Dell World Opens, Nvidia Eyes the Gulf, IREN 5GW Deal: AI Infrastructure Week in Review

Table of contents
Dell Technologies World 2026

1. Dell Technologies World 2026 (May 18): Jensen Huang and Michael Dell Unveil Dell PowerRack and AI Factory Advances

Dell Technologies World 2026 opened this morning, May 18, at the MGM Grand in Las Vegas, with a joint keynote by Dell CEO Michael Dell and Nvidia CEO Jensen Huang. Dell announced a broad set of advancements to the Dell AI Factory with NVIDIA, framing the release as the transition from "AI ambition to AI outcomes." The flagship hardware announcement is Dell PowerRack - a new turnkey, rack-scale AI infrastructure product that integrates compute, networking, and storage in a pre-validated unit capable of going from delivery to running live AI or HPC workloads in as little as six and a half hours.

Dell PowerRack for Dell PowerSwitch networking will be available in September 2026, while Dell PowerRack for Dell Exascale storage is planned for the second half of 2026. Additional Day 1 announcements include: new solutions with Google, Hugging Face, OpenAI, Palantir, Reflection, ServiceNow, and SpaceXAI; a new Dell AI Ecosystem Program to standardize partner integration; intent-based networking for large-scale AI fabrics; and the industry's most efficient rack-mount CDU (coolant distribution unit). The Dell Integrated Rack Controller and updated OpenManage Enterprise releases are available now in May 2026. Jensen Huang stated on stage that AI factories are "foundational infrastructure for the global economy," with networking now functioning as "part of the compute fabric itself."

2. Nvidia & Saudi Arabia HUMAIN Deal (May 13): 18,000 GB200 Superchips, 500MW AI Factory Pipeline

On May 13, at the Saudi-U.S. Investment Forum in Riyadh, Nvidia CEO Jensen Huang - attending alongside President Trump, Elon Musk, Sam Altman, and AMD's Lisa Su - announced a landmark partnership with HUMAIN, an AI subsidiary of Saudi Arabia's Public Investment Fund. Nvidia will supply its most advanced AI chips over the next five years; the first phase covers 18,000 units of the GB200 Grace Blackwell AI supercomputer, each equipped with Nvidia's InfiniBand networking, to be deployed in HUMAIN-operated AI factories across the Kingdom with a projected capacity of up to 500 megawatts.

The Saudi Data & AI Authority (SDAIA) will separately deploy up to 5,000 Blackwell GPUs for a sovereign AI factory, while Aramco Digital will develop additional AI infrastructure in partnership with Nvidia. The deal forms part of a broader $600 billion U.S.-Saudi investment framework and was described by Wedbush Securities analyst Dan Ives as a "watershed moment" in the global AI race - placing Saudi Arabia "at the front of the line" for advanced chip access. Nvidia stock surged approximately 5.6% on the day. The deal marks the first major Nvidia hardware commitment to the Gulf since the U.S. rescinded the Biden-era AI diffusion rule (see story 6), removing the previous presumption-of-denial framework for chip exports to Saudi Arabia, the UAE, and Vietnam.

3. Nvidia & IREN Partner on 5 Gigawatts of AI Infrastructure (May 7): $2.1B Investment Right, $3.4B Cloud Contract

On May 7, Nvidia and IREN Limited announced a strategic partnership to deploy up to 5 gigawatts of Nvidia DSX-aligned AI infrastructure across IREN's global data center pipeline. As part of the agreement, IREN issued Nvidia a five-year warrant to purchase up to 30 million ordinary shares at $70 per share - a potential investment of up to $2.1 billion, subject to conditions including regulatory approvals. Nvidia's rights under the warrant vest only as GPU infrastructure is actually deployed across IREN campuses, and fully vest only upon deployment of 600,000 GPUs - tying the investment directly to execution milestones rather than a passive financial commitment.

The anchor deployment will be IREN's 2-gigawatt Sweetwater campus in Texas, which is expected to serve as the flagship implementation of Nvidia's DSX AI factory reference architecture. Alongside the partnership announcement, IREN disclosed a $3.4 billion AI Cloud contract supporting Nvidia's own internal workloads - making Nvidia itself a paying customer of IREN's compute infrastructure. This deal represents Nvidia's most significant move to date from chip supplier to full-stack AI infrastructure integrator, and analysts noted that it follows a clear pattern: after IREN, Corning (manufacturing), Siemens (industrial AI), and OpenAI (mega-scale compute) have all entered structured partnership frameworks with Nvidia in recent months.

4. Dell Expands AI Platform with AMD (May 7): Instinct MI350P PCIe GPUs Added to PowerEdge Servers

On May 7 - one week before Dell Technologies World - Dell Technologies announced an expanded Dell AI Platform with AMD, adding support for AMD Instinct MI350P PCIe GPUs in the PowerEdge XE7745 and R7725 servers. The announcement signals Dell's deliberate positioning of its AI server portfolio as a multi-vendor GPU ecosystem: customers can now choose between Nvidia Blackwell, AMD Instinct MI350P, and Intel Gaudi accelerators depending on workload and cost requirements, all within the same Dell PowerEdge management and deployment framework.

AMD will present dedicated sessions at Dell Technologies World 2026, including EPYC server processor sessions on power and space efficiency, and Instinct MI350 GPU demos for generative AI workloads. The multi-GPU vendor strategy reflects a broader industry trend: while Nvidia dominates large-scale AI training and inference, AMD's MI350 series has gained traction in enterprise deployments - particularly for inference workloads where the cost-per-token economics are favorable compared to Blackwell at equivalent accuracy levels. For enterprise IT buyers, this means Dell-based AI deployments no longer require a single-vendor GPU commitment.

5. Nvidia Q1 FY2027 Earnings Set for Today (May 20): $78B Revenue Consensus, China Revenue Zero, Blackwell Supply in Focus

Nvidia reports Q1 FY2027 financial results today, May 20, after the U.S. market close, with the analyst conference call at 5:00 PM ET. Wall Street consensus is approximately $78 billion in revenue - a 77% year-over-year increase - with data center revenue near $73 billion and non-GAAP EPS of approximately $1.74-$1.77. The guidance explicitly excludes all China Data Center compute revenue, following the indefinite H20 export restrictions imposed in April 2025 that resulted in a $5.5 billion charge. Jensen Huang has estimated the Chinese market at approximately $50 billion in annual addressable revenue, with no clear timeline for restoration.

The three metrics analysts say will move the stock more than the headline beat: Q2 FY2027 revenue guidance (Wall Street modeling $86.6 billion, implying 85% year-over-year growth), gross margin direction (guided around 75%, with any compression a negative signal), and commentary on Blackwell GB300 Ultra production readiness and Vera Rubin NVL72 H2 2026 shipment trajectory. Nvidia stock has beaten revenue consensus for six consecutive quarters yet closed lower on four of the last five earnings reports - a pattern driven by elevated pre-earnings rallies rather than fundamental disappointment. As of the latest available price data, NVDA is trading at approximately $236, up approximately 20% over the past month ahead of the print.

6. U.S. Rescinds Biden-Era AI Diffusion Rule - Gulf Allies Move to Front of the Global Chip Queue

The Trump administration's Commerce Department formally rescinded the Biden-era "AI diffusion rule" in the weeks preceding the Saudi AI summit - a policy that had required special export licenses with a presumption-of-denial framework for AI chip shipments to a broad tier of countries, including Saudi Arabia, the UAE, and Vietnam. The repeal clears the legal path for large-scale Nvidia and AMD GPU exports to Gulf-region sovereign AI programs, which are now being structured as priority strategic partnerships rather than heavily scrutinized commercial transactions.

The policy shift has direct commercial implications for the AI hardware supply chain. With China effectively walled off as a customer for Nvidia's latest GPUs, the Gulf region - Saudi Arabia, UAE, Qatar, and Kuwait - is rapidly emerging as the second-largest addressable geography for high-end AI chip exports, after the United States itself. Analysts at Wedbush and Morgan Stanley noted following the Saudi summit announcements that the Gulf countries collectively represent a potential multi-hundred-billion-dollar GPU and AI infrastructure market over the next five to seven years, and that this demand will flow through the same OEM channel - including Dell PowerEdge servers, HPE Compute systems, and Nvidia networking fabrics - that services U.S. hyperscale deployments.

Editor's Summary

This week delivers the densest cluster of consequential AI infrastructure news since Nvidia GTC in March. Dell Technologies World 2026 opened with the PowerRack - the most significant new product in Dell's AI hardware lineup since the AI Factory launch - and a joint keynote that framed networking as an integral component of compute rather than a separate layer. Nvidia's Saudi Arabia deal and the IREN 5GW partnership together signal that Nvidia is systematically building a vertically integrated global AI factory network, using equity warrants, cloud contracts, and GPU supply commitments to lock in infrastructure partners at scale. The U.S. policy reversal on AI diffusion rules has opened a new geography for GPU exports just as China revenue has gone to zero - a strategic pivot that is already beginning to show up in signed deals. And today's Nvidia earnings report will determine whether the $78 billion Q1 revenue bar and the $86 billion Q2 outlook are achievable in a world that no longer includes China as a customer.

Network-Switch.com is a global professional distributor of networking equipment from Cisco, Huawei, Ruijie, H3C, and our own NS brand - including switches, routers, firewalls, wireless APs, optical modules, and fiber patch cables. Our CCIE, HCIE, H3CIE, and RCNP certified engineers deliver end-to-end technical support and complete network solutions for enterprise, campus, and large-scale infrastructure deployments worldwide.

Visit us: https://network-switch.com

Frequently asked questions (FAQs)

What is the newly announced Dell PowerRack?

Dell PowerRack is a turnkey, rack-scale AI infrastructure product announced by Michael Dell and Jensen Huang at Dell Technologies World 2026. It integrates compute, networking, and storage into a single, pre-validated unit. Its standout feature is speed to deployment: organizations can go from receiving the delivery to running live AI or HPC workloads in just six and a half hours. The networking-focused version will be available in September 2026.

Does Dell's AI platform strictly require Nvidia GPUs?

No. Dell has officially transitioned its AI server portfolio to a multi-vendor GPU ecosystem. In addition to Nvidia's Blackwell chips, Dell's PowerEdge servers (specifically the XE7745 and R7725 models) now support AMD Instinct MI350P PCIe GPUs and Intel Gaudi accelerators. This gives enterprise IT buyers the flexibility to choose their hardware based on specific workload needs and cost-per-token economics.

How is Nvidia legally able to supply massive quantities of AI chips to Saudi Arabia?

The landmark deal to supply 18,000 GB200 Grace Blackwell superchips to Saudi Arabia's HUMAIN was made possible because the U.S. Commerce Department recently rescinded the Biden-era "AI diffusion rule." This rule previously enforced a presumption-of-denial framework for AI chip exports to countries in the Gulf region. With the rule lifted, Saudi Arabia has moved to the front of the line for advanced chip access.

How are U.S. export restrictions to China affecting Nvidia's Q1 FY2027 earnings?

Nvidia's Q1 FY2027 financial guidance assumes zero China Data Center compute revenue. Indefinite export restrictions on H20 chips imposed in April 2025 resulted in a $5.5 billion charge and effectively closed off an estimated $50 billion annual market. However, thanks to surging demand from U.S. hyperscalers and emerging markets like the Gulf region, Wall Street consensus still projects a massive $78 billion in overall Q1 revenue.

What is significant about the Nvidia and IREN partnership?

The 5-gigawatt partnership highlights Nvidia's strategic shift from merely supplying chips to becoming a full-stack AI infrastructure integrator. The deal is highly intertwined: Nvidia receives a five-year warrant for a potential $2.1 billion investment in IREN (tied strictly to deployment milestones), while IREN concurrently secured a $3.4 billion cloud contract to support Nvidia's own internal workloads. Essentially, Nvidia is simultaneously investing in and buying from IREN's compute infrastructure.

Sources