Geoffrey Hinton: The Future of AI and the Risk of Losing Control

The artificial intelligence pioneer warns of the dangers of a future in which machines surpass human intelligence.

Artificial intelligence (AI) is here to stay, and with it, a series of questions and concerns about its evolution and control. Geoffrey Hinton, one of the leading computer scientists behind the development of neural networks, has been a key figure in creating the current technologies that we now see present in tools such as speech recognition, computer vision and language models. However, after years of working at Google, Hinton has made it clear that his concern about the future of AI goes beyond technological advances, warning of the risks of artificial intelligence that could surpass human intelligence.

Since his departure from the technology giant, Hinton has become one of the most vocal disseminators regarding the dangers of out-of-control AI. In various interviews, the scientist has raised disturbing scenarios about what could happen if machines reach intelligence much greater than that of humans. In a recent talk with the American media CNN, Hinton left a shocking statement: “If we don't find a way to coexist with artificial intelligences more powerful than us, we will be finished.” This statement, full of alarm, has resonated strongly in social networks and the scientific world, where the debate about the future of AI intensifies.

In the conversation with CNN, the interviewer asked Hinton if he saw a 10% to 20% chance that AI will wipe out humanity. His response was direct and forceful: “Yes, that possibility exists.” For Hinton, if humans cannot find a solution that allows them to stay in control, the situation could become unsustainable. This message is not just a warning, but a wake-up call for those who develop these technologies. Although AI is already an indispensable tool in many fields, its rapid advancement could endanger the balance we know.

What Hinton proposes as a solution is a curious but intriguing concept: giving machines a “maternal instinct” towards human beings. The scientist highlights that, throughout evolution, nature has developed protection mechanisms in weaker beings through the maternal instinct. According to Hinton, if we create AIs that are smarter than us, we must ensure that they have an empathetic relationship with human beings, something we have not yet managed to do. “Intelligence is only part of a being,” he emphasizes, “we need AIs to have empathy for us, and we still don't know how to do it. But evolution achieved it with mothers, and we should be able to do it too.”

These types of statements highlight the ethical and technical challenge that AI scientists face. While progress in developing more complex models is evident, Hinton warns that without careful focus, we could be building intelligence that surpasses us without being able to control it or engage with it safely. The path to advanced AI poses risks that developers must consider seriously to avoid unforeseeable consequences.

The debate about AI and its future continues to be one of the most discussed worldwide. While some believe that artificial intelligence will bring about an era of unprecedented progress, others, like Hinton, have serious doubts about how we can ensure our safety in a world where machines become more powerful than their creators. Will we be able to control the future we have begun to build, or will machines take control?

FAQ

What is Geoffrey Hinton and what is his relevance in the field of AI?
Geoffrey Hinton is a computer scientist who has been crucial in the development of neural networks, the basis of many current artificial intelligence technologies. He is known as one of the 'godfathers' of AI.
Why is Geoffrey Hinton worried about AI?
Hinton has expressed concern about the risks of AI that can surpass human intelligence, warning of the dangers of losing control of these advanced technologies.
What solution does Geoffrey Hinton propose to avoid total control of AI by machines?
Hinton suggests that AIs must develop a “maternal instinct” toward humans, similar to what mothers have to protect their children, ensuring that machines have empathy toward humans.
What do other experts think about Hinton's concerns?
Other AI experts also share concerns about the risks of developing machines smarter than humans, although not all agree on the need for a “maternal instinct.” Ethics in AI is a highly debated topic.
Is it likely that AI will surpass human intelligence?
Many experts believe that, in the coming decades, AI could reach levels of intelligence higher than that of humans. However, the likelihood of this occurring and its consequences are debated.
What risks does AI smarter than humans entail?
The main risks include loss of control over decisions made by machines, which could act in unpredictable or even hostile ways towards humans if not managed properly.
How can humanity prepare for advanced AI?
Experts suggest that we must address the regulation of AI, the creation of ethical systems, and the development of technologies that ensure safety and human control, while maintaining technological progress.
What is the relationship between maternal instinct and artificial intelligence according to Hinton?
Hinton argues that, just as nature developed protective instincts in mothers, we should be able to implant in AI a form of empathy towards humans to ensure our survival.
Could AI pose a threat to humanity in the near future?
Some experts, like Hinton, warn that there is a real possibility that AI could pose a threat, but the degree of that threat depends on how technological advances are managed.
How can I learn more about the risks of AI?
To stay informed on these topics, we invite you to join our Telegram channel, where we discuss everything related to artificial intelligence and its future implications: Join here.