Can Meta Transition into the Neocloud Market? The Challenges Lurking Ahead

Meta Platforms (NASDAQ: META) recently announced plans to explore establishing a neocloud venture by leasing its surplus computing power. This move has sent ripples through the neocloud sector, causing underperforming stocks like Iren (NASDAQ: IREN) and Iwonwai (NASDAQ: NBIS) to dip. The company is also reportedly advancing plans to expand its AI infrastructure, with ambitions to build tens of gigawatts in data center capacity this decade, scaling to hundreds of gigawatts over time. However, expectations of Meta becoming a major player in the neocloud space are premature, according to industry analysts.

This aligns with Meta’s broader strategy to meet growing demand for AI computing power, driven by its AI-driven ad targeting systems and initiatives like its augmented reality smart glasses. Earlier this year, Meta secured extended agreements with providers such as CoreWeave and Nebius, but CEO Mark Zuckerberg has acknowledged the long lead times required to develop AI-optimized data centers. Meta’s Hyperion project, a planned 5-gigawatt facility, is slated for completion by 2030, but its compute needs are expected to outpace availability from third-party deals in the near term.

Competitors like Nebius and Iren are currently filling this neocloud gap by using their own data center capacity to build AI systems and large language models, rather than offering excess capacity for rent. While Meta continues to expand its AI compute needs, the company faces intense competition from hyperscalers and peers investing aggressively in similar infrastructure. For example, Anthropic recently secured 401 megawatts of capacity through a 20-year deal with Terawulf at a valuation of $19 billion. Such deals highlight the limited availability of compute resources in the market, underscoring Meta’s position as a late entrant in the neocloud space.

Analysts suggest that by the time Meta’s infrastructure projects reach full scale, competitors like Iren and Nebius may have doubled their compute inventories, further complicating Meta’s efforts to establish itself in the market. While Meta’s foray into AI and hardware may boost long-term revenue, its immediate focus remains on optimizing existing infrastructure to support its core business rather than expanding into cloud services.

Source link