SoftBank Group Corp. has sold its entire stake in Nvidia Corp., locking in profits of $5.8 billion just as founder Masayoshi Son prepares a fresh wave of investments aimed at building a global ecosystem around artificial intelligence.
The Tokyo-based conglomerate had increased its Nvidia holdings to about $3 billion by March, and the windfall from that sale—combined with strong returns from its Vision Fund—helped SoftBank post a surprise ¥2.5 trillion ($16.2 billion) net income for the fiscal second quarter, far exceeding analyst expectations of ¥418.2 billion, according to Bloomberg News.
In a move reflecting confidence in its growth trajectory, SoftBank also announced a 4-for-1 stock split, effective January 1, aimed at broadening retail investor participation.
Son, 68, is now positioning SoftBank at the center of the AI revolution. The company’s portfolio includes major stakes in OpenAI and Oracle Corp., both of which have benefited from the surge in AI-driven enterprise demand. These holdings have propelled SoftBank’s shares up 78% in the three months ending September—its best performance since 2005.
Analysts remain optimistic. Citigroup’s Keiichi Yoneshima raised his target price for SoftBank shares to ¥27,100, citing the rising valuation of OpenAI, which he projects could reach $500 billion to $1 trillion in the coming years.
Son’s next big bets include a $30 billion commitment to OpenAI, a $20 billion AI data center initiative dubbed Stargate, and talks with TSMC to participate in a $1 trillion AI manufacturing hub in Arizona. SoftBank has also explored acquiring Marvell Technology and plans to spend $6.5 billion to acquire Ampere Computing.
However, the challenge lies in financing these massive ventures and managing exposure to soaring AI valuations. Analysts warn that SoftBank’s recent rally may have already priced in much of the upside.
As Finimize Research noted on Smartkarma: “SoftBank was once a cheap way to gain exposure to Arm and the AI boom—but with valuations stretched and the discount gone, it may be time to book profits.”
AI future and Nvidia stock
All eyes are on Nvidia ahead of its Q3 earnings release on November 19, as the chipmaker’s performance is seen as a proxy for the global AI industry’s health. CEO Jensen Huang insists that artificial intelligence isn’t a bubble like the dot-com craze, even as investor skepticism rises.
According to The Financial Times, Nvidia has already secured $500 billion in AI chip orders over the next five quarters—underscoring its dominance. With a market capitalization of $4.83 trillion, Nvidia is now worth more than Germany’s 2024 GDP, prompting questions about how much higher it can go.
Despite its scale, Nvidia continues to post start-up-like growth. Fiscal second-quarter revenue surged 56% year-on-year to $46.7 billion, driven by its booming data center business, where it sells powerful AI chips such as Blackwell. The company’s gross margin of 72% outpaces even software giants like Microsoft, thanks to its CUDA ecosystem, which locks clients into Nvidia’s hardware and software infrastructure.
However, risks are mounting. Big Tech firms—including Alphabet, Meta, Microsoft, and Amazon—have spent over $360 billion in the past year, much of it on Nvidia hardware. If the generative AI wave fails to deliver profitability, such spending could slow sharply. An MIT study recently found that 95% of AI projects fail to generate value, and reports suggest OpenAI alone lost over $11 billion due to high operational costs.
Analysts say Nvidia remains a growth powerhouse, but with valuations stretched and AI returns uncertain, investors may need to tread carefully.
Where could Nvidia stock be in five years?
Analysts at Motley Fool Stock Advisor caution that while Nvidia’s valuation appears justified by its exceptional growth and profitability, its heavy reliance on the generative AI boom introduces significant risk. If corporate spending on AI slows or the technology’s commercial promise fades, Nvidia’s earnings momentum could stall. In essence, the stock’s fundamentals remain strong—but its future performance hinges on sustained AI demand. For cautious investors, it may be prudent to wait and watch rather than jump in at current levels.