NVIDIA did announce an 18,000-unit AI infrastructure deployment with Saudi Arabia’s HUMAIN on May 13, 2025—but calling it a completed sale of 18,000 chips is misleading. The first phase was described as an NVIDIA GB300 Grace Blackwell AI supercomputer connected with InfiniBand networking. HUMAIN, an artificial-intelligence company owned by Saudi Arabia’s Public Investment Fund, also outlined a much larger, multiyear buildout.
Later announcements covered U.S. export authorization, an expansion target of up to 600,000 NVIDIA AI infrastructure technologies and a reported first shipment. Those developments show that the project progressed, but the available public record does not establish that all 18,000 systems were delivered, installed and operating at full capacity.
As an Amazon Associate I earn from qualifying purchases.
What NVIDIA actually announced
On May 13, 2025, NVIDIA announced a strategic partnership with HUMAIN, a Saudi AI company created under the country’s Public Investment Fund.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchWindows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallThe first phase was described as an 18,000-NVIDIA GB300 Grace Blackwell AI supercomputer deployment using NVIDIA InfiniBand networking. NVIDIA also said HUMAIN planned AI “factories” with projected capacity of up to 500 megawatts and several hundred thousand NVIDIA GPUs over five years.
#1 Best Overall
- ATTENTION BUYERS: EXPORTING this item outside the U.S. requires strict compliance with the Bureau of Industry and Security (BIS) U.S. Export Administration Regulations (EAR). Non-compliance carries fines up to $1 MILLION PER VIOLATION AND UP TO 20 YEARS IMPRISONMENT for both the buyer and seller. To uphold our regulatory obligations, we strictly follow BIS "Know Your Customer" guidance by screening each order and immediately aborting transactions showing unresolved regulatory red flags. ***
- [Double-Flow-Through Design] The RTX PRO 6000 Blackwell features a double-flow-through cooling design, optimizing efficiency and airflow to sustain peak performance under 600W power loads. | [5th Gen Tensor Cores] Deliver up to 3X the performance of the previous generation and support for FP4 precision for faster AI model processing times with reduced memory usage, enabling local fine-tuning of LLMs and generative AI | [4th Gen Ray Tracing Cores] Double the ray-triangle intersection rate of the previous generation to create photoreal, physically accurate scenes and immersive 3D designs with RTX Mega Geometry, which enables up to 100X more ray-traced triangles.
- [PCIe Gen 5] Support for PCIe Gen 5 provides double the bandwidth of PCIe Gen 4, improving data-transfer speeds from CPU memory and unlocking faster performance for data-intensive tasks like AI, data science, and 3D modeling. | [GDDR7 Memory] With 96 GB of GPU memory and 1.8 TB ps bandwidth, it can tackle massive 3D and AI projects, fine-tune AI models locally, explore large-scale VR environments, and drive larger multi-app workflows.
- [DisplayPort 2.1] Achieve unparalleled visual clarity and performance, driving high resolution displays at up to 8K at 240 Hz and 16K at 60 Hz. Increased bandwidth enables seamless multi-monitor setups while HDR and higher color depth support ensures superior color accuracy for precision work, such as video editing, 3D design, and live broadcasting.
- [NVIDIA Blackwell Streaming Multiprocessor] The new SM features increased processing throughput, and new neural shaders that integrate neural networks inside of programmable shaders | DLSS 4: Multi Frame Generation ensures ultra-smooth frame pacing for lifelike simulations. | [Universal MIG] Divide a single RTX PRO 6000 Blackwell into multiple isolated instances, each with dedicated resources, allowing for concurrent execution of multiple workloads, optimized GPU utilization, and secure isolation of different applications or users. [WARRANTY] 3 YR Manufacturer's Warranty. Bulk OEM Packaging. Retail Packaging is NOT included.
These figures describe different layers of the proposal:
- 18,000 GB300 systems: the announced first-phase deployment.
- 500 MW: projected capacity for the broader AI-factory program, not necessarily the power draw of the initial installation.
- Several hundred thousand GPUs: NVIDIA’s longer-term five-year ambition for the program.
- Up to 600,000 AI infrastructure technologies: a later expansion target announced by HUMAIN and NVIDIA, not a confirmed delivery total.
The original announcement is available in NVIDIA’s investor release.
Who is buying the systems?
The headline’s “Saudi Arabia” shorthand hides an important distinction. The named partner is HUMAIN, not a retail-style government procurement office or an ordinary independent startup. HUMAIN is state-backed through Saudi Arabia’s Public Investment Fund and was positioned as a full-stack AI company covering infrastructure, models, applications and services.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
That ownership matters because the project fits Saudi Arabia’s broader effort to diversify its economy, build domestic technology capability and become a regional AI infrastructure hub. Still, state backing does not mean the project is guaranteed to generate commercial returns or technological independence.
What GB300 infrastructure means
“18,000 AI chips” is understandable headline language, but it is technically imprecise. The announcement concerned an integrated data-center AI supercomputer deployment built around NVIDIA GB300 Grace Blackwell systems—not 18,000 ordinary standalone consumer graphics cards.
A deployment at this scale involves much more than accelerator silicon. It requires compute servers, high-bandwidth memory, networking, storage, software, power delivery, cooling, data-center construction and specialist operations. NVIDIA’s InfiniBand networking is intended to connect the systems so that large AI workloads can be distributed across many machines.
Consequently, the number of accelerators alone does not reveal how much useful computing Saudi Arabia will obtain. The commercial value depends on whether the infrastructure is installed, available to customers, supplied with electricity and cooling, and used efficiently for real workloads.
Free tools Windows power users keep installed
One-click scans. No signup required.
What will the computing power be used for?
NVIDIA and HUMAIN described several intended applications:
- Training and deploying sovereign AI models.
- Enterprise AI services and cloud computing.
- Arabic-language and regionally relevant AI services.
- Robotics and “physical AI.”
- Digital twins and industrial simulation.
- Manufacturing, logistics and energy applications.
- AI inference for domestic and international customers.
HUMAIN was also expected to use NVIDIA Omniverse to simulate physical environments and support digital twins, robotics and Industry 4.0 projects. Those are planned use cases, however—not independently measured results. The announcement does not by itself prove that the systems have already produced specific productivity gains, deployed autonomous factories or trained a particular commercial model.
Timeline: announcement, authorization, expansion and shipment
- May 13, 2025: NVIDIA and HUMAIN announce the strategic partnership and the first-phase 18,000-GB300 deployment, alongside projected 500 MW capacity and a broader multiyear GPU buildout.
- November 19, 2025: The U.S. Department of Commerce announces authorization for HUMAIN to purchase the equivalent of up to 35,000 NVIDIA Blackwell GB300 chips, subject to security and reporting requirements.
- November 2025: HUMAIN and NVIDIA announce plans for up to 600,000 NVIDIA AI infrastructure technologies over three years across data centers in Saudi Arabia and the United States, with involvement from xAI, Global AI and AWS.
- Late December 2025: A Reuters report syndicated by TradingView says HUMAIN received a first shipment of the latest NVIDIA AI chips.
The chronology is important. A partnership announcement, an export authorization, an expansion target and a reported first shipment are separate events. None of those facts alone proves that the full original 18,000-unit phase is installed and operating at full capacity.
Rank #2
- NVIDIA Volta GV100 Architecture — 4,608 CUDA Cores, 640 1st-Gen Tensor Cores delivering 14 TFLOPS FP32 and 112 TFLOPS deep learning performance for AI training, inference, HPC, and scientific computing workloads
- 32GB HBM2 ECC Memory — 900 GB/s Bandwidth — High-bandwidth memory on a 4096-bit bus with ECC error correction provides the memory capacity and throughput required for the largest AI models, simulations, and datasets
- PCIe 3.0 x16 Interface — 250W TDP — Standard PCIe Gen3 connectivity with passive cooling designed for enterprise rack server deployment in HPE ProLiant, Dell PowerEdge, and Supermicro platforms with adequate chassis airflow
- NVLink — Scale to 96GB Unified Memory — Connect two V100 GPUs via NVLink at 300 GB/s bi-directional bandwidth to scale GPU memory from 32GB to 96GB for larger AI training and HPC workloads
- Multi-Precision Computing — Supports FP64 (7 TFLOPS), FP32 (14 TFLOPS), FP16 (112 TFLOPS) and INT8 precision modes for flexible deployment across training, inference, and scientific simulation workloads
Was this a completed sale?
The most accurate description is a phased, regulated infrastructure buildout:
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Repair Windows errors before they cause bigger problemsFix Now →| Status | What the public record supports |
|---|---|
| Confirmed | NVIDIA and HUMAIN announced the partnership and the planned 18,000-GB300 first phase. |
| Confirmed later | U.S. authorities authorized HUMAIN to purchase the equivalent of up to 35,000 GB300s under specified conditions. |
| Reported | HUMAIN received a first shipment by late December 2025. |
| Not established | That all 18,000 systems were delivered, accepted, commissioned and operating as one completed supercomputer. |
The public announcements also do not disclose the total purchase price, the price per system, the complete delivery schedule, the exact amount of capacity online, named commercial customers or revenue generated by the infrastructure.
Why U.S. export controls matter
Advanced AI systems are subject to U.S. export licensing rules in certain destinations. NVIDIA’s fiscal 2026 annual filing identifies Saudi Arabia among countries affected by U.S. controls on products above specified performance thresholds and explains that some advanced data-center systems require licensing.
The later Commerce Department authorization is therefore central to the story. It confirms that HUMAIN had permission to purchase up to the equivalent of 35,000 GB300s, but it also attached security and reporting conditions. An announced commercial partnership and an approved export transaction are related, but they are not identical.
Those conditions can involve end-use monitoring, reporting and restrictions designed to reduce diversion or unauthorized access. The authorization should not be interpreted as U.S. approval of every later expansion target, including the proposed 600,000 technologies.
For the regulatory background, see NVIDIA’s SEC filing and the Commerce Department statement.
Saudi Arabia is pursuing a multi-vendor AI strategy
The NVIDIA partnership is one part of a broader Saudi infrastructure push rather than an exclusive bet on one chip supplier.
- AMD: AMD announced a $10 billion collaboration with HUMAIN involving planned AI computing centers and up to 500 MW of AMD-based infrastructure across Saudi Arabia and the United States.
- AWS: AWS and HUMAIN announced an AI Zone involving plans to provide, deploy and manage up to 150,000 AI accelerators, including NVIDIA GB300 infrastructure and AWS Trainium chips. See the AWS announcement.
- Qualcomm: Qualcomm announced a memorandum of understanding with HUMAIN covering AI data centers and cloud-to-edge services.
A multi-vendor approach can give Saudi Arabia more negotiating leverage and reduce dependence on one hardware platform. It also creates operational complexity: software compatibility, model optimization, networking, staffing and workload portability become important alongside raw accelerator capacity.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What the project could mean strategically
If the infrastructure becomes operational and attracts paying users, Saudi Arabia could gain:
Recommended Free Tools
- A larger domestic supply of AI training and inference capacity.
- Less reliance on overseas cloud regions for sensitive workloads.
- A platform for Arabic-language models and regional applications.
- Infrastructure for energy, logistics, manufacturing, robotics and smart-city projects.
- The ability to sell AI capacity or managed services to customers in the region and elsewhere.
NVIDIA gains a major strategic customer and a foothold in a market where governments are investing heavily in computing infrastructure. The United States gains a commercial relationship that could keep advanced AI infrastructure within an American-aligned technology ecosystem.
Rank #3
- Professional GPU with Blackwell Architecture
- Blackwell Architecture
- 24GB GDDR7 with PCIe 5.0 & Ray Tracing
- AI Workstation
But hardware does not automatically create a competitive AI industry. Saudi Arabia will still depend on U.S.-designed technology, global manufacturing capacity, networking equipment, software ecosystems, energy and cooling infrastructure, data-center construction and specialist talent. Export controls may also limit how systems can be used and expanded.
How to judge whether the project is substantive
Headline accelerator counts are only the starting point. A more useful assessment asks five questions:
- How much hardware was delivered? Distinguish authorized quantities from shipped quantities.
- How much capacity is operational? Systems must be installed, powered, networked and made available for workloads.
- Who is using it? Look for paying cloud customers, model developers, enterprises or public-sector workloads.
- Is domestic capability developing? Measure local engineering, operations, software and model expertise—not just imported equipment.
- Are compliance requirements being met? Export-control reporting, security and end-use monitoring are part of the project’s practical viability.
Until those questions are answered with operating data, customer disclosures or independent verification, the strongest evidence concerns what was announced and authorized—not the final economic impact.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Does this amount to a “new industrial revolution”?
NVIDIA and HUMAIN used “new industrial revolution” as part of their promotional framing for AI factories and the transformation of industries through AI, digital twins, robotics and physical AI.
The infrastructure could eventually support industrial change, especially in energy, logistics, manufacturing and simulation. But the phrase is not an independently verified economic result. To justify it, the project would need to demonstrate functioning facilities, sustained customer workloads, useful products and measurable improvements in productivity, costs or output.
For now, the defensible conclusion is narrower: Saudi Arabia is attempting to build a large, state-backed AI infrastructure base, and NVIDIA’s 18,000-GB300 first phase is a significant component of that effort.
What this means for companies considering AI infrastructure
Businesses should not assume that a headline-sized national deployment makes equivalent infrastructure immediately available to them. Companies evaluating AI capacity should compare the exact accelerator generation, memory, interconnects, region, data-residency terms, storage and network charges, support model and software compatibility.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Scan for outdated or missing drivers - takes under a minute3Clear out junk files and repair common Windows errorsRelevant options include NVIDIA AI Enterprise for enterprise software, NVIDIA DGX Cloud for managed NVIDIA infrastructure, AWS accelerated-computing instances and AWS Trainium, or AMD Instinct platforms.
These are not consumer products, and enterprise pricing varies by configuration, region, contract and utilization. The Saudi-HUMAIN announcements do not establish a public per-chip or per-system price that can be applied to other buyers.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
