200 megawatts and a hybrid architecture: how artificial intelligence is leaving Saudi data centers for factories?

The real test of infrastructure begins when algorithms leave climate-controlled server rooms to confront factory dust, field energy consumption, and timing delays on production lines. This specific shift is what Qualcomm has placed at the core of its activities in Saudi Arabia, announcing that sovereign data centers and industrial networks constitute the largest growth drivers targeted in the kingdom and the region, according to أكاش بالخيوالا, chief executive officer and chief financial and operating officer of the company, speaking on the sidelines of the SnapDragon summit in Hawaii.

This vision did not come in isolation from a concrete infrastructure plan, as the approach relies on a declared partnership between Qualcomm and Huawei to build AI computing capacity of up to 200 megawatts starting in 2026, supported by the creation of a dedicated AI engineering center in Riyadh. Dr. Musleh Al-Harithi, Saudi AI advisor, explains that the landscape has moved from a technology-adoption phase to establishing an integrated system that links computing and data centers with field applications, driven by direct operational demand in the energy, industry, mining, smart city and communications sectors, rather than merely importing ready-made cloud solutions.

The shift in computing weight mandates moving inference processing from central servers to edge chips.This technical principle rests on edge artificial intelligence, whereby inference workloads are partially shifted to cameras, vehicles, industrial sensors and medical devices. The primary goal is to address the energy challenge faced by data centers, relieve network pressure, reduce response times, and enhance data privacy by processing data locally instead of sending it entirely to the cloud. In this context, the hybrid computing model that links the cloud with edge devices emerges, including the integration of Arabic language models such as “Allam” with hardware based on Qualcomm processors.

This expansion builds on an existing communications infrastructure that recorded about 13.85 million IoT subscriptions in the kingdom during 2025, supported by more than 30 000 narrow-band stations and achieving 99 percent urban population coverage. This foundation enables five main sectors to benefit directly: energy, oil and gas; industry and mining; smart cities; logistics, transport and tourism; and healthcare. From this base, the next phase emerges in robotics, which combine cameras, wireless connectivity, low-power computing and object tracking.

The operational equation for regional factories shifts from renting cloud interfaces to deploying distributed computing on site.This shift means that tech leaders and engineering teams in the Gulf will see the next challenge extend beyond invoking models via cloud APIs to managing integrated systems that run inference algorithms locally under strict energy and temperature constraints. It also compels industrial and banking establishments to reallocate their budgets between cloud consumption and investment in edge hardware, ensuring that sensitive operational data remain within the facility’s geographic boundaries and avoiding data-transfer costs and latency levels that critical environments cannot tolerate.

The joint effort between Qualcomm and Huawei confirms that building sovereign capacity is no longer limited to reserving central computing capacity, but extends to linking data centers with the smallest devices on the industrial edge. With the rollout of the 200-megawatt architecture and the inauguration of the engineering center in Riyadh approaching, the real bet lies in how Arabic models, led by “Allam”, perform efficiently on field devices and robots, turning computing from a distant central concept into an operational tool that controls production details.