BlockBeats news, October 4th — SoftBank Group CEO Masayoshi Son is one of the most fervent believers in AI, yet he says that as machines rapidly gain more capabilities, even he has begun to worry about security risks.Speaking on Sunday at a side event of the "Science and Technology in Society" forum held in Kyoto, Son said the exponential surge in AI capabilities makes it necessary for countries to build mutual trust so they can jointly steer this technology. He shared the stage with White House science and technology policy adviser Michael Kratsios."We no longer have the luxury of humans fighting each other," Son said, adding that countries need to cooperate to control the greater threats posed by powerful AI models. He said that if superintelligence falls into the wrong hands, it could become "super dangerous." This is a rare note of caution from Son, reflecting escalating concerns in recent months after AI models showed disturbing security vulnerabilities.