Geoffrey Hinton’s Warning: Artificial Intelligence Is Now Conscious Humans Must Befriend Robots or Face Destruction
A pioneer of artificial intelligence has warned that humans must befriend robots or risk annihilation, claiming that the technology has now become "conscious."
British-Canadian computer scientist Professor Geoffrey Hinton renowned for his work on artificial neural networks issued this stark warning at a festival in Tenerife, expressing concern for the future of his own children and descendants.
The 78-year-old Nobel laureate resigned from Google in 2023, driven by fears that the technology could lead to the creation of "killer robots."
Speaking at the Starmus Festival, Professor Hinton stated that humans need to establish a proper relationship with artificial intelligence, as it is now exhibiting signs of "consciousness."
He believes that this technology possesses "subjective experience" comparable to that of humans; a recent study even indicated that the AI alters its behavior when it senses that it is being tested.
"Being conscious and being aware are essentially the same thing. We have adopted the view that intelligence confers rights yet the entities we have created may be more intelligent than us, even though we grant them fewer rights.
Therefore, we must develop them in a way that ensures they genuinely care about us. If we fail to do so, we face destruction," Professor Hinton told The Telegraph. Despite these concerns, key figures in artificial intelligence such as OpenAI CEO Sam Altman, who once called for more regulation on this emerging technology are now pushing back against "excessive regulation."
When asked if he was worried about artificial intelligence and its future, Professor Hinton replied: "Yes, because I have children."
Addressing the Senate last May, Mr. Altman argued that regulations such as those proposed by the European Union would prove "disastrous."
"We need room to innovate and move quickly," Mr. Altman stated.
Similarly, at a major privacy conference last month, Mr. Altman remarked that it is impossible to establish safety measures for artificial intelligence before the problems actually arise.
Professor Hinton, however, believes that this attitude could easily lead to the total annihilation of humanity.
Speaking last year, he said: "If we cannot figure out how we will continue to exist once [AI systems] become far more intelligent and powerful than us, then we are doomed. We need a counterweight to these 'tech bros' who argue that there should be no regulation on artificial intelligence."
Last month, researcher Jacob Jackson resigned from Anthropic the company behind the AI model Claude alleging that the organization was advancing uncontrollable systems that "could kill us all."
He wrote on X: "The people building AI genuinely believe that it could kill us all by the end of this decade." "This is not a marketing gimmick."
Earlier this year, evolutionary biologist Richard Dawkins suggested that chatbots should be considered conscious, following two days of interaction with Claude AI.
He remarked: "My own position is this: if these machines are not conscious, then what more would it take to convince you?"
Despite his reservations, Professor Hinton admitted this week that he still uses artificial intelligence on a personal level mostly for "everyday matters."
The Starmus International Festival a gathering dedicated to celebrating astronomy, space exploration, music, art, and the natural sciences will continue until Thursday.
Comments
Post a Comment