Nscale Seeks $3.5 Billion Before IPO and Redefines AI in Europe

A British artificial intelligence infrastructure startup founded just two years ago is negotiating a $3.5 billion funding round before going public. Nscale, which has already raised over $1.2 billion in private rounds, may conduct its IPO as early as this month. This move illustrates something broader than an isolated operation: the global race for computing power has become the central axis of the AI economy.

The ongoing deal involves two fronts. The first is the sale of $1.5 billion in convertible notes, a type of debt that can be converted into equity in the future, aimed at a group of institutional investors. The second is an additional $2 billion funding from Nvidia, which had already participated in the company’s Series B round in March of this year.

The growth speed of Nscale is, in itself, a relevant data point. The Series A round, closed in December 2024, raised $155 million. A few months later, the Series B, led by the Aker investment fund, raised $1.1 billion. The company itself classified the funding as the largest Series B round in European history.

Now, with the possibility of an imminent IPO and a pre-opening funding of $3.5 billion, Nscale positions itself as one of the most capitalized AI infrastructure companies on the continent. To contextualize, European technology companies rarely reach this level of fundraising at such an early stage. The technology startup scene in Europe has historically lagged behind Silicon Valley in capital volume, but the AI sector has been changing this dynamic.

The factor that most attracts investors' attention is the commercial relationship with major players. Nscale recently signed a contract with Anthropic, the developer of the Claude model, valued at approximately $45 billion. According to market information, the company has communicated to potential investors a projected revenue of around $103 billion, based on infrastructure leasing contracts already signed with clients.

It is essential to understand the nature of this number. The $103 billion does not represent current sales. It is a projection based on long-term contracts signed with clients who lease computing capacity from Nscale. In other words, it is contracted revenue, not recognized revenue.

This type of metric is common in infrastructure companies that operate with multi-year contracts, such as data center providers and cloud operators. Amazon Web Services and Microsoft Azure, for example, also report contractual backlog values that far exceed quarterly revenue. The difference is that these giants have decades of operation and billions in proven recurring revenue.

For a two-year-old startup, the gap between contracted revenue and revenue actually recognized is a risk that investors need to weigh. However, Nvidia's involvement as an investor and potential $2 billion funder acts as a sort of technical validation seal. As we analyzed in previous articles about Nvidia's dominance in the AI chip market, the GPU manufacturer has been selective in investing capital in infrastructure companies, choosing partners that expand demand for its processors.

Nscale's growth is symptomatic of a structural transformation. Computing capacity has become what industry analysts call the "competitive currency" of the artificial intelligence era. Companies like OpenAI, Anthropic, Google DeepMind, and Meta need increasingly larger volumes of processing to train and operate their language models.

This has created a value chain where infrastructure providers have become as strategic as the model developers themselves. The closest historical analogy is the California gold rush: those who profited the most were not the miners, but the sellers of shovels and pickaxes. In the AI ecosystem, the "shovels" are GPUs, data centers, and cloud capacity.

The $45 billion contract with Anthropic is emblematic of this phenomenon. The developer of Claude needs to secure long-term access to computing power to compete with OpenAI and is willing to lock in billion-dollar commitments to do so. The volume of investments in artificial intelligence continues to accelerate, with projections that global spending on AI infrastructure will exceed $300 billion annually in the coming years.

If realized this month, Nscale's IPO will be one of the first major tests of public market appetite for AI infrastructure companies in 2025. Unlike software or language model companies, Nscale operates at the hardware and computing services layer, which implies different margins and more intensive capital needs.

The timing is relevant. The technology IPO market is experiencing a gradual recovery after two years of contraction. Companies that can demonstrate robust contracts and accelerated growth have found favorable windows for going public. Nvidia's support and the significant contractual backlog may serve as convincing arguments for public market investors.

For those following the sector, Nscale's case offers a clear lesson: in the AI economy, those who control the infrastructure control the game. The remaining question is whether the public market will price this thesis with the same generosity that venture capital investors have shown.

This content is provided for general informational purposes only and doesn't constitute financial, investment, legal, or tax advice. Any events, rewards, online promotions, or related information mentioned herein should not be considered a recommendation, solicitation, or invitation to purchase, sell, trade, or otherwise deal in any crypto assets. Crypto assets are highly volatile and may result in loss. The availability of WEEX services, products, and related events may vary by region. You are responsible for ensuring that your participation is in accordance with applicable local laws and regulations.