Artificial Intelligence pioneer and Nobel Peace Prize winner Geoffrey Hinton has called on governments around the world to cooperate in training AI that is not harmful to humans, RT reported. He warned that this technology is developing at a rapid pace and is likely to surpass human intelligence in the near future.
Speaking at the World Artificial Intelligence Conference in Shanghai (China) last weekend, Mr. Hinton said that although countries have their own interests, none of them want AI to dominate humanity. Although he believes that international cooperation in attack applications such as automatic weapons of war, cyber attacks or fake videos to manipulate public opinion is unlikely, he still believes that countries can jointly create an institutional network to orient the development of AI in the direction of not being human-prone.
Mr. Hinton compared the development of AI to raising a baby tiger that could become dangerous as an adult. According to him, there are only two options: train it without killing the owner, or eliminate it. He warned that when AI is smarter than humans, turning it off will no longer be a viable option. We will be like three-year-olds, and they will be like adults, he said.
In a press conference later, Mr. Hinton said that mental countries could easily cooperate on this issue, although he said it could be difficult for the US under the current administration.
In contrast to Mr. Hinton's call, the White House has recently announced an action plan to achieve "global dominance" in the field of AI, through investments, subsidies and the removal of legal barriers to this technology.
Meanwhile, Chinese Prime Minister Li Qiang also announced at the conference that the country will establish an organization to coordinate international cooperation on AI, calling for the formation of a global governance framework for this technology.
