Power & Sites · United States

Trump Onsite Power Push Meets AI Data Center Reality

Credited to Data Center Knowledge · datacenterknowledge.com · 4h ago

Photograph · Data Center Knowledge

Trump’s onsite power push highlights a hard reality: most AI data centers will likely need hybrid energy strategies rather than fully off-grid designs.

With data centers dominating industry and policy discussions, it was only a matter of time before President Donald Trump weighed in.

In an Oct. 21 Truth Social post, Trump said new, large-scale data centers should generate their own electricity and return excess power to the local community or grid. He also suggested that developers make financial arrangements with host communities, including payments to residents:

“As the President of the United States, I have been watching Data Centers very closely, and have come to a very strong conclusion. New and large, in some cases gigantic, Data Centers are mandated by me to make their own electricity, and give or sell any excess back to the local community, or grid, and therefore there is electric price increase. In fact, there could be a substantial decrease. Other elements likewise! If companies want to make their Data Centers popular, pay the community, and/or individuals within the community, a ‘dividend.’ I have watched the process closely, and Data Centers that were highly unpopular in a certain area, have become extremely popular when a financial deal is struck. MAKE DEALS, and get them built!!! GREAT FOR THE U.S.A. President DJT”

Related: Could the Societal Value of Workloads Shape Data Center Policy?

The comments are likely to intensify debate across the sector. Many projects already under construction have secured grid connections, signed power purchase agreements (PPAs), or otherwise locked in their energy strategy. If Trump’s position were translated into formal policy, developers would face difficult questions about whether those plans would need to change.

Equipment availability would be another major challenge. Many of the components required for gas-fired, nuclear, and battery-backed onsite power systems are already in tight supply, with some turbine models effectively sold out through 2031.

That would present a serious challenge for data center designers, engineers, developers, and operators, said Patrick Hughes, senior vice president of strategy, technical, and industry affairs at the National Electrical Manufacturers Association (NEMA). Hughes was involved in developing the AI Data Center Energy Performance Framework,co-developed by NEMA, the American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE), and Pacific Northwest National Laboratory (PNNL).

“It’s possible for a standalone AI data center to provide sufficient onsite power to fuel its own operations – whether it’s a combined-cycle plant or solar plus storage, we have the technology,” Hughes said. “But that doesn’t mean it’s feasible for a single AI data center – let alone the whole fleet of AI data centers – to supply 100% of their own power and operate reliably 24/7; that requires significant operational management and maintenance.”

Related: Reliability Standards for AI Data Centers, Crypto Miners Move Closer to Adoption

Permitting timelines, fuel and equipment supply chains, interconnection rules, and community considerations vary widely by location, Hughes added. Those factors will shape what is practical at any given site. For many facilities, the most likely outcome may be a hybrid approach that combines onsite generation and storage with a grid connection.

Why AI Data Centers Can’t Easily Go Off-Grid

There are substantial technical and logistical barriers to any broad requirement that AI data centers supply all of their own power. AI training and inference facilities have power densities and load profiles that differ sharply from those of traditional data centers and from facilities that have historically relied on co-located generation and storage.

At the same time, the AI infrastructure buildout is grappling with enormous demand for electrical equipment and skilled labor across utilities, buildings, transportation, and industrial manufacturing. Whether or not developers are pushed toward a bring-your-own-power model, they are already competing with other sectors for turbines, switchgear, transformers, batteries, and specialized labor.

Related: Permits, Not Power, Halt Google’s Finland Data Centers

Regardless of the onsite resource selected, the AI Data Center Energy Performance Framework offers technical standards and guiding principles for safety, resilience, and grid interaction. Data center designers, engineers, developers, and operators can use it to improve energy and water efficiency while supporting grid reliability and resilience through load flexibility and other grid-interactive operations.

Existing Power Deals and Renewable Commitments Complicate the Shift

A blanket onsite power requirement would also have to contend with existing clean-energy mandates, corporate sustainability goals, and regional differences in the availability of renewable resources. Many AI data centers already have agreements in place to draw power from solar and wind projects.

The Stargate site in Abilene, Texas, offers one example. The project is already using natural gas, fuel cells, and gas engines during the construction phase and to power the first data centers now operating there. Developers are also planning to incorporate solar and wind, and the area’s solar generation potential was one of the reasons the location was selected. Abilene offers strong irradiance, abundant undeveloped land, and no major urban loads nearby. With developers scrambling to find 1.2 GW once all phases are complete, wind and solar are expected to supplement other power sources.

Applied Digital’s AI data center project in North Dakota provides another example. The company has arranged for 400 MW of stranded wind power to serve its Ellendale campus and continues to emphasize renewables in its energy strategy. It also recently signed a deal with data center developer Soluna to use output from Soluna’s 150 MW Briscoe wind farm for a new data center in Texas.

"We have said for a long time that the constraint in AI is energy access, not chips," Soluna CEO John Belizaire said in a press release. "We built Soluna to bring load to the generation, so the power would be ready when the demand arrived."

A Hybrid Power Model May Be the Most Practical Path

If policymakers decide to pursue Trump’s idea, they will likely need to translate it into a more flexible and workable policy. The AI Data Center Energy Performance Framework accounts for that kind of flexibility, avoiding a rigid preference for either bring-your-own-power or grid power. Its guidance on grid-interactive design, onsite energy, and thermal efficiency addresses the role of onsite and offsite renewables, as well as intelligent energy storage, in reducing strain on the grid.

“There’s no single answer, as the right onsite generation strategy will inevitably depend on each facility’s load, site, and regulatory landscape,” Hughes said.

About the Author

Drew Robb has been a full-time professional writer and editor for more than twenty years. He currently works freelance for a number of IT publications.

Want more Data Center Knowledge stories in your Google search results?

Editor's Choice

Industry Voices

Join the entire data center ecosystem September 21 – 23, 2026, at the Gaylord Texan Resort & Convention Center in Dallas, TX.

Chips & SupplyUnited States

Innolight Begins NPO Production on Tower Semiconductor’s Silicon Photonics

Credited to Converge Digest · convergedigest.com5h ago

Useful

Innolight has begun early production of its first near-package optics (NPO) device using Tower Semiconductor’s PH18M silicon photonics process at the foundry’s Newport Beach, California facility. The milestone moves Innolight’s NPO portfolio from development into initial manufacturing, while the company has also completed tape-outs of multiple additional NPO products on Tower’s 300mm silicon photonics platform. The two companies are extending their manufacturing relationship to support higher-density optical interconnects for AI and hyperscale data centers.

The collaboration addresses a significant architectural transition in AI networking: moving optical interfaces closer to the processors and switching silicon they serve. NPO architectures position optical engines near the main ASIC package, potentially reducing the electrical signaling distances associated with conventional front-panel pluggable transceivers. These designs typically require substantially higher optical channel counts, often integrating 16 or more channels within a single device. Tower’s silicon photonics platforms support monolithic integration of multiple optical functions, including wavelength-division multiplexing and demultiplexing, allowing designers to combine more functionality within compact photonic integrated circuits.

Innolight is now a manufacturing customer across four Tower Semiconductor facilities: Newport Beach, California; San Antonio, Texas; Israel; and Japan. The initial NPO production uses Tower’s 200mm PH18M process, while the additional 300mm tape-outs establish a development path for subsequent product generations. The companies have not disclosed shipment volumes, device-level bandwidth, customer deployments, or a schedule for volume production of the newly taped-out designs. The announcement follows Tower’s November 2 earnings update, in which management outlined expanding silicon photonics capacity and expectations for NPO demand during 2027.

Key Technical and Manufacturing Highlights

First NPO device — Early production on PH18M silicon photonics

300mm roadmap — Multiple additional NPO products taped out

Channel density — NPO architectures commonly use 16 or more optical channels per device

Photonic integration — Monolithic WDM multiplexing and demultiplexing

Manufacturing footprint — Newport Beach, San Antonio, Israel and Japan

Deployment status — Early production; no disclosed shipment volumes or hyperscaler deployment

Market application — High-density AI and cloud data center optical interconnects

“Our collaboration with Tower provides us with industry leading capabilities enabling us to accelerate the development of NPO solutions to meet the evolving needs of hyperscale and AI data center customers.”

The importance of this announcement is the transition from NPO design activity to initial foundry production at a major optical transceiver supplier. Conventional pluggable optics remain central to AI data center networking, but NPO offers an alternative packaging approach for higher channel densities and shorter electrical connections. The production milestone establishes that Innolight has progressed beyond design tape-out on Tower’s 200mm platform, although commercial shipment volumes and system-level performance remain undisclosed.

The manufacturing roadmap is equally significant. Tower’s 200mm PH18M process supports the first Innolight device, while its 300mm platform provides a separate path for additional designs. Scaling photonic integrated circuits requires consistent wafer-level process control, optical performance and manufacturing yield, particularly as more channels and WDM functions are integrated onto a single die.

The announcement also adds context to Tower’s broader AI optics expansion. Its recent collaborations with NewPhotonics, OpenLight and Semtech span laser-integrated photonic ICs, advanced photonic design platforms and high-speed analog interfaces. Innolight’s early NPO production represents another manufacturing milestone within that ecosystem. The next indicators to watch will be volume production qualification, customer adoption, channel configurations and the progression of Innolight’s 300mm designs into manufacturing.

Company Profile — Tower Semiconductor

Core technologies — Silicon photonics, SiGe BiCMOS, RF CMOS, CMOS image sensors, power management and mixed-signal

SiPho platforms — PH18M; PH18DA; 300mm silicon photonics

Manufacturing — 200mm and 300mm fabs in the U.S., Israel, Japan and Italy

AI optics ecosystem — Innolight, NewPhotonics, OpenLight, Semtech and Marvell

Applications — AI optical interconnects, data center networking, coherent optics, NPO and CPO

Related Converge Digest Coverage

Nov 2, 2026 | Silicon Photonics Tower Semi Targets $2B in 2026 Revenue as Silicon Photonics Surges Capacity expansion, NPO demand and 2027 manufacturing targets.

Oct 13, 2026 | Analog / SiGe Tower and Semtech Secure SiGe Capacity for AI Optical Interconnects Dedicated capacity for TIAs, drivers and equalizers.

Sep 17, 2026 | Photonic ICs NewPhotonics and Tower Move 800G/1.6T Laser-Integrated PICs into High-Volume High-volume PIC shipments on the PH18DA platform.

Aug 11, 2026 | NPO / CPO Tower and OpenLight Target 1.6T, NPO and CPO with Cadence Integrated photonics design ecosystem.

Jun 18, 2026 | Coherent Optics Tower + Marvell Milestone: 5 Million Coherent PICs Shipped Commercial scale in coherent silicon photonics.

🌐 We’re tracking the latest developments in optical networking and AI infrastructure at Converge Digest.

Original · Converge Digest

Front

Chips & Supply · United States

Entegris Enters Silicon Carbide and Diamond-Enabled Thermal Management Materials Market for AI Advanced Packaging

Credited to Semiconductor Digest · semiconductor-digest.com · 2h ago

Entegris, Inc., a supplier of critical advanced materials and process solutions for the semiconductor and other high-technology industries, today announced two important milestones in its strategic expansion into the market for silicon carbide (SiC) and diamond-enabled thermal management materials solutions for next-generation AI advanced packaging.

“Growth in AI compute, from model training to inference, is intensifying power and heat-management demands in data centers,” said Olivier Blachier, President, Materials Solution and SVP, Chief Innovation Officer at Entegris. “We are excited to announce that Entegris is entering this fast-growing market to bring necessary innovation to meet our customers’ chip and package level materials demands.”

First, Entegris has completed its acquisition of Advanced Abrasives Corporation, a New Jersey -based specialty producer of diamond-made precision super-abrasive consumables used in hard-substrate polishing processes. The acquisition builds on Entegris’ existing SiC CMP solutions capabilities that span slurries, pads, cleans and surface conditioning, as well as on its intellectual property portfolio in materials science and surface engineering. In connection with the acquisition and following strong customer engagement, Entegris plans to invest in a new manufacturing line in Taiwan to bring these SiC solutions to market, while continuing U.S.-based exploratory research and development for next-generation thermal interface materials.

Furthermore, Entegris has entered into a long-term development partnership with Element Six, a global leader in synthetic diamond, part of De Beers Group, designed to accelerate the integration of diamond-enabled thermal management solutions into advanced packaging. The strategic collaboration will focus on scalable, cost-effective wafer-level and heterogeneous integration approaches that can reduce manufacturing complexity, minimize thermal boundary resistance and support the performance requirements of next-generation logic and memory chips and their 3-D integration.

“Thermal management is emerging as a critical industry challenge. We believe the move from conventional heat-spreading materials to novel materials and engineered thermal interfaces will mark a major shift in semiconductor manufacturing,” said Mr. Blachier. “Our goal is to facilitate better heat dissipation without compromising performance or reliability. We’re helping customers make that transition through materials innovation and integration at scale.”

At its upcoming Capital Markets Day on November 9, 2026, in New York, Entegris will present its advanced thermal management platform strategy.