Masayoshi Son, the billionaire founder of SoftBank Group, delivered a rare and pointed warning about artificial intelligence safety on Sunday. Speaking at the Science and Technology in Society forum in Kyoto, Son stated that the exponential surge in AI capabilities makes it imperative for countries to build trust to harness the technology together. His comments stand in stark contrast to his usual bullish rhetoric, signaling a shift in tone from one of AI's most significant financial backers.
The $65 Billion Stake and Rising Stakes
Son's caution carries weight because of his massive financial exposure to the sector. SoftBank has directed funds totaling nearly $65 billion into ChatGPT creator OpenAI. The 69-year-old Japanese billionaire has spent years championing AI's ability to deliver investment returns and help humanity. However, he noted that superintelligence in the wrong hands can become 'super dangerous,' urging nations to move past human-versus-human conflicts to address the threat posed by powerful AI models.
Global Push for Automated Science
The warning coincided with an announcement by White House science and tech policy advisor Michael Kratsios, who was in Kyoto to launch a U.S.-led 17-country initiative. This push aims to use AI to accelerate scientific research, with representatives from Japan, South Korea, the U.K., Germany, Singapore, and the UAE joining the stage. Notably, China was not included in the initial list of nations endorsing the White House statement. Kratsios described a future where automated cloud laboratories use robotic systems to conduct hundreds of parallel experiments, feeding data back for real-time analysis.
From Code to Physical Robots
Son outlined a three-phase evolution for AI capabilities. He noted that current chatbot technology is already helping AI services browse sites and write code using general-purpose chips. The next phase, which Son predicted will begin next year, involves AI moving toward helping robots and other machines interact with the physical world. This transition from digital agents to physical embodiment marks a critical infrastructure milestone that developers and enterprise architects should watch closely.
Key Takeaways
- Masayoshi Son warned that superintelligence in the wrong hands is 'super dangerous,' urging global cooperation.
- SoftBank has invested nearly $65 billion in OpenAI, making Son's cautionary note significant for market sentiment.
- A new 17-country U.S.-led initiative aims to use AI for automated scientific research, excluding China initially.
- Son projects that AI and learning machines will account for at least 20% of global GDP by 2040.
The Bottom Line
When the guy betting $65 billion on OpenAI starts talking about 'danger,' itβs not just hypeβitβs a risk management signal. Developers building on these models should expect stricter safety guardrails and geopolitical fragmentation in AI infrastructure.