Saudi Arabia’s Humain Is Raising $5.3 Billion to Build a Gigawatt‑Scale AI Compute Hub

Reports indicate the build‑out could target around 2 gigawatts (GW) of compute capacity, which would represent roughly one‑third of Humain’s longer‑term infrastructure ambitions through 2034.

Gigawatt‑scale AI campuses are becoming the new standard for hyperscale compute facilities, reflecting the massive power and hardware requirements of modern AI models.

Building a Sovereign AI Infrastructure Platform

Humain was created as part of Saudi Arabia’s effort to develop sovereign AI capabilities while also offering compute capacity to global customers. The company is backed by the country’s sovereign wealth fund, the Public Investment Fund, which has made AI infrastructure a key pillar of the Kingdom’s Vision 2030 economic diversification strategy.

The Goldman‑advised financing would complement existing sovereign and infrastructure funding already committed to Humain’s projects. For example, the company previously announced a financing framework of up to $1.2 billion with Saudi Arabia’s National Infrastructure Fund to expand digital and AI infrastructure in the country.

That earlier agreement targeted the development of up to 250 megawatts of hyperscale AI data‑center capacity, demonstrating how Humain is layering multiple financing sources as it scales its infrastructure footprint.

Partnerships With Nvidia, AMD, AWS, and Others

Financing is only one part of Humain’s strategy. The company has also formed partnerships with major global technology firms to secure hardware, networking, and cloud capabilities.

Examples include:

• Nvidia: Humain plans to deploy up to 600,000 Nvidia GPU systems over three years to power AI infrastructure in Saudi Arabia and abroad.
• AMD and Cisco: A joint venture aims to deploy up to 1 GW of AI infrastructure by 2030, beginning with an initial 100‑MW deployment in Saudi Arabia.
• AWS: Collaboration plans include building large AI data‑center zones in the Kingdom to support cloud and AI workloads.
• Qualcomm: The companies are working on AI infrastructure for global inferencing services.

Together, these partnerships create a full stack that combines compute hardware, networking systems, cloud platforms, and AI software ecosystems.

Early Data Centers Already Under Construction

Humain has already started building its first AI facilities in the Kingdom. Initial sites in Riyadh and Dammam are expected to launch with roughly 100 megawatts of capacity each, with operations targeted for 2026.

These early deployments serve as pilot infrastructure for the much larger multi‑gigawatt expansion planned later in the decade.

Why the Global AI Infrastructure Race Matters

Humain’s expansion comes during what analysts describe as a global AI data‑center investment supercycle. Artificial intelligence workloads are driving unprecedented demand for compute capacity, electricity, and specialized chips.

Research from Moody’s estimates that global data‑center investment could reach at least $3 trillion over the next five years, reflecting the enormous capital requirements of AI infrastructure.

Other forecasts suggest rapid growth in capacity as well. Knight Frank projects global data‑center capacity could expand from about 62 GW in 2025 to more than 110 GW by 2028, fueled largely by AI workloads.

Within that context, Saudi Arabia is attempting to position itself as a competitive alternative to established compute hubs in the United States, Europe, and parts of Asia.

The Strategic Bet: Energy, Capital, and Geography

Saudi Arabia’s strategy relies on several structural advantages:

• Abundant and relatively low‑cost energy to power energy‑intensive AI data centers
• Sovereign capital from the Public Investment Fund to seed large infrastructure projects
• Partnerships with global chip and cloud companies to access advanced hardware and software ecosystems

If Humain succeeds in securing project financing and deploying the required GPU infrastructure, the Kingdom could emerge as a new regional hub for AI training and inference workloads.

However, execution challenges remain. Large‑scale AI infrastructure depends on reliable power delivery, chip supply chains, cooling systems, and sufficient global demand for compute services. The proposed Goldman‑advised financing package is therefore an important step—but not the final one—in Saudi Arabia’s effort to build a global AI compute platform.