SoftBank Borrows $11.87 Billion to Double Down on OpenAI as AI Doubts Sink Its Shares
SoftBank has taken out an $11.87 billion loan to deepen its bet on OpenAI, but the move has rattled investors and sent its stock falling as questions mount over whether massive AI spending will ever pay off.
The Japanese technology giant secured the $11.87 billion loan to increase its investment in OpenAI, the company behind ChatGPT, according to a report [1]. SoftBank has already committed billions to the artificial intelligence startup, and the new borrowing adds to its growing exposure [1].
But investors are worried. SoftBank's shares fell as concerns about AI spending and returns hit the market [1]. Some analysts question whether AI investments will pay off soon enough, and that uncertainty is weighing on the company's stock [1]. SoftBank has not commented publicly on the report [1].
The jitters around SoftBank are not isolated. Across the technology industry, the enormous sums being poured into artificial intelligence are facing new scrutiny. Nvidia, the world's most valuable chip company, has built a $99 billion investment portfolio, backing firms that depend on its own AI hardware [2]. That strategy has put Intel and CoreWeave under pressure to prove they can generate enough revenue to justify the investment [2]. Analysts warn that Nvidia is not just a passive investor — it is actively shaping the market by funding companies that buy its products, meaning a downturn in AI spending would hurt the chip giant twice [2].
In China, Alibaba has committed $10.2 billion to its AI division, but a new competitor is making investors nervous [3]. DeepSeek, a Chinese AI startup, recently released a model that rivals Alibaba's own technology at a fraction of the cost [3]. The news raised doubts about whether Alibaba's massive spending will pay off, and the company's stock has struggled to recover from regulatory crackdowns in recent years [3].
Even the leaders of the AI industry are urging caution. Dario Amodei, chief executive of Anthropic, has called on artificial intelligence companies to slow down development of the powerful technology amid growing worries about the risks of superintelligent systems — machines that could think and improve themselves beyond human control [4]. His proposal has won support from OpenAI CEO Sam Altman, Google DeepMind chair Demis Hassabis, and Elon Musk [5]. Amodei has described the idea as "pacing the frontier" — deliberately slowing the race to build the most advanced AI systems rather than stopping development entirely [6].
But Amodei also warns that China poses the "toughest dilemma" for his plan. If the United States slows down alone, China could move ahead [5]. That tension between safety and competition remains unresolved [5].
Meanwhile, the rush toward AI is reshaping labor markets in unexpected ways. In Hong Kong, the labour chief warned that young people face a "paradox of AI fluency" — knowing how to use artificial intelligence is no longer a bonus but a basic requirement that still fails to guarantee a job [7]. Singapore's trade minister urged workers to adopt a "continuous learning" mindset, saying skills become outdated quickly and only those who keep upgrading themselves will stay relevant [7].
In Britain, a pilot project in Preston is using AI training to tackle youth unemployment. The government has launched an "AI bootcamp" for people aged 16 to 24, focusing on young people not in education, employment, or training. After completing the three-week course, participants are offered an apprenticeship [8].