What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Nvidia’s recent AI-infrastructure agreements are best understood as multi-year buildout plans, not as gigawatts of data-center capacity already operating. They connect processor and networking demand with land, construction, electricity generation, grid upgrades, and long-term customers. The announcements indicate a large potential pipeline, but each figure has to be read alongside its project phase, capacity definition, financing, and commissioning status.
They also arrive as electricity demand from data centers is accelerating worldwide. The International Energy Agency (IEA) reported that data-center electricity use grew 17% in 2025, compared with 3% growth in global electricity demand. The IEA expects data-center consumption to double by 2030, with AI-focused facilities using roughly three times their current amount. Those are sector-wide figures, not electricity attributable to Nvidia’s deals alone.
As an Amazon Associate I earn from qualifying purchases.
What the deals actually signal
Nvidia is helping turn expected demand for AI computing into physical infrastructure commitments. A typical agreement may cover accelerators and networking equipment, a customer’s future workload, a data-center campus, a power supply, transmission or distribution work, and the company that will finance and operate the facility.
That bundled structure matters because an announced “gigawatt” can describe several different things:
- IT capacity: the electrical load available to servers and other information-technology equipment.
- Generation capacity: the nameplate output of power plants or other sources that support a site or group of sites.
- Pipeline capacity: a target, option, or prospective deployment that still requires construction, equipment, permits, and grid connection.
- Operating capacity: equipment that has actually been commissioned and is serving workloads.
These categories are not interchangeable. A generation figure can be larger than the IT load it supports, and a pipeline number can remain unbuilt for years. The announcements discussed here do not establish how much of the planned capacity is operating today.
How Nvidia’s announced projects compare
| Announcement | Capacity and definition | Status and timing | Power, construction, or investment terms |
|---|---|---|---|
| OpenAI, September 22, 2025 | At least 10 GW of Nvidia systems for OpenAI’s next-generation infrastructure | Letter of intent; first phase targeted for the second half of 2026 on the Vera Rubin platform | Nvidia said it intended to invest up to $100 billion progressively as systems were deployed |
| PORTS-Pike, Pike County, Ohio, August 17, 2026 | Initial 4.25 IT-GW, plus an option for another 3.75 IT-GW; the summary also refers to 8 IT-GW of AI-factory capacity | Planned phases begin in 2028; the announcement describes secured land, power, and shell capacity rather than an operating campus | At least 10 GW of new generation is intended to support 8 IT-GW; at least $4.2 billion in regional grid infrastructure and a $1.5 billion Nvidia investment in SB Energy were announced |
| IREN, May 7, 2026 | Up to 5 GW of DSX-aligned infrastructure across IREN’s global pipeline, with future deployments expected to focus on the 2-GW Sweetwater campus in Texas | Prospective support for future deployments; not a verified operating total | The announcement describes a pipeline and expected focus, not a completed power or construction program |
The figures should not be added together. The projects have different dates, definitions, scopes, and development stages. The OpenAI and PORTS-Pike announcements are also related to the same broader infrastructure push, so treating them as independent energized capacity would overstate what is known.
OpenAI: a large equipment commitment with a staged rollout
On September 22, 2025, Nvidia and OpenAI announced a letter of intent covering at least 10 GW of Nvidia systems for OpenAI’s next-generation infrastructure. Nvidia said it intended to invest up to $100 billion progressively as those systems were deployed.
The first phase was targeted for the second half of 2026 and was associated with Nvidia’s Vera Rubin platform. The wording describes an intended deployment and a target schedule. It does not certify that 10 GW had been installed, energized, or placed into production. The investment amount is likewise progressive: it is tied to deployment rather than a single payment made at announcement.
For capacity analysis, this is primarily a signal about expected accelerator demand and the scale of facilities OpenAI may require. It becomes evidence of physical capacity only as sites receive permits, connect to the grid, install equipment, and pass commissioning tests.
PORTS-Pike: why IT load and generation must be separated
The August 17, 2026 PORTS-Pike announcement in Pike County, Ohio, is a clear example of an AI infrastructure deal combining computing, real estate, electricity, and financing. Nvidia said it secured land, power, and shell capacity through a partnership with SB Energy. OpenAI would be the customer, while SB Energy would build, own, and operate the data center under a 20-year lease.
Rank #2
- Part number 900-53651-2500-000 and model: P3651
- This is the 2 slot version for when there is no empty slots between 2 slot cards. If you have one or more empty slots between the cards or the cards are 3 slot this NVLink will not work. See the attached images showing the card layout.
- NVLink 3.0 for any brand of RTX Ampere model graphics cards: 3090, A30, A40, A100 / H100 (Requires three NVLinks), A800, A4500, A5000, A5500, A6000
- This is the same as PNY part number: NVLAMP-2SLOT-BSP and RTXA6000NVLINK-KIT
- This is the same as Dell part number: 0RWJ7Y
The initial plan calls for 4.25 IT-GW, with an option for another 3.75 IT-GW. Together, those numbers correspond to 8 IT-GW of potential AI-factory capacity. The same announcement says at least 10 GW of new generation would support that 8 IT-GW. The two measurements describe different layers of the project: IT-GW is the intended computing load, while generation GW refers to power-production capability.
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →The partners also described at least $4.2 billion in regional grid infrastructure and a $1.5 billion Nvidia investment in SB Energy. Planned phases begin in 2028. Those commitments may improve the project’s ability to secure power, but they do not mean the campus is already delivering 8 IT-GW or that the full generation figure is online.
IREN: a global pipeline figure, not a live-capacity count
On May 7, 2026, Nvidia and IREN said they intended to support up to 5 GW of DSX-aligned infrastructure across IREN’s global pipeline. Future deployments were expected to focus on IREN’s 2-GW Sweetwater campus in Texas.
“Up to” describes a ceiling for prospective deployments, not a guaranteed buildout. The 5-GW figure spans a pipeline, and the Sweetwater reference identifies a possible concentration of future work. Neither statement verifies that the entire amount has been constructed, connected, or loaded with Nvidia systems.
How much electricity are AI data centers using?
The most reliable way to answer the question is to separate global sector estimates from individual corporate announcements.
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Clear out junk files and repair common Windows errors3Scan for outdated or missing drivers - takes under a minute| Measure | IEA finding | How to interpret it |
|---|---|---|
| 2024 global data-center use | 415 TWh, about 1.5% of global electricity consumption | A worldwide estimate; local effects can be much larger where facilities cluster |
| 2025 year-over-year change | Data-center electricity demand rose 17%; global electricity demand rose 3% | A reported sector comparison, not a measurement of Nvidia’s projects |
| 2030 base case | About 945 TWh of global data-center electricity consumption | A forecast with assumptions, not a guaranteed outcome |
| AI-focused facilities | Power use is projected to reach roughly three times current levels by 2030 | A scenario projection for AI-oriented data centers worldwide |
| Drivers of net growth | Accelerated servers, mainly associated with AI adoption, could account for almost half of net growth through 2030 | A projection from the IEA’s 2025 analysis |
The IEA’s 2025 Energy and AI analysis presents the approximately 945 TWh figure as a base case and also models outcomes in which efficiency improvements reduce demand or faster AI adoption and more flexible infrastructure increase it. The agency’s April 2026 update adds newer observations and bottleneck analysis. None of these estimates isolates electricity caused by Nvidia’s OpenAI, PORTS-Pike, or IREN agreements.
Rank #3
- Video/Sound Cards
- Passive Cooling
What will supply the additional power?
There is no single global answer because generation mixes and grid conditions differ by region. In the IEA’s 2025 supply analysis, renewables are the fastest-growing source for data-center electricity. Renewables were projected to meet nearly half of demand growth from 2024 to 2030, while natural gas and coal together were projected to meet more than 40% of additional demand in that period. Nuclear power is expected to contribute more after 2030.
Those are scenario results, not observations of future plants. A project can contract renewable energy while still relying on a wider regional grid whose marginal supply includes fossil generation. Conversely, a site with dedicated generation may have a different mix from the surrounding system. The source and timing of electricity therefore matter as much as the headline data-center load.
Can the power grid keep up?
The immediate constraint is often not a shortage of announced demand but the time required to connect and serve it. The IEA’s 2026 update identifies tightening supply chains for gas turbines, transformers, advanced chips, and other IT components. Planning and regulatory systems are also under strain, delaying grid connections and approvals.
Grid connections and permits
A campus can have a customer and a site yet wait for transmission or distribution work, environmental reviews, interconnection studies, and construction permits. These steps determine when a proposed IT load can become a firm operating load.
Equipment availability
Transformers, turbines, advanced chips, and networking equipment all have different manufacturing schedules. A delay in any one of them can leave a building partially complete or prevent a planned phase from entering service.
Variable and rapidly changing demand
AI facilities can produce rapid, large swings in electricity demand. That variability complicates procurement and reliability planning, especially when several large campuses seek service in the same area. The IEA says onsite natural-gas projects remain early-stage in many cases and face technical and financial hurdles; they are not a universal quick fix.
Rank #4
- CUDA Cores: 4608 / NVIDIA Tensor Cores: 576 / NVIDIA RT Cores: 72
- GPU Memory: 24 GB GDDR6 with ECC / Bandwidth: 624 GB/Sec
- System Interface: PCI Express 3.0 x16
- Four DisplayPort 1.4 Connectors
- 3D Stereo Support with Stereo Connector
Storage and flexibility
Onsite battery storage is one emerging response. In September 2026, Nvidia, Google, and Emerald AI announced the AI Energy Management Alliance. Nvidia’s description says a flexible facility could shift workloads, discharge storage, use paired generation, or respond to system contingencies. Such measures may use existing grid capacity more efficiently and reduce stress, but the alliance’s stated benefits are intended outcomes, not proof that flexibility eliminates the need for new generation or grid upgrades.
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 reinstallWill these projects raise electricity bills?
Large data centers can affect local electricity systems, but the available evidence does not establish one bill impact for every community. Costs depend on who pays for generation, substations, transmission, distribution upgrades, contracts, and system balancing. A project that funds dedicated infrastructure may have a different effect from one that relies on broader rate-base investment.
The IEA notes that global data-center use was about 1.5% of electricity consumption in 2024 while warning that local impacts can be much larger because facilities are geographically concentrated. That distinction is important: a national percentage cannot predict a household bill in a county hosting a new campus. Local utility filings, interconnection agreements, tariff rules, and construction schedules would be needed for a project-specific answer.
A practical checklist for reading the next Nvidia capacity announcement
- Identify the date and status. Mark whether the language says letter of intent, partnership, option, planned phase, under construction, commissioned, or operating.
- Classify the gigawatts. Determine whether the number is IT load, generation nameplate capacity, a campus maximum, or a multi-project pipeline.
- Check whether figures overlap. Related announcements may describe the customer, equipment, site, and power supply for one buildout.
- Locate the geography. Grid congestion, generation mix, permitting, and weather conditions are local factors.
- Read the timeline literally. A phase beginning in 2028 is not capacity available in 2026; a target for the second half of 2026 is not confirmation of delivery.
- Assign responsibilities. Note who supplies systems, who owns the land and building, who operates the site, who funds grid work, and who bears the lease obligation.
- Separate sector forecasts from deal totals. IEA TWh estimates describe global data centers; company announcements describe intended projects. Neither can be substituted for the other.
What the announcements mean for capacity planning
For Nvidia, the deals create a route from chip demand to long-duration infrastructure commitments. For data-center developers, they can provide an anchor customer and a clearer case for financing campuses, generation, and grid work. For utilities and regulators, they represent concentrated loads that may require new substations, transmission capacity, generation, storage, and operating rules.
For communities, the relevant questions are practical rather than promotional: when will construction start, what load is contractually committed, which upgrades are customer-funded, how will reliability be maintained, and what happens if a planned phase is delayed? A headline gigawatt number cannot answer those questions by itself.
Bottom line
Nvidia’s AI infrastructure deals point to a substantial potential expansion of data-center capacity, but they are not a tally of energized facilities. The OpenAI agreement is a staged equipment and investment intention; PORTS-Pike combines 8 IT-GW of planned AI capacity with at least 10 GW of proposed new generation and major grid investment; and IREN describes up to 5 GW across a future pipeline. Meanwhile, the IEA’s sector data confirms fast-rising electricity demand and a possible doubling of data-center consumption by 2030.
The deciding factors will be commissioning status, interconnection and permits, equipment supply, generation mix, storage, and workload flexibility. Until those details are delivered project by project, Nvidia’s gigawatt announcements should be treated as indicators of future demand and infrastructure planning—not as electricity already being consumed.
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.
