Professor Geoffrey Hinton, known as the "Father of Artificial Intelligence," has updated his predictions regarding the potential threat of AI bringing about humanity's end. He now estimates that the risk could rise to 20% within the next 30 years, up from his previous 10% assessment.
Speaking on BBC Radio 4's Today program, Hinton explained that the change in his outlook wasn't significant, noting, "There isn't much difference, it's between 10 and 20." The program host pointed out that these increasing percentages are becoming more concerning.
Hinton compared humanity's situation in relation to AI to a child facing an adult, suggesting that in the future, humans would be as vulnerable as a three-year-old compared to AI.
Hinton has long been an outspoken critic of uncontrolled AI development. Earlier this year, he resigned from his position at Google to more openly express his concerns.
He highlighted the possibility of AI surpassing human intelligence, a notion that was initially dismissed by many, but now, he warns, this transition might occur sooner than expected.
Hinton also commented on the management crisis at OpenAI and the dismissal of CEO Sam Altman. He praised Ilya Sutskever, the former student who fired Altman, for his emphasis on AI safety.
However, Hinton criticized OpenAI for shifting its focus from safety to profit, calling this shift "very unfortunate."
Hinton's warnings serve as an alarm for tech leaders and lawmakers alike. He stressed the rapid development of AI and the urgent need for stricter control mechanisms and ethical guidelines to safeguard humanity's future.
In recent years, China's advances in AI, particularly with large language models (LLMs) and open-source projects, have become a significant challenge to the U.S. China views its AI progress as a competitive advantage.
Hinton's alerts underscore the critical need for greater awareness of the risks associated with AI, with the actions taken in this field potentially marking a pivotal moment for humanity.
