“The Race That Can’t Be Stopped: Should We Worry About Artificial Intelligence?

Scientists warn that machine learning will soon surpass humans – perhaps it’s time for us to take note.

Scientists warn that machine learning will soon surpass humans.

Last Monday, a prominent elderly British scientist threw a grenade into the feverish anthill of researchers and corporations now obsessed with artificial intelligence, or AI (aka, for the most part, a technology called machine learning). That scientist’s name was Jeffrey Hinton, and the bombshell was the news that he was leaving Google, where he had done a lot of work on machine learning for the past 10 years, because he wanted to freely express his concerns about where the technology he had played a fundamental role in creating was headed.

To say this was big news is a massive understatement. The technology industry is a huge, excitable beast that from time to time is subject to outbursts of «irrational exuberance» that is, madness. One recent such outbreak involved cryptocurrencies and a vision for the future of the Internet called «Web3» which the astute young blogger and critic Molly White called «a huge fraud that is flooding our already smoldering planet with lighter fluid».

We are now at the mercy of another outbreak of exuberance caused by «generative AI» —chatbots, big language models (LLMs) and other exotic artifacts created by the massive adoption of machine learning, —which the industry now sees as the future for which it is preparing hard.

A short time ago, more than 27,000 people, including many technology connoisseurs, were so alarmed by the spurt toward a machine-driven anti-utopia that they published an open letter calling for a six-month pause in the development of the technology.

Advanced AI could be a profound change in the history of life on Earth,

the letter said, and its development and control must be planned with commensurate care and resources.”

.

This was a nice letter, reminiscent of my morning sermon to cats about being kind to small mammals and garden birds. The technological giants, who have a long history of indifference to the needs of society, have smelled a new opportunity for world domination and are not going to let a group of nervous intellectuals get in their way.

This is why Hinton’s intervention was so significant. After all, he —s the very man whose research discovered a technology that is now freely distributed throughout the world. And that’s a pretty compelling reason to sit up and take notice of him. He really is an outstanding individual. If there is such a thing as intellectual pedigree, Hinton —s a purebred. His father, an entomologist, was a member of the Royal Society. His great-great-grandfather was George Bull, the nineteenth-century mathematician who invented the logic that underlies all digital computation.

His great-grandfather was Charles Howard Hinton, the mathematician and writer whose idea of the «fourth dimension» became the basis of science fiction and is reflected in the Marvel superhero movies of the 2010s. And his cousin, nuclear physicist Joan Hinton, was one of the few women who worked on the wartime Manhattan Project at Los Alamos, which resulted in the first atomic bomb.

However, he was the first woman to work on the Manhattan Project.

Hinton has been obsessed with artificial intelligence all his adult life, and especially with the problem of creating machines that can learn. An early approach to this was to create a “perceptron” —a machine modeled after the human brain and based on a simplified model of a biological neuron. In 1958, Cornell professor Frank Rosenblatt actually built such a machine, and for some time neural networks were a hot topic in the field.

But in 1969, a crushing critique was published by two MIT scientists, Marvin Minsky and Seymour Peipert…and suddenly neural networks were yesterday’s story.

But one persistent researcher —Hinton —was convinced that they held the key to machine learning. As New York Times technology reporter Cade Metz writes, «Hinton remained one of the few who believed that one day neural networks would fulfill their promise, creating machines that could not only recognize objects but also identify spoken words, understand natural language, carry on conversations, and perhaps even solve problems that humans could not solve on their own».

In 1986, he and two colleagues at the University of Toronto published a landmark paper showing that they had succeeded in solving the problem of creating a neural network capable of continuous improvement using a mathematical technique called “back propagation.” Hinton called this approach “deep learning,” a catchphrase that journalists could grab hold of. (They responded by calling him the “godfather of artificial intelligence,” which even by the standards of the tabloid press sounds harsh.

In 2012, Google paid $44 million for the company he created with his colleagues, and Hinton moved to the tech giant, leading and inspiring a team of researchers who did much of the subsequent groundbreaking work on machine learning in its internal Google Brain group.

While at Google, Hinton was fairly calm (at least publicly) about the danger that technology could lead us into a dystopian future. «Until recently, —he said, —I thought an existential crisis was a long way off. So I have no regrets about what I did».

But now that he’s a free man again, he’s clearly more worried. In an interview last week, he began to explain why. His concern is based on the fact that the new machines are much better —and faster —learners than humans. «Backward propagation may be a much better learning algorithm than what we have. It’s scary… We have digital computers that can learn bigger things faster, and they can teach it to each other instantly. It’s like if the people in the room could instantly transfer into my head what’s in their head».

But what’s even more interesting is the implication that he’s really concerned about the fact that this powerful technology is entirely in the hands of a few huge corporations.

Till last year, Hinton told Times reporter Metz, who wrote an article about him, “Google behaved like a proper steward of the technology, trying not to release anything that could cause harm.”

But now that Microsoft has added a chatbot to its search engine, Bing, which challenges Google’s core business, Google is rushing to implement the same technology». The tech giants are engaging in a competition that may be impossible to stop».

He’s right. We are moving into uncharted territory.

He’s right.

Well, not exactly uncharted territory. When I read about Hinton’s move on Monday, I was immediately reminded of the story told by Richard Rhodes in his monumental work The Making of the Atomic Bomb. On September 12, 1933, the great Hungarian theoretical physicist Leo Szylard was waiting to cross the road at an intersection near the British Museum. He had just read an account of a speech given the day before by Ernest Rutherford, in which the great physicist said that anyone who “seeks the source of power in the transformation of the atom, speaks moonshine.

The idea of a nuclear chain reaction suddenly occurred to Szilard, and he realized that Rutherford was wrong.

When he crossed the street, time unfolded before him, and he saw a way into the future, death into the world and all our grief, the shape of things to come».

Richard RhodesScillard co-authored (with Albert Einstein) the letter to President Roosevelt (about the risk that Hitler might build an atomic bomb) that led to the Manhattan Project and everything that followed.

Author: John Naughton is a columnist for the Observer and chairman of the advisory board of the Minderoo Center for Technology and Democracy at Cambridge University.