Summary
- Nscale has filed an S-1 for a proposed NYSE flotation under the ticker NSCL.
- Its filing shows $140.6m first-half revenue alongside a $1.02bn net loss and heavy capital requirements.
- The company says it has 55MW of active capacity and a development pipeline extending beyond 10GW.
London-headquartered Nscale has filed for an initial public offering in New York, opening a new source of potential equity funding for one of Europe’s fastest-expanding AI infrastructure developers.
The company has submitted a registration statement on Form S-1 to the US Securities and Exchange Commission and applied to list its ordinary shares on the New York Stock Exchange under the ticker NSCL.
Nscale has not yet set the number of shares to be offered or an expected price range. Goldman Sachs, J.P. Morgan, and Morgan Stanley are acting as lead bookrunners, with a larger banking group participating in the proposed transaction.
The filing puts public-market scrutiny around a business whose growth depends on turning a large pipeline of planned electrical capacity into operating AI infrastructure.
Nscale reported revenue of $140.6m for the first half of 2026, up sharply from the comparable period, while recording a net loss of $1.02bn. The scale of that loss partly reflects the capital-intensive expansion of a company building facilities, leasing capacity, acquiring hardware, and signing long-term infrastructure contracts ahead of full utilisation.
The business says it has 55MW of active computing capacity across owned and leased infrastructure and a power pipeline above 10GW. It has also disclosed more than $100bn of active and contracted value, including large arrangements involving Microsoft, Anthropic, and other AI customers.
Those figures underline both the opportunity and the financing requirement. A gigawatt-scale development pipeline represents far more than servers. Each project requires suitable land, grid connections or on-site generation, substations, cooling plant, buildings, network infrastructure, and expensive computing equipment.
Nscale has already been raising debt and equity at pace. Its recent financing includes billions of dollars associated with US projects, a revolving credit facility, and earlier project financing for infrastructure in Norway.
The proposed IPO would broaden that funding base, but it will also expose the relationship between contracted demand and actual delivery more clearly. Data centre developers can announce very large power pipelines, yet usable capacity is only created after permitting, connection, financing, construction, commissioning, and customer deployment are completed.
The filing also shows the concentration inherent in parts of the AI infrastructure market. Reuters reported that more than half of Nscale’s current revenue comes from one customer. Large future contracts can support development, but concentration increases the financial effect of a change in customer timing, deployment strategy, or credit quality.
Nscale is competing in a market that includes CoreWeave, Nebius, Crusoe, Lambda, hyperscale cloud companies, and other specialist AI infrastructure providers. Their common challenge is that demand is growing quickly while the physical infrastructure required to serve it takes considerably longer to build than software capacity can be sold.
For the UK, Nscale has become part of the wider discussion around sovereign AI infrastructure and domestic compute. The company is involved in British projects while simultaneously raising capital and expanding internationally.
Its decision to seek a New York rather than London listing also reflects the depth of US capital markets for AI infrastructure, where investors have become accustomed to financing very large compute and data centre programmes.
The IPO does not itself solve Nscale’s build-out risk. The value of the company will remain tied to how successfully it converts secured capital, customer contracts, and power positions into live computing capacity — and how much additional financing is required to reach the scale set out in its development pipeline.
