Sonrise again: SoftBank’s AI ‘stack’ is ready for whatever comes next

Illustrations by Hannah Rose Geraghty

As with any investment, your capital is at risk.

It was the dawn of President Trump’s new term and, once again, Masayoshi Son was in the right place at the right time. On his first full day, Trump summoned three tech titans to face the press in the White House’s Roosevelt Room.

There was Sam Altman of OpenAI, Larry Ellison of Oracle and, introduced as “my friend Masa”, the beaming, diminutive figure of the SoftBank founder.

The trio was unveiling a $500bn investment in Stargate, the US’s OpenAI-led megaproject. Son, the perennial outsider who somehow always ends up at the centre of the next technological revolution, committed SoftBank to a $100bn investment in “the largest infrastructure project in history”. The proposed public-private network of datacentres and strategic partnerships is a big, bold attempt to assert US AI leadership.

But what became clear was that Son was thinking beyond current breakthroughs in artificial general intelligence (AGI), when computers think like humans, towards a “golden age” of artificial super-intelligence (ASI), when computers go beyond our imagination.

“Artificial super-intelligence will come to solve the issues that mankind never ever thought we could solve,” Son said.

Learning to expect the unexpected 

It was no surprise to Matthew Brett to see the entrepreneur publicly hailing the next technological revolution – Baillie Gifford has held shares in SoftBank for nearly two decades.

Brett has come to expect drama from Son, whose rise from a Korean-Japanese ghetto in Japan’s far south is the stuff of business legend. This is the man who connected his adopted nation to the internet via Yahoo! Japan during the dial-up era, backed Alibaba at the dawn of China’s digital boom and brought the iPhone to his homeland thanks to his partnership with Apple’s Steve Jobs.

He infuriated UK tech investors (notably James Anderson, then Baillie Gifford’s leading investment manager) by snapping up British chip designer Arm at the dawn of the AI era.

“The day after Trump’s inauguration, there he was again on the biggest stage,” Brett says. “In every technological age since the 1980s, Mr Son has managed to back or partner with a winner, in this case OpenAI. With AI, size matters. Clearly, he wants to benefit from the scale-up.”

Son, Brett notes, is putting SoftBank in pole position, “working his magic to get these investments at key points in the technology cycle”. In addition to OpenAI, to which SoftBank has committed up to $40bn since 2024, Son has been assembling an AI empire – partly through the Vision Fund investment vehicle – with Arm as its “crown jewels”.

The scalability of the Cambridge-based firm’s intellectual property and its relevance to AI innovation have put Arm at the centre of the new industrial landscape. Son acquired it for $32bn in 2016. It’s worth more than four times as much today.

Moving into AI in the real world

Now a new phase is opening up. Last October, SoftBank announced the $5.4bn acquisition of the robotics division of Swiss industrial company ABB. Once completed this year, the deal will allow SoftBank to determine the rollout of ready-to-use industrial systems and widen the firm’s robotic footprint in the three-dimensional world.

In this, it provides an essential piece of SoftBank’s chips-to-robots ‘physical AI’ platform, intended to bring world-changing AI technology into factories, warehouses and, eventually, the wider world we all live in.

Brett spells it out: “Son’s interested in using those factories and mechatronics [mixing mechanical engineering and electronics] to make AI-enabled robots.”

For our manager, SoftBank’s ‘full-stack’ approach makes sense. At the pace things are now moving, he thinks it unlikely that Son has a clear idea of which part of the AI agglomeration has the most commercial potential. But he’s determined that SoftBank keeps its options open.

“If the optimism about AI – that it could be bigger than the Industrial Revolution, bigger than the internet – is right, then a blue ocean of opportunity is opening up,” Brett says.

“Son has shown a consistent trend of trying to identify who the major players are in the next technology evolution and getting invested in them. Like the ice hockey star Wayne Gretzky, whoinspired Steve Jobs, he ‘skates to where the puck is going’.”

Brett notes how Son has been a constant presence through half a century of tech booms and busts – always on top of each new wave of innovation, often executing triumphant, era-defining deals, but also backing spectacular wrong horses and shrugging off gasp-inducing losses. 

Market misconceptions of Masa

Brett admires his vision of where technology will take humanity, even more so Son’s ability to focus firepower on what’s investable in the short term.

“It’s always been a volatile performance,” he says. “Some things go well, some go badly. He’s prepared to say: ‘Actually, we were wrong, stop it.’ Hitting reverse is perhaps the most difficult thing to do for an entrepreneur. That’s how, over all these cycles, he’s invested big but never come unstuck.”

Having met him several times, Brett finds Japan’s top billionaire “surprisingly humble in conversation”. A psychology PhD himself, he’s intrigued by Son’s combination of long-term strategy and do-it-now! urgency.

But almost as fascinating, in his view, is the market’s misperception of Son as a recklessly undiscriminating ‘gambling man’ – to borrow the title of Lionel Barber’s recent biography.

“In the stock market’s mind, there’s a kind of a Masa sentiment gauge that swings from wild man to genius. But that perception changes far more than the reality,” Brett adds.

“Over the 20 years I’ve been investing in Japan, SoftBank Group has delivered a compounded return for shareholders of a bit over 18 per cent a year. Tokyo’s stock market has delivered a bit over 6 per cent.

“That’s a Berkshire Hathaway level of performance. But while people hang on Warren Buffett’s every word, with Son, who’s been succeeding for 40-plus years, many still think of him as ‘that crazy guy from Japan’. It’s a bit odd.”

It’s the breadth of scope that sets Son apart, Brett suggests. While Silicon Valley pours billions into AI, it’s mostly digging deeper into specialisms. Microsoft, for example, has a relationship with OpenAI to run datacentres, but seemingly lacks the ambition to build robots.

‘Full-stack’ vision and a ruthless focus on what works, Brett argues, are Son’s superpowers.

“It’s both a strength and a weakness that SoftBank doesn’t have many of its own products. If you’re, say, a Microsoft and you’ve built your own AI datacentres, even if you start to lose a bit to a better model, you’d probably keep your own going to see if things improve.

“But with him, if a better large language model partnering opportunity than OpenAI’s emerges, he’d quickly go with that instead.

“The longer we’ve invested, the less likely it becomes that SoftBank’s successes are down to luck. Because he keeps on getting lucky.”

What comes next?

Is there another act for SoftBank after Son (68) retires? Brett doesn’t see it happening soon. Great entrepreneurs tend to keep going.

SoftBank has the pick of Japanese talent, and Brett expects it may one day become a more conventional holding company. But even if Son were to disappear tomorrow, his past big moves would take a decade or more to play out. In the meantime, being a long-term investor in SoftBank means trusting Son – whom Brett likens to a racing driver – to keep control at high speed.

“The challenge with SoftBank is not to anchor yourself too much on where the company is today, because that could change. In 10 years’ time, it could be the robotics businesses that are the crown jewels. Or something completely different.

“All the value Masayoshi Son has created comes from executing these pivots. We want him to keep doing that, and we have to stay calm while he does.”

Son’s ‘stack’: SoftBank’s decade of AI-related buys

*Amounts shown are the headline transaction value, SoftBank’s exact cheque can be smaller where co-investors participated.

Risk factors

The views expressed should not be considered as advice or a recommendation to buy, sell or hold a particular investment. They reflect opinion and should not be taken as statements of fact nor should any reliance be placed on them when making investment decisions.

This communication was produced and approved in March 2026 and has not been updated subsequently. It represents views held at the time of writing and may not reflect current thinking.

Potential for profit and loss

All investment strategies have the potential for profit and loss, your or your clients’ capital may be at risk. Past performance is not a guide to future returns.

This communication contains information on investments which does not constitute independent research. Accordingly, it is not subject to the protections afforded to independent research, but is classified as advertising under Art 68 of the Financial Services Act (‘FinSA’) and Baillie Gifford and its staff may have dealt in the investments concerned.

All information is sourced from Baillie Gifford & Co and is current unless otherwise stated.

The images used in this communication are for illustrative purposes only.

191813  10061945