SoftBank’s CEO Masayoshi Son says, ‘Superintelligent AI could become super dangerous’

Masayoshi Son's comments represent a notable shift in tone from a figure who has invested nearly $65 billion of SoftBank funds into OpenAI and consistently highlighted AI's potential benefits. His concern centers on the rapid, exponential growth of AI capabilities, which he believes necessitates international trust and collaboration to manage the associated risks. This perspective, coming from a major AI investor, suggests a growing recognition within the industry of the complex challenges accompanying technological advancement.

Son's call for global cooperation against the threat of advanced AI, rather than human conflict, underscores a perceived urgency regarding AI governance. His remarks, made at an event on the sidelines of the Science and Technology in Society forum, align with broader discussions about the need for safeguards and coordinated efforts as AI models become more powerful. The presence of Michael Kratsios, White House science and technology policy advisor, at the same gathering further highlights the high-level attention these issues are receiving.

Despite his caution, Son maintains a long-term optimistic outlook, viewing current AI applications like chatbots that browse websites and write code as merely the initial phase. He envisions a future involving physical-world robots. This dual perspective, acknowledging immediate risks while retaining long-term bullishness, is characteristic of many industry leaders. It suggests that while the transformative potential of AI remains a core belief, the practicalities of its safe development and deployment are becoming increasingly critical considerations.