Superintelligence could turn 'super dangerous' says SoftBank's Masayoshi Son

SoftBank Group Corp. chief Masayoshi Son, one of the biggest supporters of artificial intelligence, has flagged the risks linked to more advanced AI systems. Speaking at an event on the sidelines of the Science and Technology in Society forum in Kyoto on October 4, Son said the rapid and exponential rise in AI capabilities made global trust and cooperation essential. He called on governments to work together to regulate the technology and put safeguards in place, according to a Bloomberg report.

Son appeared alongside White House science and technology policy adviser Michael Kratsios. “We don’t have any more luxury of having human fight against human,” Son said, while arguing that countries must work together to manage the much larger threat from powerful AI models. Superintelligence in the wrong hands can become “super dangerous,” he said.

Must read: Trump-Xi meet: Warm words, sharp divide on AI as US-China moves into ‘superintelligence’ age

His remarks were notable at a time when concerns around AI have grown after security breaches involving AI models in recent months. Son has committed nearly $65 billion of SoftBank funds to ChatGPT maker OpenAI and has long argued that AI can deliver investment returns and wider benefits for humanity.

At the same event, Kratsios spoke about how AI could be used in research. “In automated cloud laboratories operating at industrial scale, robotic systems will conduct hundreds of parallel experiments, generating data that feeds back for real-time analysis and hypothesis refinement,” Kratsios said.

Must read: Former Anthropic researcher Jacob Coxon set to testify as New York City weighs AI rules

Son said AI services will expand at an exponential rate with services like browsing websites or writing code, while using larger amounts of general-purpose chips. He added that the next phase for such programmes would begin next year, with systems helping robots and other machines interact with the physical world. “This is just the beginning," Son said.

Son also repeated his long-term forecast that AI and machines with learning capabilities would make up at least 20 per cent of global gross domestic product by 2040, or $46 trillion.