NextFin News - As the 56th World Economic Forum (WEF) Annual Meeting concluded in Davos, Switzerland, on January 23, 2026, the global financial and technological elite left the Alps with a clear mandate: the era of AI experimentation has ended, and the era of massive infrastructure build-out has begun. The week-long summit, which brought together heads of state, CEOs of hyperscalers, and institutional investors, served as the stage for multi-billion dollar announcements that redefine the data center not merely as a utility, but as a critical instrument of national security and economic statecraft.
Among the most significant developments, Hussain Sajwani, founder of Dubai-based real estate giant DAMAC, announced that his company has secured U.S. land and power capacity worth $12 billion for AI data centers. This investment, executed through DAMAC’s digital arm Edgnex, represents more than half of a planned $20 billion expansion in the United States. Simultaneously, the Saudi-based AI firm HUMAIN and the National Infrastructure Fund (Infra) signed a Strategic Financing Framework Agreement of up to $1.2 billion to develop 250 MW of hyperscale AI data center capacity in the Kingdom. These moves, according to reports from The CFO Middle East and Saudi Gazette, underscore a broader trend where Middle Eastern capital is aggressively underwriting the physical backbone of the Western and regional AI economies.
The shift in discourse at Davos 2026 was palpable. While previous years focused on the generative capabilities of Large Language Models (LLMs), this year’s sessions, such as "Economies of the Future: Fast-Forward to 2050," centered on the "AI stack"—the layered infrastructure of energy, hardware, and specialized real estate. U.S. President Trump, who has championed domestic infrastructure growth since his inauguration on January 20, 2025, previously called Sajwani’s commitment an "honor" that keeps the U.S. on the cutting edge. This political alignment highlights how data centers have transitioned from back-office assets to front-line geopolitical priorities.
Deep analysis of these developments reveals that the primary constraint on AI growth has shifted from silicon to sockets. Capital is no longer the bottleneck; electricity and land are. Sajwani’s strategy of locking in one gigawatt (1GW) of power-ready sites gives DAMAC a significant first-mover advantage. In a market where Big Tech and cloud providers are scrambling for grid interconnection, "power-ready" land is the new gold. According to data cited by Oliver Wyman Vice Chair Huw van Steenis, the securitization of U.S. data center debt reached $27 billion in 2025 alone, reflecting a massive appetite for asset-backed lending in this sector. In contrast, the European Union has lagged significantly, with only $0.8 billion in similar securitizations, prompting urgent calls at Davos for European reforms to avoid a permanent "digital gap."
Furthermore, the emergence of "Sovereign AI" was a dominant theme. The HUMAIN-Infra deal in Saudi Arabia is a prime example of how nations are treating compute capacity as a sovereign resource, similar to oil reserves or strategic grain silos. By developing local hyperscale capacity, the Kingdom is ensuring that its data and AI training remain within its jurisdictional control while supporting its Vision 2030 goals. This trend toward "economic statecraft," as Steenis noted, suggests that the global order is moving toward de-risking and strategic autonomy. Investors are no longer looking for the most efficient global supply chain, but the most resilient one.
The financial implications are equally profound. Private credit and bankers at Davos expressed a growing pickiness regarding which data centers to finance. The "valuation gap" is widening between facilities backed by hyperscalers (like Microsoft or Google) and speculative "zombie" data centers that lack guaranteed tenants or swift grid access. As AI compute requirements double every few months, the energy intensity of these projects is forcing a rethink of the "E" in ESG. The launch of the Business4Land Champions’ Council at Davos, aimed at land restoration and drought resilience, indicates that the industry is beginning to grapple with the environmental footprint of its massive physical expansion.
Looking forward, the next 24 to 36 months will be defined by execution risk. While $12 billion in land and power is a formidable start for players like Sajwani, turning that into operational capacity requires navigating a tightening global supply chain for transformers, cooling systems, and specialized GPUs. Moreover, as AI moves from "micro" productivity tasks to "macro" economic drivers, the impact on labor markets will become a central concern for policymakers. The Davos consensus suggests that while AI infrastructure spending may only represent roughly 1% of global GDP—comparable to the U.S. shale boom—its "fat tail" risks, such as sudden shifts in employment or energy shocks, require a philosophical and disciplined approach to risk management. The road from Davos to the upcoming Global Collaboration and Growth Meeting in Jeddah this April will likely see even more aggressive capital deployment as the race for the physical foundation of the AI era intensifies.
Explore more exclusive insights at nextfin.ai.
