Geoffrey Hinton, Nobel Prize winner in physics, warns of the dangers of AI: “It will take control of our lives”

At 76, the computer scientist Geoffrey Hinton has just been recognized by the Nobel Prize of Physics. The scientist from the University of Toronto, accompanied by a colleague from Princeton, John Hopfield, was rewarded for “his fundamental discoveries and inventions” which enabled Artificial intelligence can learn.

But, Paradoxically, he himself warns of the danger that this technology represents for humanity.

Now that he is part of the most recognized group of scientists, along with Albert Einstein, Hinton – who is also nicknamed “the godfather of artificial intelligence” – used the media coverage to warn of the dangers of the AI ​​he helped develop.

“We have no experience of what it means to have things smarter than us” declared during the press conference of Nobel Prize winners .

And, to his knowledge, It is possible that in a few more years, AI will be so developed that it will primarily attempt to take control of human life.

Hinton worked in the offices of Google until 2023. At that time, Concerns about the rapid development of AI forced him to leave his job, as he favors stopping the development of this technology, at least until it is safe to continue growing.

Although acknowledged that AI is “going to be wonderful in many ways, in areas like health care,” The British computer scientist said that the possible risks this poses should not be ignored.

“We also have to worry about a number of possible negative consequences. Particularly the threat of these things spiraling out of control.

According to the scientist, The more AI advances, the more dangerous it will become and could even go so far as to rebel against humanity, a plot similar to that offered by science fiction in films like Terminator .

“Look at what it was five years ago and what it is today. At the moment, (artificial intelligence systems) are not smarter than us, but I think they will be soon,” the expert told BBC .

As things progress, the Nobel Prize winner in physics warned that “Within 5 to 20 years, there will be a 50% chance that we will face the problem of artificial intelligence trying to take control of our lives. “.

And while his award recognizes that he himself made the feat of enabling machine learning in artificial neural networks possible, Hinton He said he found comfort “with the normal excuse.” If I hadn’t done it, someone else would have done it.” declared to The New York Times .

A possible uprising in artificial intelligence could be brewing in the coming years, according to forecasts by scientist Geoffrey Hinton, but This technology also raises short-term concerns.

One of them is that The Internet is full of fake photos, videos and texts that an average user will not be able to recognize.

“They will no longer be able to know which is true. »

Additionally, the combination of chatbots that are smarter than humans and being “bad actors” is a “scary” situation for Hinton, since it could allow “authoritarian leaders to manipulate their voters, things like that,” said in a conversation with BBC.

Another of their concerns is how artificial intelligence will begin to revolutionize the job market: His ideal was that this technology could “complement” humans, but all indications are that it will simply replace them and eliminate millions of jobs.

“Takes away the bulk of the work”, but also “I could carry more than that.”

But Hinton’s biggest concern is that technology threatens humanity, that it could create its own autonomous weapons and that they could become “killer robots.”

After working hard in the field, he realized that AI systems “often learn unexpected behaviors, due to the large amount of data they analyze.”

And in general, Companies would let the technology start generating its own codes, he warned.

“People and businesses are enabling AI systems to not only generate their own code, but also execute that code on their own. Some people believed in the idea that these things could become more intelligent than humans. But most thought that was too far.

Continuous: “I myself thought that it was very far away, that it was between 30 and 50 years or more. Obviously, I don’t think like that anymore.”

In this sense, he fears that the competition generated between large companies like Google and Microsoft becoming a global race that is not regulated internationally.

Unlike the creation and management of nuclear weapons, There is now no way to know whether companies or governments are working on AI confidentially.

Although he assured that Google was acting “very responsibly” with AI until 2022, without intending to launch products that could harm humanity, since Microsoft entered the race, the two companies constantly challenge each other.

It is for this reason that he resigned from his position at Google, to be able to freely express his opinions.

“I left so I could talk about the dangers of Artificial Intelligence without considering how it affects Google.

Today, after winning the Nobel Prize, His message is that scientists around the world can collaborate to develop strategies and ways to control AI, before it is too late.