfield have just become winners of this year’s Nobel Prize in Physics thanks to their great contributions to machine learning technology.

Hinton made headlines when he left Google in 2023, raising concerns about the wave of misinformation, the potential for AI to disrupt the job market and the existential risks posed by creating digital intelligence. actually caused.

This year, when he was honored by the Royal Swedish Academy of Sciences as the winner of the Nobel Prize in Physics, Hinton was extremely surprised. When asked about the potential of the technology he helped develop, he said “AI will have a huge impact” on society.

“It can be compared to the industrial revolution. But instead of surpassing humans in physical strength, it will surpass humans in intellectual ability. We have no experience of what will happen when there are things that are smarter than humans,” he said in a phone interview after the Nobel announcement.

Hinton was born in Wimbledon, London, England in 1947. The path that Hinton pursued was perhaps inevitable, because he came from a family rich in scientific tradition. His great-grandfather, George Boole, was the mathematician who invented Boolean algebra, which laid the foundation for modern computers. His cousin, Joan Hinton, was a nuclear physicist who worked on the Manhattan Project, which built the world’s first nuclear weapons during World War II. His father was Geoffrey Taylor, a respected scholar and member of the Royal Society (London’s Royal Society for the Advancement of Natural Knowledge) – the world’s oldest scientific academy.

Hinton once remembered what his mother told him as a child: “Become a scholar or be a failure.” And it seems he followed this advice.

Hinton studied at the University of Cambridge and switched between physiology, philosophy and physics before earning a degree in experimental psychology in 1970. He then studied at the University of Edinburgh and received his doctorate. in AI in 1978. Despite being discouraged by professors, Hinton still decided to develop a non-traditional computer network that imitates the neural nodes and structure of the human brain. He began researching systems called “artificial neural networks” and completed his postdoctoral studies at the University of California, San Diego.

In 1982, Hinton joined the teaching staff at Carnegie Mellon University. There, he worked with psychologist David Rumelhart and computer scientist Ronald J. Williams to develop an algorithm that works backwards from output to input when measuring error. This process is called “backpropagation” and was published by the trio in 1986 in an influential paper that helped lay the foundation for the development of artificial neural networks.

Backpropagation is a neural network training technique, hailed as the “missing piece of the mathematical puzzle” that helps rapidly improve machine learning technology. Thanks to this technique, humans do not need to constantly tinker with neural networks to improve performance, they can do it themselves. This is a key factor with chatbots that millions of people are using every day. Each chatbot relies on a neural network structure trained on large amounts of text data to interpret commands and generate responses.

The famous chatbot ChatGPT is also well aware of the importance of backpropagation for its development. It describes the technique as a “key breakthrough that enables ChatGPT to adjust parameters so that predictions or responses become more accurate over time”.

In his research life, he often made unexpected decisions. In 1987, while working as an associate professor at Carnegie Mellon University, USA, he left his position and went to Canada. One of the reasons for this decision is technology ethics. At that time, most US AI research was funded by the US Department of Defense and Hinton opposed the use of AI for warfare. In Canada, he continued his research as a professor at the University of Toronto.

Another notable milestone came in 2012, when Hinton and two other researchers — including later OpenAI co-founder Ilya Sutskever — won a competition to build a computer vision system. Can recognize hundreds of objects in images. Together with Alex Krizhevsky and Sutskever, Hinton founded DNNresearch to direct joint work on machine learning. The image recognition system they developed, called AlexNet, was so successful that it attracted the attention of “giant” Google.

Google acquired DDNresearch in 2013. Hinton joined Google Brain, the company’s AI research group, eventually being named senior engineer and vice president. He continues to receive prestigious awards for his contributions, including the 2018 Turing Award for groundbreaking research on artificial neural networks, and the 2022 Royal Society Royal Medal for his work. Pioneer in deep learning.

Hinton predicts AI will revolutionize sectors like healthcare, leading to significant increases in productivity. “But we should also worry about the potentially bad consequences, especially the risk of them getting out of control. I worry that the overall consequence could be systems that are smarter than us will eventually take control,” he warned.

Thu Thao (Synthetic)