The nascent “neocloud” sector, focused on building dedicated AI computing infrastructure, is generating significant market buzz. These companies represent the cutting edge of AI investment, offering specialized services in contrast to the established, multi-purpose hyperscalers like Amazon Web Services, Microsoft Azure, and Google Cloud Platform.
Neoclouds are aggressively expanding their capacity, often by issuing substantial amounts of debt. While they aim to carve out a profitable niche by offering specialized, cost-effective AI workloads, industry experts caution that achieving profitability may take longer than anticipated. This high-risk, high-reward environment has led to considerable stock volatility. For instance, CoreWeave, a prominent neocloud player, has experienced dramatic price swings, with its stock fluctuating significantly within short periods.
McKinsey consultants have previously highlighted the fragility of the bare-metal-as-a-service (BMaaS) economic model employed by neoclouds, which initially emerged to alleviate GPU shortages. Despite much of the neocloud activity occurring outside public markets, analysts suggest that retail investors should pay closer attention as the AI revolution unfolds. “The heavy lifting on AI development and monetization is happening in places public equity investors can’t access, but public markets’ eyes are moving to AI beneficiaries, and neoclouds are becoming increasingly relevant,” wrote Wolfe Research analysts in a recent note.
Key players in the emerging neocloud space include Lambda Labs, WhiteFiber, Nebius, Crusoe, TensorWave, and Genesis Cloud, with CoreWeave being the largest. CoreWeave, which debuted on Nasdaq in 2025, has seen substantial stock appreciation. Analysts at Wolfe Research have projected significant further upside for the stock, citing its ability to secure crucial GPU capacity and provide rapid compute access to customers grappling with ongoing supply chain bottlenecks.
Amsterdam-based Nebius, also a Nasdaq-listed company, is being recognized by equity researchers as an “emerging AI hyperscaler.” Analysts at Citi have set a positive one-year target price for Nebius, pointing to its strong balance sheet and well-defined funding strategy. The company’s core AI cloud segment has recently achieved EBITDA profitability and projects healthy medium-term EBIT margins, supported by secured capital expenditure plans.
The neocloud strategy hinges on the belief that by focusing exclusively on AI workloads, they can undercut hyperscalers on price. Some industry observers estimate their target pricing to be as low as a quarter of average hyperscaler costs. However, this aggressive pricing strategy is accompanied by a sharp increase in debt levels, raising concerns among both industry veterans and Wall Street analysts.
“I think the neocloud providers are a little delusional about how quickly they think this market is going to take off,” commented David Linthicum, a former chief cloud strategy officer at Deloitte. He believes the market inflection point will be more of a five to ten-year progression, similar to the initial development of cloud computing. CoreWeave, for example, carries a significant debt-to-EBITDA ratio, with aggregate debt estimated to be in the tens of billions. Nebius recently issued billions in debt to fund data center expansion, which initially impacted its stock performance.
The primary concern for neoclouds, as articulated by Linthicum, is the “runway risk”—the possibility that lenders will demand repayment before the companies achieve profitability. This scenario could force a sale at a significantly reduced valuation, potentially leading to acquisition by larger players.
Beyond competitive pressures, neoclouds face the overarching risk that AI may not achieve widespread commercial adoption, leading to demand plateaus before capital expenditures can generate sufficient returns to service debt. Nevertheless, data analytics firms report tangible demand from clients, with AI increasingly integrated into commercial workflows.
“It’s not a closed loop. It’s not fake. People really do need these massive, incredibly expensive chips,” stated Jed Dougherty, senior vice president of AI and platform at Dataiku. He cited the example of SoftBank, a client that significantly enhanced its sales pipeline efficiency through Dataiku’s AI agents, estimating annual savings of 250,000 selling hours. This real-world application underscores the growing necessity for advanced AI computing power and the potential for companies that can deliver it effectively.
Original article, Author: Tobias. If you wish to reprint this article, please indicate the source:https://aicnbc.com/21044.html
