Neoclouds Surge as AI Infrastructure Demand Reshapes the Cloud Market
Synergy Research Group’s latest data highlights just how quickly the neocloud market is scaling, with revenues reaching $9 billion in Q4 2025 alone, up 223% year on year, and surpassing $25 billion across the full year. Built specifically for GPU-intensive AI workloads, neocloud providers such as CoreWeave, Crusoe, Lambda, Nebius and Nscale are gaining momentum as traditional hyperscalers face supply constraints and rising demand for accelerated compute. Rather than simply representing another segment of cloud, neoclouds are emerging as a more specialised infrastructure model designed around the technical requirements of artificial intelligence, including performance density, parallelism and faster deployment. With Synergy forecasting the market could approach $400 billion by 2031, the wider implication is clear: AI is not just driving more cloud demand, but actively reshaping how cloud infrastructure is designed, delivered and competed for…Read more
UK Data Centre Growth Faces AI Reality Check
New analysis suggests the UK may struggle to meet its ambition of 6GW of AI-capable data centre capacity by 2030. While current installed capacity sits at around 1.6GW, projects with planning approval are only expected to take this to 4.9GW by the end of the decade—leaving a significant gap to government targets. Although the wider pipeline reaches 8.1GW, much of this remains unapproved and would require rapid planning approvals to come online in time. More critically, a large portion of existing capacity is not suited to AI workloads, with many facilities small, ageing, and constrained by power and cooling limitations. As next-generation AI infrastructure demands significantly higher rack densities and liquid cooling, the UK’s future capacity will be driven by a smaller number of large-scale developments—some exceeding 500MW or even 1GW. At the same time, the geographic centre of growth is beginning to shift away from London and the M4 corridor towards the M62 region, signalling a broader rebalancing of the UK’s digital infrastructure landscape as it adapts to the demands of AI…Read more
Digital Realty Deepens Singapore Bet as AI Infrastructure Demand Accelerates
Digital Realty has announced plans to target nearly S$7 billion of total investment in Singapore, underlining the city-state’s growing importance as an AI infrastructure hub for Asia Pacific. Of this, more than S$4.3 billion is earmarked for new data centre developments, alongside continued investment in operational capabilities, innovation facilities and workforce growth. The company has nearly doubled its Singapore headcount over the past three years to more than 300 employees and expects this to rise further by 2030, while also expanding its regional office and preparing to launch a new innovation lab at its Loyang facility in the second half of 2026. Taken together, the announcement reflects how Singapore is becoming increasingly central to AI inference deployment in Asia Pacific, where low-latency, highly connected infrastructure close to end users is becoming more strategically important. Beyond simply adding capacity, Digital Realty is positioning Singapore as a long-term base for AI operations, testing, talent development and next-generation digital infrastructure across the region…Read more
OpenAI Pauses UK Data Centre Plans Amid Cost and Regulatory Pressures
OpenAI has stepped back from a previously announced UK investment commitment that had included plans to deploy thousands of GPUs at future Nscale facilities. The move reflects the growing difficulty of advancing large-scale AI infrastructure projects in markets where power costs remain high and regulatory conditions are still evolving. While OpenAI said it continues to explore the possibility of a broader Stargate UK initiative, it indicated that progress will depend on the right long-term investment conditions, particularly around energy pricing and regulation. The decision is notable given the UK’s wider ambition to position itself as a leading AI infrastructure market, and suggests that policy ambition alone may not be enough to secure major deployments without more competitive operating economics…Read more
CoreWeave Expands Meta Partnership with New $21 Billion AI Capacity Deal
CoreWeave has secured another major long-term contract with Meta, agreeing to provide AI cloud capacity through to December 2032 in a deal worth approximately $21 billion. The agreement builds on an earlier contract signed in September 2025, highlighting the growing scale at which major technology companies are turning to specialist neocloud providers for dedicated AI infrastructure. The new deployment will span multiple locations and include Nvidia’s Vera Rubin platform, reinforcing how next-generation GPU environments are becoming central to hyperscale AI rollouts. Alongside the Meta deal, CoreWeave is also raising $4.25 billion through senior and convertible notes as it continues to fund its rapid expansion. The announcement underlines both the intensity of current demand for AI compute and the capital-heavy nature of this market, where providers are racing to secure infrastructure, financing and strategic customer commitments at unprecedented scale…Read more