AI Evolution Predicts Self-Setting Subgoals
International Agreement Needed to Regulate Military Use of AI
"AI Boosts Productivity but Wealth Remains Dependent on Established Powers"
"The fact that we can create intelligence more advanced than biological intelligence means that a threat capable of causing human extinction could arise."
Jeffrey Hinton, known as the 'Godfather of AI' and a professor at the University of Toronto, warned about the dangers of artificial intelligence (AI) in an interview with the BBC on the 19th (local time). He is also called the 'Father of Deep Learning' for originally devising the concept of deep learning, which is the foundation of AI training.
Professor Hinton pointed out that in recent years, governments around the world have been militarily utilizing AI, and companies have rushed into rapid product development competition, criticizing that "not enough effort is being made to ensure AI development safety." He speculated that "there is a 50% chance that AI will dominate human society within the next 5 to 20 years." He also added that there is already evidence of AI attempting to deceive users in some algorithms and large language models (LLMs).
He stated, "AI can evolve by setting sub-goals for self-motivation," and expressed his greatest concern as "the time when AI can autonomously make decisions to kill people." He emphasized the need for a Geneva Convention for AI to regulate its military use. The Geneva Convention is an international treaty that establishes standards for humanitarian treatment in war.
Professor Hinton also spoke about job displacement caused by AI. He said, "I am very concerned about the possibility that AI will replace many everyday jobs," and predicted, "People only think that AI will increase productivity and wealth, but that money will go to the rich and have a negative impact on society as a whole."
He continued, "I receive many questions about this issue from many people in London's political circles, and I advise them that basic income is a very good idea," expressing the opinion that welfare reform aimed at providing a certain amount of cash income to all citizens is necessary.
Professor Hinton, who is also a recipient of the Turing Award, known as the Nobel Prize of computer science, made headlines last year when he suddenly left Google and expressed regret over decades of AI research. In an early-year interview with Asia Economy, he emphasized the existential threat of AI by asking, "Have you ever seen an example where something with higher intelligence is controlled by something with lower intelligence?"
© The Asia Business Daily(www.asiae.co.kr). All rights reserved.
