SoftBank-Backed SB Energy Slows $50 Billion IPO as AI Data Centre Boom Faces Investor Doubts
SB Energy, the SoftBank-backed developer building large-scale infrastructure for artificial intelligence, has slowed the timetable for its planned U.S. stock-market debut as investors question the pace and economics of the AI data-centre boom.
SB Energy, backed by SoftBank, OpenAI and Nvidia, has slowed preparations for its planned U.S. initial public offering, which could value the company at about $50 billion. The company is waiting for investor sentiment toward AI infrastructure and data centres to stabilize after concerns emerged over demand for computing power, financing requirements and the pace of AI development. SB Energy still intends to pursue the IPO.
The artificial-intelligence boom has created a new race to build the enormous computing infrastructure required to power the next generation of AI systems.
But one of the companies betting heavily on that future has now encountered a more cautious public market.
SB Energy, a U.S. data-centre developer backed by Japanese investment group SoftBank, has slowed the process for its planned initial public offering after investor interest proved weaker than expected. The company is still pursuing the listing, but is waiting for market conditions to become more supportive.
A $50 Billion Valuation Target
SB Energy is seeking to enter the U.S. public markets at a valuation of roughly $50 billion, according to previous reports. Its proposed offering is expected to raise billions of dollars to help finance the company's expansion of data centres, power generation and related infrastructure.
The company filed its IPO prospectus in September but did not initially provide a final listing date or offering price. Its planned Nasdaq listing would make SB Energy one of the largest publicly traded companies focused on the infrastructure required by the AI industry.
The size of the proposed valuation has attracted attention because SB Energy's data-centre operations are still in the development stage. The company has not yet brought a data centre online, although it has a substantial pipeline of projects under construction and development.
Why the IPO Has Slowed
The immediate problem is investor uncertainty over the future demand for AI computing infrastructure.
AI companies have spent enormous amounts on computing capacity as they develop increasingly powerful models. That spending has fueled a rapid expansion in data-centre construction, electricity generation and semiconductor demand.
Investors are now examining whether that growth can continue at its current pace. Concerns include improvements in AI model efficiency, the possibility of slower development by major AI companies, regulatory pressure and resistance to new data-centre projects because of their enormous energy requirements.
Potential valuation: About $50 billion
Data-centre capacity under contract: 8.8 gigawatts
Planned investment required for current projects: About $174 billion
Nvidia investment: $3 billion in total, including a recent $1.5 billion investment
Potential additional debt: About $4.9 billion
The OpenAI Connection
One of the most important parts of SB Energy's strategy is its relationship with OpenAI.
The two companies are connected through major data-centre projects, including a huge planned campus in Ohio that is being developed for OpenAI. SB Energy's own IPO documents say the company is substantially dependent on OpenAI as both a tenant and strategic partner.
The concentration creates both an opportunity and a risk. A long-term customer such as OpenAI can provide visibility for a massive infrastructure project, but heavy reliance on a single customer also means changes in that customer's finances, strategy or computing requirements could affect SB Energy's business.
OpenAI has also been reported to be slowing some aspects of its development strategy, adding to questions among investors about how quickly demand for new computing capacity will grow.
Nvidia Has Put Billions Behind the Strategy
Nvidia is another major financial partner. The chipmaker has invested a total of $3 billion in SB Energy and has committed substantial financial support connected to the company's Ohio data-centre project.
Nvidia has agreed to provide up to $105 billion of backstop support for SB Energy's planned Ohio campus, which is being leased to OpenAI for 20 years. The project is expected to become one of the world's largest data-centre developments.
The involvement of Nvidia and OpenAI illustrates how closely the companies building AI models, supplying AI chips and constructing AI infrastructure are becoming financially connected.
A Huge Pipeline — and a Huge Funding Requirement
SB Energy's ambitions are enormous. The company says its current contracts involve 8.8 gigawatts of data-centre capacity, with much of the pipeline concentrated in its planned Ohio development.
The company estimates that the projects needed to fulfil its current contracts will require approximately $174 billion in investment. Most of that funding is expected to come through project-level debt rather than equity.
That financing model creates another challenge. If credit markets become more expensive or investors become less willing to finance data-centre projects, SB Energy could face difficulty raising the capital required to build its pipeline.
SB Energy's planned expansion requires infrastructure spending on a scale far beyond its current operating business. The company has warned in its IPO filing that securing the required amount of financing could become difficult if debt markets tighten.
The Company Behind the AI Infrastructure
SB Energy did not begin as a pure data-centre company. It was founded as a renewable-energy business and later shifted heavily toward AI infrastructure, including data-centre development and power generation.
Its existing operating assets include solar and battery-storage projects in Texas and California, while additional data-centre projects are under construction in Texas.
The strategy reflects a broader change taking place across the energy and technology industries: data centres increasingly require companies to secure not only buildings and servers but also reliable supplies of electricity.
The Bigger AI Infrastructure Question
SB Energy's IPO slowdown does not by itself prove that the AI infrastructure boom is ending. The company still intends to pursue the listing, and demand for computing capacity remains a major driver of investment in data centres around the world.
But the development shows that investors are becoming more selective about how the AI boom is financed and valued.
Building AI infrastructure requires enormous amounts of capital before facilities begin generating revenue. That creates a difficult balance: developers must secure financing years ahead of demand while investors must decide whether future AI computing requirements justify today's valuations and construction commitments.
The question is no longer simply whether AI will require more computing power. It is also whether the growth in demand will be fast enough, durable enough and profitable enough to justify the enormous infrastructure investment now being planned.
What Comes Next
SB Energy still plans to proceed with its IPO and is evaluating market conditions before moving forward. The company is also considering additional debt financing to support its infrastructure pipeline.
The eventual public-market valuation will provide an important test of investor confidence in the AI infrastructure economy. If the company can attract sufficient demand for its shares, it could unlock another major source of capital for its data-centre expansion.
If investor caution persists, however, SB Energy may have to reconsider the timing, pricing or financing structure of its expansion plans.
SB Energy's delayed IPO is a sign that the AI infrastructure race is entering a more demanding phase. The appetite for data centres remains enormous, but investors are now being asked to finance projects that require billions of dollars long before their promised AI demand fully arrives.
AI • Technology • Business • Global Industry
