Sijbrandij Foundation
Playbook12345

The Data Center
Power
Playbook.

A living document, provided as an open free resource for the broader AI infrastructure industry to de-risk and streamline power · Updated Nov 11, 2026 (v4)

The AI revolution is now an energy and infrastructure race. Power is the primary bottleneck to scaling compute, and the industry's current defaults of grid-dependent procurement and on-site gas baseload face structural constraints & mounting community opposition that compound over time.

But there is a faster, more reliable, lower-risk path that also earns social license: solar+storage centric energy co-development paired with flexible grid interconnection, enabling firm compute capacity in months rather than years. This is the Playbook for how to execute.

The Problem

Power is now the binding constraint of the AI infrastructure race, interwoven with social license gating the ability to build at all. Energy velocity, paired with enforceable community benefits, will determine who wins it.

Target Scale

500MW+

Per site, per deployment

Time to Power

12–18mo

Not years

Scroll to Explore
Power Playbook
Clean Energy✦AI Infrastructure✦Data Center Energy Initiative✦Power Co-Development✦Flexible Interconnection✦Sustainable Compute✦Energy Velocity✦Bankable Commitments✦Responsible Development✦Sijbrandij Foundation✦Clean Energy✦AI Infrastructure✦Data Center Energy Initiative✦Power Co-Development✦Flexible Interconnection✦Sustainable Compute✦Energy Velocity✦Bankable Commitments✦Responsible Development✦Sijbrandij Foundation✦
Section 1 · The Challenge

The Challenge

Why Now

“Power is now the binding constraint of the AI infrastructure race.”

The exponential growth of AI has created a structural compute shortage. The expansion of compute capacity is constrained by power. Industry forecasts, consistently revised upward, expect US electricity load growth as high as 50% through 2035, with AI data center load on track to reach 9.5–15.3% of total US electricity by 2030 (Lawrence Berkeley National Laboratory, June 2026) – requiring >50GW of new capacity through the end of the decade – and BloombergNEF now projecting ~12% by 2030 and ~20% by 2035 reaching 194GW (US Data Center Capacity Outlook, July 2026), an 83% upward revision against their December forecast.

As of August 2026, US hyperscaler capital expenditure is tracking towards ~$790B in 2026, roughly double 2025 (Currence) — though an increasing share of CapEx growth is driven by component inflation (including memory, transformers, switchgear) beyond volume growth, as hyperscalers directly attributed on Q2 earnings calls when raising CapEx guidance. Annual AI infrastructure spending now exceeds 2% of US GDP, representing the second-largest capital mobilization in American history (Nat Bullard, Halcyon).

AI infrastructure demand is also categorically different from what came before. Traditional data centers moved and stored data, while AI data centers function as factories that output tokens. That shifts the economics of everything downstream, with facility performance instead measured in token output per unit of power. The value of that output is what makes speed, flexibility, and community standing economic variables rather than compliance costs.

As server racks continue to climb in power density, from ~130kW toward ~600kW by 2027 and 1MW by the end of the decade, compute power demand is further amplified. The grid was not built for the pace and scale demanded by AI. Median US grid interconnection timelines now stretch past 5 years and continue to increase.

Power is now the binding constraint of the AI infrastructure race, and energy velocity — defined as the speed that firm, reliable, scalable power can be secured — will determine who wins it.

50%

US electricity load growth through 2035

Driven primarily by AI data centers

~$790B

Hyperscaler CapEx in 2026 alone

~2x YoY; >2% of US GDP

>50GW

New data center capacity needed by 2030

Up from ~25GW in 2024

5+ yrs

Median US grid interconnection timeline

Up from under 2 years pre-2008

Deeper Dives

The exponential growth in AI advancements and usage has created a structural shortage of accelerated compute, with expansion constrained by power. Total global accelerated compute base (compute attributable to AI) stands at ~8GW of installed capacity in 2025, with industry forecasts expecting a 10x increase to ~80GW globally by 2030. Annual performance improvements support an even higher increase in compute output (50x increase in compute expected to be supported by the 10x increase in power by 2030).

The scale of this demand growth is underscored by data from hyperscalers. In May 2026, Google reported processing more than 3.2 quadrillion tokens per month, a 7x year-over-year increase from ~480 trillion in May 2025 (and just ~9.7 trillion the year prior). As of late October, Google’s first-party model APIs alone were processing ~32 billion tokens per minute (Alphabet Q3 2026 Earnings Call). Even with algorithmic and hardware advances improving efficiency, Google was estimated to have needed to increase accelerated compute dedicated to token generation by an estimated 3–10x to meet the 2025 ramp (Altimeter), meaning that a portion of accelerated compute needed to shift away from training to token generation to serve its users. As general purpose computing continues to shift to accelerated computing, hyperscalers shift from CPUs to GPUs, and new AI applications arise, demand will only continue to outpace supply, worsening the structural compute shortage.

The capital commitments match the demand signal. US hyperscaler capital expenditure is now tracking ~$790B in 2026 alone, roughly double 2025 (Currence), with cumulative commitments well past $1T — worth noting that an increasing share of CapEx growth is driven by price rather than volume, as component inflation across memory, transformers, and switchgear drove Q2 2026 guidance raises beyond compute demand (Amazon attributed $20B increase to memory costs on its Q2 earnings call). Google alone has committed $175–185B in annual CapEx, with the majority directed at AI infrastructure. At 2.1% of US GDP, annual AI infrastructure spending now represents the second-largest capital mobilization in American history (% share of US GDP for year specified) — second only to the Louisiana Purchase exceeding the US railroad buildout, the Interstate Highway System, the telecom buildout, and the Apollo program (Nat Bullard, Halcyon, Transition-AI 2026).

Nvidia CEO Jensen Huang estimates $3–4T CapEx on AI infrastructure by end of decade, with 60% directed to chips and computing hardware. Presuming $1.8–2.4T spent on compute server rack modules, this implies 600,000–800,000 server racks ($3M total rack ASP). Power density per rack is accelerating — server racks are currently designed to operate at ~130kW, with industry roadmaps guiding to ~600kW racks by 2027 and beyond 1MW by the end of the decade (Nvidia's platform roadmap through 2030 reaches ~646kW with Rubin Ultra and >1.5MW with Feynman) . At current power density, this implies +80–105GW of additional power demand. As racks become increasingly power dense, this amplifies power demand further.

Underscoring the structural compute and power supply constraint point, there are no dark GPUs sitting idle, every rack deployed is running at max utilization.

What Does This Mean for Power Demand?

Industry forecasts project US electricity load growth as high as 50% through 2035, with AI data center load projected to represent 75% of this total load growth. AI data center consumption alone is now projected at 9.5–15.3% of total US electricity by 2030 (LBNL) vs. 4–5% today. The US now accounts for nearly 40% of global data center power demand, consuming ~313TWh in 2025, more than the entire national generation of the UK, Italy, Spain, or Australia (Energy Institute Statistical Review). BloombergNEF now projects ~12% by 2030 and ~20% by 2035, reaching 194GW in 2035, representing an 83% upward revision against their December forecast. The US needs >50GW of new data center capacity by 2030 (from ~25GW in 2024), consuming >606 TWh of electricity annually. Just meeting industry forecasts for global data center capacity growth through 2030 would require delivering as much new power supply over the next five years as the past two decades of cloud infrastructure development combined.

Third-party modeling from Aurora Energy Research projects 139 to 280GW of US data center load by 2030, with the range driven primarily by pipeline conversion rates rather than underlying demand. Even the low end of this bracket exceeds the entire capacity of any single US utility today; the high end represents nearly two-thirds of current peak US electricity demand. To serve this load, Aurora projects total AI infrastructure investment at $2T for basic data center build-out, $3T for AI-ready deployment with advanced tooling, and $6T fully loaded — approximately 25% of US GDP. The scale of demand is not the question, but rather what fraction of the announced pipeline is actually delivered. Roughly 145GW of the announced pipeline is attributed to sponsors with zero operational data center capacity — entities whose announced projects exceed their execution track record by orders of magnitude. This speculative overhang is the primary mechanical driver of the gap between announced and deliverable capacity.

As of Aug 2026, ~51% of the announced data center development pipeline has yet to break ground, meaning roughly 13GW announced for 2026 delivery cannot energize this year and slips to 2027+. Currence rates 55% of the 283.7GW announced pipeline through 2030 ‘at risk’, with 36% of announced projects currently disclosing no power strategy at all — after credibility-weighting, the pipeline could still add 100+GW through 2030 (Currence Q3 2026 Data Center Outlook, Aug 2026).

Verification economics are now doing the filtering, as utilities attach financial commitments to reduce speculative requests. For example, Exelon's data center interconnection pipeline fell from ~43GW to ~25GW in a single quarter once collateral was attached, with just 4GW backed by signed transmission security agreements and ~$1B in collateral (Exelon Q2 2026). Meanwhile in Texas, PUCT’s adopted large-load rules (effective Oct 2026) set refundable $50,000-per-MW deposits for large loads.

Execution credibility is key when considering announced pipeline numbers, tens of GWs contracted with utilities represent commitments, not yet connected load. Even so, demand forecasts for compute capacity keep proving to be understated.

Why Can’t Supply Keep Up?

Regardless, the physical grid is not built to meet this demand — but this is not an energy shortage. Generation supply exists in abundance; what's scarce is the interconnection, procurement, and delivery infrastructure to convert available energy into firm, fast capacity. Median US grid interconnection timelines exceed 5 years and are increasing — down from under 2 years pre-2008.

PJM queues exceed 5+ years. Texas has offered 18–24 months for large loads, though its framework is currently in transition amid a statewide large-load audit followed by a change in administration. Globally, Japan has delayed data center interconnection to 2029; while Ireland implemented a moratorium on new data center grid connections, but reopened under its 2026 Large Energy User connection policy, requiring dispatchable on-site or proximate generation and creating flexible, non-firm connection paths (Ireland CRU).

The scale is without precedent at the utility level. Arizona Public Service took 140 years to build 8GW of peak load. It now faces over 20GW of uncommitted data center requests, more than double its entire existing system in a fraction of the time (4.5GW committed) (Nat Bullard, Halcyon, Transition-AI 2026).

Equipment procurement compounds the constraint — high-voltage transformer lead times have stretched to 4–5 years, and grid upgrades are increasingly expensive and chronically delayed. Even where generation supply and permitting are cleared, execution is labor-bound. Construction timelines for large energy projects have tripled — from roughly six months to eighteen months to break ground — driven by pre-existing skilled-labor constraints that predated the AI buildout and have been severely compounded by it. EPC and skilled-labor availability — not turbine or equipment supply — is now widely cited alongside power as the greatest binding constraints on speed-to-power. Regulatory reform alone does not compress this constraint.

Even as industry forecasts highlight significant data center load growth, they still likely understate actual demand — forecasts have consistently been revised upward, and current projections likely embed supply constraints as ceilings. Ultimately, the binding constraint is execution, not demand.

AI infrastructure is categorically different from traditional data centers that move and store data. AI data centers function as factories that output tokens. Factory economics are set by output per unit of input. The input is power. How performance is measured has shifted accordingly toward token output per unit of power, and commercial terms are increasingly denominated the same way (tokens per watt, time to first token, dollars per million tokens).

A gigawatt of AI compute capacity carries different revenue figures depending on which layer of the stack is measured (and how much of the value chain is held). At the compute layer, figures cited run $13–15M per megawatt-year (what cloud providers charge to rent out capacity). At the token layer, figures cited can range from $36–100M per megawatt-year (what is earned by serving intelligence on that capacity). While figures may vary, the delay math that follows is consistent, a month of lost time-to-power forfeits a month of revenue at whichever layer is held.

That value explains why speed is worth so much and nearly every behavior documented in the Sections that follow. When the facility is a profit center rather than a cost center, investments that deliver speed and protect the license to operate (flexibility, additionality, community commitments) serve to directly improve facility yield. The defining competition is not who is able to build the largest facilities, but who can power facilities the fastest.

Energy velocity is the defining metric of the AI infrastructure race. It measures the speed at which firm, reliable, scalable power can be secured. As tokenomics above show, energy velocity is where economics concentrate.

The AI compute race will be decided by energy strategy, not server procurement. Four strategic priorities, in order of precedence:

1

Access to Power (Firm Envelope)

Without power, data centers cannot operate. Compute requires a firm power envelope but firmness is a property of the portfolio vs. single source.

Site selection and infrastructure deployment must follow power-first logic. Identify the energization path to a firm power envelope before committing to facility locations. This inverts traditional fiber-first development.

2

Speed to Power (Time to Deployment)

Rapid access to reliable power is strategically decisive. The speed at which power can be secured AND scaled determines time-to-revenue. A six-month advantage in power procurement translates directly to months of accelerated compute deployment, which at the layer economics represents value measured in billions at campus scale.

3

Reliability of Power (Continuity)

GPU infrastructure carries high capital amortization costs. Downtime equals stranded revenue. Firm, reliable power envelope (strongest as a multi-source portfolio) ensures maximum utilization and eliminates the forced shutdown risk inherent in fully grid-dependent or single-source BTM power deployments.

4

Economics of Compute (Optimization)

Cost flexibility exists only within the constraints of the preceding three priorities. Economics must be optimized for total compute cost, not levelized cost of energy (LCOE) in isolation. Power economics are ultimately inputs to compute economics.

This priority ordering is validated by how leading data center developers actually evaluate power decisions today (under a Three C framework): 1) Capacity first — can it deliver firm power on the required timeline; 2) Community second — will it survive regulatory and local scrutiny; 3) Cost third. Energy velocity and social license are the binding constraints, not LCOE.

2026 demonstrated that the fastest path to power must also run through the community. Community and political opposition is now the leading driver of project cancellations (52% of all cancellations — Currence), while responsible-development commitments are increasingly what unlock expedited interconnection, from Pennsylvania's GRID standards to the state frameworks covered in Section 2. The second C is quickly becoming equally as important to the first. A site that clears capacity but fails community still gets cancelled. Responsible development is not a constraint on speed; it is how speed is earned.

Energy is the new currency of compute. There is a direct correlation between energy access, compute deployment, token processing, and revenue generation. Energy strategy is no longer optional. Energy velocity will decide the winners of the AI race.

Industry energy defaults are forming now. The frameworks and standards established today will determine how the next 50GW of global compute capacity gets powered. First-mover advantage accrues to those who solve energy velocity now.

Section 2 · The Risk

The Risk

Why the Current Defaults Fail

“Data centers have become the visible face of rising electricity costs and other resource strains. The industry faces the risk of this reputation setting in permanently, and of community backlash hardening policy from temporary pauses into permanent restrictions.”

Energy ranks among the biggest, most interwoven risks for the data center industry. AI infrastructure is scaling faster than the grid can support, and the data center industry has largely realized that the old power playbook — passive grid-first procurement as a standard industrial load — is broken at GW scale.

The industry's immediate default response has been a rush to procure on-site gas turbines to deliver 24/7 baseload power. However, overreliance on on-site gas baseload recreates and compounds the same infrastructure constraints the industry is trying to overcome while amplifying community backlash — concentrating energy and political risk at the exact moment the industry needs to de-risk.

With current defaults, the data center industry is on a collision course with the communities, regulators, and grid systems it depends on. Energy affordability has already proven to be a top political priority across federal, regional, state, and local levels, as policy increasingly mandates additionality and curtailment to address it. Community and political opposition is now the leading driver of project cancellations industry-wide (52% of all cancellations, Currence). Backlash has consolidated into enacted policy, driving a recent wave of moratoriums.

The window to be proactive is narrowing with most moratoriums being implemented today expiring into decisions by mid-2027. Two policy tracks are being written in parallel today: outright moratoriums, and responsible-development frameworks that set a bar the industry can clear. How the industry meets that bar is what determines whether momentum shifts from the first track to the second. Our expectation coming out of the Nov 2026 election cycle is that more state-level moratoria will come, but that standards likely develop out of them. Data center operators who proactively address energy and community together, now, convert the industry’s largest liabilities into a durable competitive advantage; those who wait will absorb the regulatory costs, lose the permitting battles, and inherit the blame for negative externalities — eroding the timeline to scale compute, or worse, facing outright project cancellation.

~189GW

US gas-fired capacity in development for data centers

Nearly doubled in H1 2026, from 97GW announced through construction (GEM, Aug 2026)

+50%

Gas turbine order prices, 2026 vs 2025 (per kW)

OEM order books now stretch into the 2030s (GE Vernova, Oct 2026)

~$198B

Data center projects blocked or delayed by community opposition in H1 2026

~27% above all of 2025 (Data Center Watch, Sept 2026)

830+

Restrictive local measures passed in 2026 alone

Bipartisan across 42+ states (Heatmap, Nov 2026)

Deeper Dives

As the grid interconnection bottleneck has become increasingly undeniable, the AI data center industry's immediate default response has been to procure on-site gas turbine engines to deliver 24/7 baseload power independent of grid timelines.

On paper, this appears to be effective in addressing the speed-to-power challenge. In practice, this recreates and compounds the same infrastructure constraints the industry is trying to escape. These constraints are only amplified further as the data center industry increasingly shifts to on-site gas generation as a new default. Halcyon Gas Power Plant Tracker indicates 77GW of new gas-fired generation planned for addition between 2026 and 2033, the largest wave of thermal buildout in over a decade.

The development pipeline (announced through construction) behind it is growing faster still, with US gas capacity in development for data centers nearly doubling from 97GW to ~189GW, within 378GW of total US gas development, itself up 50% in six months (Global Energy Monitor). GEM’s own caveat, mirroring the execution-credibility emphasis in our Section 1A, is that ++most of this pipeline is early-stage and turbine-constrained — meaning the announced gas development pipeline carries the same delivery uncertainty as the announced data center pipeline it serves++.

Six specific risks undermine the on-site gas baseload thesis. These risks are correlated and compound upon one another. Data center operators committing to heavy on-site gas baseload are simultaneously exposed to equipment lead times, capital cost inflation, fuel price swings, pipeline bottlenecks, community opposition, and gas system cost and reliability externalities. This creates concentrated risk on a single energy pathway at a moment when the industry's strategic imperative is to diversify and de-risk.

1

Surging Equipment Lead Times

  • Full-year 2026 gas turbine orders are pricing roughly 50% higher per kilowatt than 2025 orders (GE Vernova, Oct 2026), and Wood Mackenzie projects equipment prices will reach nearly triple 2019 levels by 2027, with equipment contributing 20–30% of total power plant cost (Wood Mackenzie, Apr 2026). Cumulative Q3 order books at the three major OEMs — GE Vernova at 122GW, Siemens Energy at 95GW (69GW firm + 26GW reserved), and Mitsubishi Power at 35GW — now total >250GW against global manufacturing capacity of just ~65GW per year: nearly four years of production already sold. New orders are now booking delivery in 2031 and beyond. Six-year delivery cycles are now standard with OEMs requiring 10% up-front slot-reservation deposits years converting procurement into a multi-year capital commitment before any revenue begins. Without pre-existing orders in place, the speed advantage over grid interconnection has effectively disappeared.
  • Notably, of the 253GW of gas-fired capacity in US development as of late 2025 — a pipeline that has since grown to 378GW in mid-2026 (GEM) — 146GW (more than half) had not announced a turbine manufacturer. The three major suppliers (GE Vernova, Siemens Energy, Mitsubishi Power) accounted for just 91GW of identified orders (Global Energy Monitor, H2 2025), underscoring the gap between announced gas projects and actual equipment procurement.
  • Equipment Damage: If gas turbine equipment installed breaks, equipment supply constraints mean that there are no spare turbines immediately available for replacement. This is especially relevant given volatility in AI compute loads that can swing from full power towards zero in milliseconds, cycling ten to twenty times per second — a demand profile no thermal generator can follow. Fast-ramping gas turbines need one to two minutes; the mismatch spans three orders of magnitude. At these frequencies, periodic load profiles create resonant torsional stress on rotating equipment, accelerating fatigue and mechanical failure. Training workloads specifically generate measured oscillations of up to 70% of rated load at millisecond timescales — making battery buffering a structural requirement, not an optional enhancement.
  • In contrast, the US now has solar module manufacturing capacity to produce nearly 65GW as of year-end 2025 (SEIA), up over 50% from the prior year, with domestic factories now able to fully meet US demand for the first time in over a decade. Lead times for solar panels are measured in weeks, not years.
2

Inflating Capital Costs

  • As of June 2026, new CCGT CapEx build costs have climbed to ~$3,000/kW and as high as ~$4,000/kW — roughly triple 2022 levels — driven by supply-chain constraints, materials inflation, and surging demand. But capex understates the problem, because what determines whether gas pencils is the price of delivered power, with the market transacting at the equivalent of ~$130–160/MWh (under tolling agreements where the offtaker is subject to fuel price volatility exposure risk). The capital intensity of gas baseload is approaching levels that fundamentally challenge project economics, especially for speculative data center developments without committed compute offtake.
3

Gas Pipeline Interconnection Constraints

  • On-site gas generation requires firm natural gas supply, which means pipeline interconnection. Similarly to the grid, existing gas pipeline capacity is also becoming increasingly constrained. New pipeline construction faces its own permitting timelines (often 2–5 years), environmental review requirements, and community opposition. The data is visible in the records of midstream companies carrying record pipeline project backlogs. Record backlogs represent a queue, newly sanctioned capacity carries in-service dates around 2030, and securing firm capacity on new build increasingly means anchoring the pipeline itself with 20-year take-or-pay commitments.
  • Data center operators seeking to avoid electric grid interconnection constraints by shifting to an on-site gas baseload strategy are now encountering and exacerbating gas grid interconnection constraints.
4

Fuel Price Volatility

  • Natural gas prices are inherently volatile and subject to geopolitical disruption. In 2026, the Iran conflict and Strait of Hormuz disruption sent Brent crude from roughly $60 at the start of the year to peaks near $130, while LNG spot prices surged more than 140%. Even with de-escalation, Gulf flows have only recovered to only 80–85% of pre-war levels as prices remain well above where the year began.
  • BTM gas contracts are increasingly structured as tolling arrangements, where the data center off-taker is required to assume commodity fuel price risk, while the developer is responsible for delivering the turnkey asset with an availability guarantee.
  • In contrast, on-site solar and storage provides pricing locked through fixed long-term PPA structures, without any commodity price exposure (fuel cost is zero).
  • Compounding the fuel price risk, US natural gas production growth in the three largest shale regions, responsible for three-quarters of all US shale gas production, has slowed considerably. With infrastructure assets designed to operate for 20+ years, fuel input cost risk is a structural vulnerability.
5

Community Opposition & Permitting Risk

  • On-site gas generation produces local air emissions — NOx, CO₂, particulate matter — that trigger air quality permitting requirements and community opposition. Multiple data center projects have already faced lawsuits and backlash over air pollution from on-site gas turbines, with communities organizing against developments. Examples include:
  • xAI's Colossus build-out drew federal lawsuits from environmental groups and local community opposition over air quality, as its newer Southaven, Mississippi site grew to roughly 47–57 on-site gas turbines through mid-2026, — leading the Department of Justice to intervene on behalf of xAI in June 2026, citing national security (The Atlantic).
  • The focus extends beyond air quality (and beyond the US). In Rocky View County, Canada, a 900MW data center campus had already secured its power through a 50/50 gas joint venture, but was rejected by the county anyways citing water use, electricity impacts, environmental concerns, and agricultural impacts, with objections specifically directed at its gas plant. Projects can secure power and financing and still fail on social license (Currence).
  • As environmental scrutiny of data centers intensifies, gas-heavy power strategies create a permitting and public relations liability that compounds over time. Secured power is no longer secured permission.
6

Gas System Cost & Reliability Externality Risk

  • Behind-the-meter gas generation does not eliminate grid dependence, it simply transfers it from the electrical grid to the natural gas grid. Natural gas generates approximately 40% of US electricity, meaning electricity prices are inherently tied closely to natural gas prices. As data center gas consumption scales, behind-the-meter operations will drive up natural gas and electricity prices for all ratepayers, resulting in the same cost-shift backlash dynamic on the gas grid that regulators are already fighting on the electric grid. Many industries (such as petrochemical, manufacturing, industrial heating) that are structurally dependent on natural gas and cannot turn to electricity or on-site renewables as data centers can will face direct cost and supply competition from data centers. On-site gas baseload overreliance simply transfers risk instead of mitigating it.
  • Further, with turbine order books full through 2028+, every unit that goes behind the fence is one fewer available for the grid, meaning data centers are not only competing for fuel supply but actively absorbing the dispatchable generation capacity that utilities need to maintain grid reliability.
  • Natural gas supply is also weather-dependent: cold winters, as Texas experienced in 2021, can freeze wellheads and crimp supplies dramatically. During Winter Storm Fern in February 2026, gas accounted for 84% of thermal generation outages in ERCOT (850+GWh of total outages), while PJM lost 22GW (16% of committed capacity). This dynamic may force suppliers to choose between keeping data centers running and letting people heat their homes, a position no supplier or regulator wants to defend.
  • The correlated-failure profile is itself a grid risk and no longer hypothetical. On July 22 2026, PJM saw its largest data center load-loss event on record: nearly 4,000MW of data center load trip offline unexpectedly in Northern Virginia, as a mechanical failure at a 230kV line sent data centers flipping to backup power en masse, causing spikes in voltage and system frequency that cascaded. While the system recovered in nine minutes, this was the third load drop occurrence in the service territory, as PJM’s committee noted facilities are disconnecting too early, on faults the grid clears normally. FERC has now ordered NERC to file the first enforceable large-load reliability standards by December 31, 2026, with minimum ride-through requirements expected in 2027, and ERCOT's ride-through requirements took effect August 1, 2026. On-site gas does not eliminate this risk; it just transfers it. Every behind-the-meter site that loses primary power drops onto the grid simultaneously, and correlated weather, gas supply, or pipeline disruption events cluster that risk further.

The operational and financial risks of gas overreliance outlined above do not exist in isolation — they feed directly into a broader political and community backlash that is reshaping the terms on which data centers can develop, regardless of power source.

Energy affordability is where the backlash flashpoint started, and it has become a forefront political priority across the United States as utility bills and electricity costs continue to rise, roughly 30-40% on average since 2020 (US EIA and Bureau of Labor Statistics). As the average US residential electricity bill hit a record high in July 2026 (against hottest month in US recorded history), regulators have logged 362 energy-affordability actions across 46 states, DC, and Puerto Rico this year. An estimated 13–21% of US households are now considered “energy-burdened” with 77% of American reporting anxiety about their energy bills (PowerLines/Ipsos, Mar 2026).

Data centers are increasingly at the center of energy affordability discussions, as the scale of AI-driven load growth has drawn direct political attention. In Nov 2025, New Jersey and Virginia elected governors who campaigned explicitly on reining in energy costs, with Virginia’s governor directly promising to make data center operators pay for their share of grid upgrades and take immediate action on energy affordability upon taking office. This issue is bipartisan, with the backlash widespread across state, RTO/ISO, and federal levels.

Affordability proved to be a defining factor in the Nov 2026 US election cycle. Governors have already acted beyond campaign rhetoric: NJ’s Governor Sherrill made utility business model reform her first executive order; Indiana has passed legislation tying utility compensation to performance instead of spend. Developers who cannot demonstrate utility rate neutrality at minimum will face political headwinds in battleground states. 

PJM at the forefront of data center energy affordability backlash

Affordability pressure is visible at every level of the system, but most acutely in PJM. The PJM regional electricity market, historically the heart of US data center development anchored by Northern Virginia, has been at the forefront of rising public and political energy affordability backlash, especially as it pertains to data centers, as it struggles with a worsening supply-demand imbalance. 

PJM's independent market monitor has attributed $29.4B in data center-related cost increases across the last four capacity auctions (46% of total auction costs) with wholesale power costs up 54% year-over-year, driven primarily by a 262% surge in capacity costs. In December 2025, PJM's 2027/28 capacity auction cleared at a record $121/kW-year, driving upward pressure on customer electricity bills by as much as 20%, while still failing to procure its 20% reserve margin for the period. The 2028/29 auction then cleared at its $325/MW-day price cap (3rd consecutive auction to hit the ceiling) and would have cleared near $555/MW-day regionwide without the collar. Despite record prices, only 525MW of new supply cleared as the reserve margin came in 6.8GW below target, the worst shortfall on record, with the market monitor attributing $6.3B (38%) of the auction's cost to data center load, calling the shift a "paradigm shift" and proposing a separate procurement track for data center demand. For scale reference, a 10MW industrial customer's monthly capacity charge is now on track to rise from roughly $6,000 in 2024 to roughly $70,000 by 2028.

Data center load growth has been cited as the primary driver of these conditions. PJM expects a 32GW increase to peak demand from 2024 to 2030, with 30GW attributed to data centers. At the individual utility level, projected data center loads are set to dominate system planning. Within PJM, Dominion Energy's data center load is expected to grow from roughly ~3GW today to ~18GW by 2035, AEP from ~2GW to ~10GW, and ComEd from near zero to 6GW (Halcyon). This imbalance is further exacerbated by PJM's inability to unclog its new generation queue (100GW+ backlog of new solar, wind, batteries) — even as more than 50GW of already-cleared projects now await build-out (Aurora Energy Research and RMI estimate that bringing 10GW of these cleared renewables projects online by 2028 would deliver $10.9B in ratepayer savings).

In response, the White House and all 13 PJM state governors issued a bipartisan “Statement of Principles” calling for an emergency PJM backstop capacity auction to address electricity affordability and reliability. PJM’s proposals have moved in the same direction, including a provision under which new large loads that do not supply their own generation will face mandatory curtailment during peak demand periods. The signal to the data center industry is clear, self-supply power or expect to be curtailed.

That signal is now being tested at FERC but the test is over who sets the terms, not whether terms are coming. PJM’s August filing would have excluded new large loads from the base capacity auction unless they secured matching new capacity (storage, demand response, and uprates included). FERC rejected most of it on jurisdictional grounds as the states’ domain, while accepting the 50MW large-load definition and a mandatory large-load registry. PJM’s backstop procurement was refiled within weeks for a March 2027 launch against a 6.8GW shortfall. While the federal docket cycles, the same terms are being written in state large load tariff proceedings and in bilateral contracts, where minimum contract terms have stretched from under 3 years to roughly 12, collateral requirements are tightening, and self-supply and flexibility commitments are becoming standard conditions of service. The binding  version of “self-supply or be curtailed” is arriving in interconnection agreements now, and with PJM capacity prices likely to stay higher for longer (Jefferies, Oct 2026), the value of self-supply and flexibility is raised regardless of how the docket resolves.

Within PJM, the PA House of Representatives passed HB 1834 (March 2026) as "a data center ratepayer protection bill" directing the state PUC to develop regulations barring electric companies from passing data center infrastructure and energy costs onto ratepayers — including regional transmission, network upgrades, grid reliability, and PJM emergency capacity procurement costs. Separately, PA Gov. Shapiro's GRID (Governor's Responsible Infrastructure Development) standards advanced by executive order, requiring data center developers to fund their own power, hire locally, maintain transparency with communities, and adhere to high water-conservation standards, with support and expedited interconnection withheld from projects that do not commit. Pennsylvania is effectively the first state to implement the PJM framework's logic in state policy.

Energy affordability and reliability risk has escalated well beyond PJM

If AI data centers continue to be perceived as contributors to rising electricity costs and worsening grid reliability, especially during extreme weather events when grid stress is most visible to the public, the industry risks becoming a permanent political scapegoat, with consequences to match. 

In its latest Long-Term Reliability Assessment, North American Electric Reliability Corporation (NERC) found that 13 of 23 US/Canada grid regions face higher blackout risks over the next decade, with 5 at high risk by 2029, as demand from new data centers and other large loads is expected to outpace energy supply growth. The reliability squeeze is already extending the assets that communities want retired most. For instance, FERC has extended reliability-must-run agreements for Maryland's Brandon Shores coal plant (1,282MW) and H.A. Wagner oil units (774MW) through 2031 (six years past planned retirement) — leaving ratepayers paying $312 and $137 per MW-day respectively to keep the units alive. When load outpaces the grid, the default outcome is not new clean capacity but old capacity, extended by mandate, at ratepayer expense. Additionality prevents this.

The risk extends beyond actual contribution to rate increases. Data centers are increasingly being blamed as a scapegoat for utility rate increases at large, even when increases in recent years have been driven primarily by aging grid infrastructure investments, to which AI load has contributed only marginally thus far. Without a proactive energy and policy engagement plan, data centers will inherit this blame and the policy consequences that follow, regardless of whether the attribution is fair.

What the rate data actually shows

The claim that data center loads inevitably result in higher electricity rates misreads utility economics. Rates are determined by total utility capital spend over total demand. New large loads that spread existing fixed costs across more kilowatt-hours (increasing the denominator relative to the numerator) drive utility rate suppression. In capacity-constrained markets like PJM, the co-development model addresses both sides: new generation expands self-supply and reliability, while flexible load expands grid demand — enabling higher grid load consumption without requiring proportionate increases to utility infrastructure upgrade spend. Recent EPRI analysis found data center growth was associated with ~6% lower retail rates on average from 2019–2024, consistent with the denominator effect described here. The exception is constrained regions where urgent buildouts outrun cost allocation (the PJM pattern, not the national one).

Bipartisan backlash at the federal level

Nonetheless, the backlash is bipartisan and consolidating federally. The "Ratepayer Protection Pledge" announced in March 2026 has expanded from its original seven signatories to roughly 200 organizations representing about 80% of US delivered power. That said, PJM's market monitor expressed that the pledge is unfulfillable under current market rules, where new-load costs socialize by design — and enforcing it requires additionality and load flexibility. Data center energy costs are now bipartisan electoral strategy — and across the tightest 2026 election races, candidates from both sides converged on stricter data center guardrails.

Affordability is where the backlash began, but not where it ends. What communities are now contesting runs the full ledger of what actually impacts them, organizing around three core focuses. The first is still their bills, whether rates increase. The second is resource strain, whether power, water, land, and air resources are depleted. The third is what daily life looks and sounds like near the fence line, including visible blight, noise, light, and ecosystem impact. 

Worsening community opposition is a structural industry risk

Public opinion has hardened fast and continues to harden further. A March 2026 Gallup poll found 71% of Americans oppose a data center being built near them — higher than opposition to a nuclear plant (53%) — with national support falling from 65% to 36% in roughly a year. In July 2026, POLITICO found that Americans’ attitudes toward data centers have grown increasingly sour and politically polarized with opposition to a data center within three miles rising from 28% to 41% between January and July — and nearly half of respondents responding that they'd be more likely to vote for a candidate proposing a statewide moratorium. While different question framings produce different opposition levels, the directional trend is unambiguous.

A meaningful share of community backlash traces to capital structure. Substantial data center land parcels have been assembled by short-hold investment vehicles structured to exit on three-to-five-year time horizons, well before long-term community relationships can mature. The result is a sector-wide externality: community opposition now attaches to the entire “data center” industry rather than the conduct of any specific operator. Long-hold sponsors inherit the political friction generated by short-hold capital, and community engagement has become a shared industry cost that shows up directly in project cap rates, permitting timelines, and interconnection queues. The commercial implication is straightforward: operators who build durable community relationships convert a sector-wide externality into a site-specific advantage.

The same capital dynamic that strains community relations also distorts grid planning. The speculative pipeline cited earlier — 145GW of announced projects held by sponsors with no operational track record — has clogged interconnection queues with filings that may never energize (Aurora Energy Research). Grid operators now face a signal-to-noise problem: forecasting which projects will actually deliver has become nearly impossible, and transmission and generation planning is delayed for real capacity alongside the vaporware. Credible operators pay the price — not for their own execution, but because they share a queue and a reference class with projects that will never break ground. Effective policy — sponsor qualification standards, interconnection queue screening, and project-readiness filters — must separate the two, so that committed capacity is not taxed by friction generated elsewhere. Those filters have begun to arrive in 2026: collateral requirements, per-MW deposits, and large-load registries are doing the necessary separation.

Opposition momentum continues

Community opposition is now the single largest driver of project value destruction for the data center industry. ~$198B in projects blocked or delayed by community opposition in H1 2026 alone, ~27% above all of 2025 (Data Center Watch, Sept 2026), in a remarkably bipartisan pattern. Opposition drives 52% of cancellations but only 3% of delays (Currence, Q3 2026), with cancelled capacity rising 66% quarter-over-quarter to 4.5GW in Q3 2026. While delays can be resolved with time, community trust has to be earned up front. 

Backlash enacted into law with a growing restriction map

The restriction map has moved from proposals to enacted law. Heatmap's tracking counts 830+ restrictive measures passed in 2026 alone (Heatmap, Nov 2026). Counting only severe constraints and outright permit bans, more than 530 restrictive local laws have been enacted across 42 states as of Aug 2026 (Heatmap) Currence counts 500+ proposed state and local measures, of which 250+ are now effective — a 75% jump from May to June, the sharpest single-month increase on record.  The pass-rate asymmetry shows at the local vs. state level with roughly 80% of local measures enacted while only 5% of state-level bills pass. 

New York became the exception that proves the pattern: Executive Order #62 (July 2026) imposed a one-year statewide moratorium on environmental permits for data centers of 50MW or more — the first statewide moratorium in force — but its exit architecture is the more instructive half: community-benefits negotiation guidance within 60 days, and a directive to modernize large-load interconnection "while ensuring those consumers pay their fair share or supply their own power." Meanwhile, Illinois Gov. Pritzker announced a two-year pause on data center tax incentives, with similar rollbacks proposed in Maryland, Michigan, Oklahoma, and Virginia; in Florida, Gov. DeSantis announced an AI "bill of rights" giving communities the right to limit data center construction; in Arizona, Gov. Hobbs has supported pulling the industry's tax incentives.

Beyond energy, broader socio-political headwinds are building. An emerging de-growth sentiment against AI — driven by societal anxiety over job displacement, privacy, and concentrated corporate power — is spurring an increasingly populist backdrop. Political backlash is the biggest risk to data center development. While community acceptance is among the most significant risks, it is also among highest-return investments in project economics, given the time-to-power value (Dive 1C Tokenomics).

An emerging thread: fire safety and insurability

Fire safety is also emerging as a distinct thread in the same social license family. Lithium-ion battery storage deployed inside data center envelopes for UPS and pulse mitigation is governed by NFPA 855 (installation) and UL 9540 (safety listing), and fire-service tactical doctrine for lithium-ion incidents is increasingly shifting toward defensive postures: perimeter containment (vs. interior fire attack), driven by thermal-runaway gas hazards and the low life-safety exposure (mostly unmanned facilities). The implications transmit commercially through insurability, permitting, and community trust. The Playbook's architecture already anticipates this thread: siting large-scale storage at the energy-campus perimeter and outside the fence rather than inside the compute envelope, and specifying non-lithium technologies (supercapacitors) for inside-the-fence pulse mitigation — fire-safety decisions as much as power decisions. We will monitor and develop this interwoven thread further in coming updates as it evolves.

The November 2026 midterm elections were the first national cycle in which data centers functioned as a named voting issue, building on 2025’s VA and NJ races where the issue first surfaced state-level. Where data centers appeared directly on ballots, restriction won almost without exception, with 40 of 41 local ballot measures passing across 10 states with passage showing no correlation to county partisan lean. Post-election, the local restriction cadence continues unabated, but statewide measures open a second front

That second front was most visible in Governor races. In Texas, Governor Abbott lost re-election to Governor-elect Hinojosa, who campaigned on a statewide moratorium. New Mexico’s incoming administration has pledged a day-one pause pending an eight-condition state framework; Georgia’s governor-elect floated a statewide pause during the campaign; and Ohio’s new administration carries a conditional-moratorium posture, with a statewide ban on large facilities aimed at the November 2027 ballot. The moratoria wave that began locally is spreading, and we expect more to come at the state level as the 2027 legislative session begins. The 2027 legislative sessions, rather than elections themselves, are where terms get written.

But, the moratoria wave is not the end-state. The strategic question rather is what policy will be written next. Almost every moratorium on the books is a pause to buy time, not a permanent decision: 78% create 6–12-month delays, only 9% have been extended so far, and the observed endgames run from permanent zoning ordinances to outright bans to serial extensions (Currence), putting a wave of expiration decisions on the calendar for mid-2027. Effectively, the majority of moratoriums are an invitation to the negotiation table with a deadline attached.

Beyond moratoriums, several states are already writing responsible-development standards that set a bar for the industry to clear, representing the exit ramp following moratorium expirations if the industry is able to meet it. What gets written across states already has visible precedent or signaling:

  • Frameworks: 1) Massachusetts has already implemented a framework, the first to mandate the bargain, via Executive Order 658 requiring data centers above 25MW to bring their own additionality, supply clean generation covering annual consumption, or pay into a Ratepayer Protection Fund; 2) Virginia with both Gov Spanberger’s accountability framework covers transparency, noise, water, and clean energy as well as its state energy plan that treats data center demand as an instrument of the state’s carbon-free buildout; 3) Pennsylvania with Gov Shapiro’s GRID standards conditioning expedited interconnection on self-funded power, local hiring, transparency, and water commitments.
  • Pause-first, building conditions: 1) New York implemented the first statewide pause via Executive Order #62 but the exit ramp design work is already underway, including a large-load tariff targeted by year-end
  • Moratoria with conditions attached: 1) New Mexico Gov-elect Haaland’s pledged pause is explicitly paired with an eight-point data center accountability framework; 2) Ohio Gov-elect Acton supports a conditional moratorium by design where developments are paused unless they meet a specific set of requirements; 3) Maryland Gov Moore expressed support for a moratorium only paired with an exception for projects meeting high standards; 4) Michigan Gov Whitmer had rejected an outright moratorium in favor of codified consumer protections plus a company pledge while her successor Gov-elect Benson has called for a statewide moratorium on new construction until strict regulatory guardrails are established, nearly identical bargain terms but commitments move from pledge to requirement

The groundwork underneath the politics also influence where each state ultimately lands. Texas is a live test with the incoming Governor expected to implement a moratorium, but groundwork for standards is already laid by TX SB 6’s power-down provisions, PUCT’s adopted large-load rules, and ERCOT’s Batch Zero process. Whether that groundwork shortens the length of a pause is an open question to watch in state data center policy heading into 2027.

The direction is clear, pauses serve as a start to standard-setting rather than an alternative to it, and these terms will be written in upcoming 2027 legislative sessions. The same cycle that enacted hundreds of restrictions has also surfaced the compliance path out of them, converging on additionality, flexibility, and enforceable community benefit. The industry recognizes it must proactively address affordability, reliability, and community impact while that window remains open. The Playbook provides the framework for how in subsequent Sections.

Against the backlash described, one layer that has already been actively written are utility large-load tariff structures and interconnection requirements that directly set project economics and timelines. These standards are being developed across every level of the grid: individual utilities, state public utility commissions, RTO/ISOs, and federal regulators. As of August 2026, 104 large-load tariffs are in place or pending across 37 states — up from 65+ across 30+ states just two months earlier — with roughly one in five now carrying an explicit curtailment or interruptibility clause. While development standards will be written in 2027 legislative sessions, interconnection agreement terms are actively being established.

From an energy perspective, regulators and grid operators are most concerned with 1) avoiding cost impacts on general ratepayers; 2) maintaining grid reliability; 3) preventing environmental impacts. AI infrastructure companies that proactively position to meet these requirements — by delivering additionality and embedding load flexibility — will be at a structural advantage. Those that wait will negotiate from a position of weakness against increasingly demanding terms.

Three primary categories of “sticks” are being implemented to address these concerns through large-load tariffs:

Take-or-Pay Minimums

Utilities are requiring data center customers to commit to paying for a minimum percentage (typically 75–85%) of their requested load, regardless of actual usage. Contract terms have extended to 10–20 years to align with transmission and capacity planning, materially de-risking utility capital deployment, with collateralized exit fees and early termination penalties. Higher minimum demand charges enhance cost recoverability, while elevated collateral requirements (as high as $1–1.5M per MW) serve as effective filters to deter speculative demand and improve pipeline quality. The intent is to prevent cost shifting: if a data center requests grid capacity, it pays for that capacity whether or not it uses it, preventing stranded costs from falling on other ratepayers.

The convergence is measurable and fast: average minimum contract terms in large-load tariffs have moved from roughly 2.5 years before 2024 to about 12 years in 2026 (Halcyon), with collateral tightening in parallel. The observed family spans the country — AEP Ohio’s approved framework (85% of demand, 12 years, exit fees — litigated and sustained), NorthWestern’s 15-year/75%-minimum terms, DTE–Google at 20 years with 80% minimums and termination protections, dedicated-supply requirements for large loads in Kentucky, and the OG&E–Google agreement (100% of connection costs and all contracted costs regardless of usage) pending commission approval. Terms are hardening at the leading edge: Georgia Power’s signed large-load contracts now set minimum bills at ≥100% of incremental cost to serve over 15-year terms with no volume at risk, and Microsoft has proposed its own ‘Ratepayer Protection Tariff’ in Nevada (Bring Your Own Power, a 2% cap on residential increases). Regulators are also converting one-off contracts into standing rules: after approving a single 220MW large-load contract, Wisconsin’s commission ordered a standardized tariff for all new loads above 100MW — exit fees, reporting requirements, and financial safeguards included.

Jefferies' tracker of binding large-load commitments reached 118.8GW in Q1 2026 (+12% QoQ), led by AEP (17.9GW) and Entergy (15.8GW).

Additionality Requirements

Regulators and grid operators are increasingly requiring that new large loads demonstrate incremental new generation supply, not simply tap into existing grid capacity that already serves other customers. The FERC PJM Co-Location Case is an early landmark: FERC's November 2024 rejection of Amazon's proposed expanded co-location at Talen Energy's Susquehanna nuclear plant cited cost-shifting to ratepayers and reliability concerns — in substance, the objection to contracting existing generation away from serving the grid without new supply. The 2025 restructuring confirmed the direction: the arrangement converted to a front-of-meter PPA serving the grid, with the co-located load transitioning as transmission work completes. Reallocation without additionality did not survive contact with the regulator 

The retreat from non-additional structures now extends beyond dockets: the proposed co-location at Maryland's Calvert Cliffs nuclear plant was abandoned outright in 2026 after local elections turned against it, without ever reaching a federal ruling. And the signal has hardened into structure, PJM’s filed framework sought to make matching new capacity the condition of market entry (FERC rejected most of it on jurisdictional grounds, but the direction persists in refilings and state proceedings), New York's moratorium exit requires supplying your own power or paying your fair share, and the expanded Ratepayer Protection Pledge commits signatories not to contract existing plants away from serving the grid. Additionality has moved from a regulatory preference to a mandatory prerequisite.

Curtailment Mandates

Increasingly required as a condition for interconnection, particularly to protect peak grid reliability. Large loads must agree to curtail grid consumption during peak demand periods on notice. Texas SB 6 has now been implemented as a statutory landmark with mandatory firm-load curtailment on 24-hour notice within the statute itself. The curtailment-service parameters are still being finalized in open rulemaking (PUCT Project 58482, proposed July 2026). More broadly, SB6 codifies flexibility as a gateway condition for any large load energy campus in Texas — operators must demonstrate the ability to pull from both grid and on-site generation, making flexibility a regulatory prerequisite for the co-development pathway, not just an operational optimization.

The mandate now comes with precedent: Texas regulators have approved the first fully-curtailable co-located AI data center in ERCOT (curtailable within 30 minutes), while the largest disclosed demand-response commitment on record arrived in Georgia, where OpenAI committed up to 1GW of curtailable demand for its Project Camellia, representing roughly a third of its proposed campus's draw, as part of an approved 25-year Georgia Power contract. PJM’s proposals have moved in the same direction, although the mechanism is being redesigned, curtailability-for-speed persists across filings. Curtailability has become the price for speed.

The pace of tariff adoption is accelerating. As of early 2026, more than 65 large-load tariffs have been proposed or approved in over 30 states (SEPA/NC Clean Energy Technology Center). The trend began in 2024, led by early movers Ohio and Indiana, expanded to Kansas, Michigan, and Virginia, and is now being debated in Illinois and Wisconsin. Some utility regulators have included requirements for data centers to “bring their own” generation or grid capacity.

Three tariff models have emerged: 1) published tariffs (transparent and replicable, but limited flexibility); 2) framework tariffs (standardized rules with negotiated economics, increasingly attractive for clean-energy pairing); 3) and bespoke contracts (customizable but opaque and non-scalable). States and utilities are each calibrating their own approach, but take-or-pay, additionality, and curtailment remain the consistent building blocks.

The signal is clear as the ratepayer protection trend has accelerated across states. The instrument set is diversifying beyond ratepayer protection. Cost-causation tariffs — the take-or-pay, collateral, and minimum-bill structures detailed above — protect existing ratepayers. Speed-for-flexibility frameworks trade expedited interconnection for additionality and curtailability (TX SB 6, PJM's proposed matching-supply condition, NJ Fair Share Act). Clean-supply requirements tie benefits to generation mix (Minnesota, Oregon, Michigan alignment with state clean energy goals). Across all three categories the direction is the same: the terms of access are being written now, and additionality plus flexibility are the currency.

This is a rapidly evolving landscape that the Playbook will continue to track closely. The contractual path these tariffs feed from first study to full energization is mapped in Section 4. And what these terms are worth when met is covered in Section 3.

Section 3 · The Opportunity

The Opportunity

What to Do Instead

“Clean flexible power co-development is the fastest, most de-risked, commercially superior path to powering AI at scale — and the path that earns the social license to keep building.”

The same dynamics that make energy the AI data center industry's biggest risk also make it the biggest opportunity for the companies that proactively lead. 

Power has become an executive and board-level constraint for AI and tech leaders, presenting a generational opportunity to not just achieve energy velocity but also to leverage AI-driven load growth as a catalyst to build a better power system for all.

The same opportunity exists with communities. Section 2 traced where community opposition is concentrated: bills, resources, and life at the fence line. Section 3 focuses on the opportunity to invert opposition by over-delivering rather than merely mitigating, being measurably additive for communities rather than just minimizing harm. That inversion is what converts a community from an opponent to a supportive counterparty.

As outlined in Section 2D, the industry can still shift the narrative, with ~78% of current moratoriums (Currence) standing as 6-12 month pauses and post-moratorium exit ramps still being decided. Responsible-development standards are now being written to set a bar the industry can clear as a productive policy path forward. Meeting that bar is how the industry shifts momentum from outright moratoriums towards this path. The industry must proactively address affordability, reliability, and community impact while the window to act is still open. 

Data centers can become the most sought-after grid assets by communities, as early movers are already demonstrating, by bringing their own clean generation, embedding load flexibility, strengthening grid reliability, and suppressing costs for all ratepayers among other enforceable community benefits (see precedent examples in Dive 3C). The policy signals around the opportunity are clear: FERC, PJM, SPP, ERCOT, DOE, etc. are all moving — through market design, interconnection rules, and emergency authorities — to prioritize large loads that provide both additionality and flexibility. AI data center developments that position for this now will earn accelerated interconnection, favorable utility tariff structures, and the regulatory goodwill to scale at speed.

Clean flexible power co-development is both the risk mitigation strategy and the fastest commercial pathway to scaling firm compute, turning energy from one of the industry's biggest constraints to its biggest opportunity. Through active co-development of energy alongside compute infrastructure, data center operators can deploy initial power in months, scale to hundreds of MWs of on-site clean generation and energy storage within 12 months from NTP, while pursuing complementary accelerated flexibility-embedded grid interconnection in parallel.

The result: 500MW+ of firm compute powered within 12–18 months per site through a repeatable framework, commercially structured to minimize upfront CapEx, at competitive costs. This is the fastest, most de-risked, commercially superior path to powering AI at scale — one that also earns the social license to build.

500MW+

Firm compute per site in 12–18 months

Through the Playbook framework

3–5 yrs

Faster grid interconnection with BYOP + flexibility

vs. standard queue timelines

76–126GW

New load absorbable on existing US grid

Tyler Norris, Duke

0.25%

Annual load flexibility to unlock 76GW of absorbable grid capacity

Tyler Norris, Duke (Feb 2025)

Deeper Dives

Clean flexible power co-development outperforms the data center industry's current defaults across every strategic priority driving energy velocity to deliver faster, more reliable, more scalable power with lower total cost of compute and mitigated policy and execution risk.

The debate between off-grid gas islands and full grid dependence presents a false binary. The Playbook represents a third path: on-site clean generation and storage behind the meter in complement to accelerated flexible grid interconnection. This framework captures the speed advantage of self-supply, the reliability backstop of the grid, the community acceptance of sustainable development, and the economic upside of bidirectional grid participation. This creates multiple independent pathways to power, staged in phases, with each phase de-risking the next.

Beyond new build, flexibility unlocks enormous stranded capacity already in the system on both sides of the meter. Most data centers today draw only a fraction of their provisioned power capacity (facility-level studies put average utilization near 30%), overbuilt for worst-case scenarios that rarely materialize. The grid itself mirrors this pattern. NVIDIA CEO Jensen Huang noted (Lex Fridman podcast, April 2026) that the US grid runs at roughly 60% of peak capacity 99% of the time. The real bottleneck to AI power, Huang argues, is not generation, not permitting, not interconnection queues — it is the five- and six-nines uptime standard that forces utilities to reserve capacity above their system maximum rather than using available excess. Data centers designed for graceful degradation — shifting workloads, accepting marginally longer latency during peak hours — paired with utilities offering tiered power products instead of one-size-fits-all contracts, could unlock capacity at a scale that dwarfs any single generation buildout. NVIDIA's DSX Flex platform targets 100GW of unlockable capacity on the existing grid; Tyler Norris's Duke University research corroborates the order of magnitude, estimating ~76–126GW depending on curtailment tolerance. The opportunity is not just in what gets built next, but also in what has already been built.

That said, although flexibility is essential, it is not sufficient alone. Flexibility buys time and unlocks faster interconnection, but it does not substitute for new generation or transmission build. This is why the Playbook's architecture is an integrated portfolio rather than single-lever — each element (on-site clean generation, storage, flexible grid interconnection, and grid-enhancing transmission investment) is complementary — each compensating for the limits of the others.

The Playbook's blended architecture unlocks more total MWs faster. Flexible grid interconnection complemented by on-site clean generation allows operators to scale power from both sources simultaneously rather than waiting on either alone. Because co-developed sites bring new clean capacity to the grid alongside new load, they improve broader system reliability and affordability for all ratepayers, turning data centers from grid burdens into grid assets.

Speed to power, decarbonization, and ratepayer affordability are not tradeoffs, but rather simultaneously achievable objectives when developers co-develop on-site clean energy campuses alongside load and flexibly optimize for existing grid capacity in parallel.

1

Speed (Time to Power)

The Playbook critically enables compute to be brought online, generating revenue in months rather than sitting idle waiting years for grid capacity. In contrast, the old playbook and current industry defaults deliver power in years. Grid-first interconnection timelines have stretched to a median of 5 years and continue to increase , from under 2 years pre-2008. The Playbook delivers power in staged phases: initial turnkey modular firm-power units within months of NTP, large-scale on-site solar + storage + complementary dispatchable power within 12 months, and flexibility-embedded accelerated grid interconnection pursued in parallel.

Notably, the supply side is not the bottleneck. Clean generation is ready to deploy — the constraint is a procurement and interconnection framework fast enough to match it to demand.  The deliverability case compounds it: a substantial share of accredited clean capacity already sits in interconnection queues, supply that can be matched to new load years faster than new-build alone. A single capacity RFP for 400 MW of accredited clean capacity in SPP drew 1.75GW of offers from over 600 developers within one month.

The national supply picture tells the same story. Solar leads all new US generation additions: 43.4GW of a record ~86GW of total planned 2026 additions, with storage adding another 24.3GW (EIA). The procurement contrast is clear: renewables and storage deploys in months from order, while a gas turbine ordered today delivers 2031+ with deposits locked years before first megawatt. Clean capacity is shovel-ready at a scale the current interconnection framework has yet to absorb.

Nor is the existing grid as constrained as it appears — if load flexibility is on the table. In what has become the cornerstone study on data center load flexibility, Tyler Norris — then at Duke's Nicholas Institute (now Head of Market Innovation at Google) — found that ~76GW of new large loads could be absorbed on the existing US grid if they curtailed just ~0.25% of annual load hours, roughly 22 hours per year. CPower's operating data across PJM, ERCOT, SPP, and MISO corroborates the same order of magnitude: data center VPP participation concentrates load reductions in the 16–40 tightest peak hours annually. The market's own accreditation math agrees with demand response rated at a 92% capacity credit in MISO's framework — higher than combined-cycle gas at 74–78% (and lower winter ratings at 60–65%) — meaning the grid formally scores flexible load as more reliable than the fuel the industry defaults to. 

EPRI’s facility-side testing closes the loop demonstrating 25% load reduction sustained for four hours through workload management alone, with a 500MW facility modeled as flexing 100–200MW for 100–200 hours per year, comfortably above the fewer-than-100 hours when transmission constraints typically bind. For sites with on-site power resources, those peak hours often require no compute trade-off at all — flexibility is simply delivered by shifting to self-consumption. At most, the trade is a rounding error on annual compute output in exchange for years of avoided transmission build and meaningfully faster time-to-power.

Site-level analysis corroborates the national picture. Aforementioned Duke University study found 100GW of generation headroom available on the US grid, and site-level analysis tells the same story across six real-world 500MW sites, delivering just 1 to 2 percent of annual power as flexibility was sufficient to unlock capacity at three of four transmission-constrained locations studied. The cost of flexibility is trivially small relative to the capacity it frees.

And even the existing transmission system itself is under-utilized. As Jigar Shah — former head of the DOE Loan Programs Office — frames it, the US is not short generation — it is short wires. A separate estimated ~100GW of existing grid capacity sits trapped behind transmission constraints (rather than generation limits) that grid-enhancing technologies (GETs), dynamic line ratings, and better operational data could unlock — without building a single new line or power plant. This is capacity that can come online in months through software and operational reform, not the decade-plus required for new transmission build. The near-term arithmetic sharpens the point: with capacity markets (PJM) clearing at record shortfalls, the only levers that operate before 2028 for the grid are load flexibility, and storage and renewables that have already cleared in interconnection queues (and on-site energy co-development for additionality).

Supply, grid capacity, and transmission itself — at every layer, the constraints are procedural, not physical. The Playbook builds from first principles for the system as it actually is — not the one conventional wisdom assumes.

2

Reliability (Continuity of Power)

The old playbook is grid-dependent with on-site diesel backup designed for limited emergency use, not 24/7 firm capacity. Aging grid infrastructure and increasing extreme weather events make grid-dependent power an increasingly unreliable single point of failure. The Playbook builds a blended on-site power portfolio — fuel cells, microgrid systems, solar, energy storage, complementary dispatchable generation — with flexible modular power conversion (SSTs) removing single points of power failure. Higher reliability of power means higher effective uptime, less stranded compute, and lower stranded revenue.

Energy storage is central to this architecture. AI compute loads create millisecond-scale transients that no thermal generator can follow — storage (including supercapacitors) is the only technology fast enough to buffer these oscillations, sitting between power sources and the load to orchestrate grid interconnection, on-site generation, and compute demand simultaneously. Without it, no co-developed campus can safely operate at AI-grade demand profiles. xAI's Colossus facility in Memphis deploys Tesla Megapacks specifically to buffer training-load volatility, validating battery storage as a core architectural component at gigawatt scale, not a theoretical fix.

Critically, this reliability logic cuts both ways: if grid-only is a single point of failure, so is off-grid-only. Even hyperscalers building dedicated on-site generation acknowledge that grid interconnection remains essential — not just for economics, but for the reliability redundancy that no single-source behind-the-meter configuration can match. Industry consensus is converging around expectations that 95% or more of data center capacity will require grid interconnection.

Capital markets have quietly reached the same verdict. Major financing firms have signaled they will not fund off-grid data center projects beyond two-year horizons — and notably, they are not drawing the line at fuel. Islanded power is treated as a short-term bridge rather than durable long-term infrastructure, regardless of whether the generation is gas, solar, or hybrid. The capital stack's verdict applies to off-grid architectures broadly — gas baseload is simply its most politically and economically exposed version today.

3

Economics (Total Cost of Compute)

The Playbook's power approach may cost more per megawatt-hour than a simple grid connection. But per-megawatt-hour cost is the wrong way to evaluate power for AI data centers (see tokenomics in Dive 1C). When a MW-year of capacity generates $13-15M at the compute layer (and multiples of that at the token layer), powers costs are overridden as the decision variable by factors including time-to-revenue, how often your GPUs are actually running, and how much capital you risk on infrastructure that may never arrive. The industry shorthand: LCOE (levelized cost of energy) favors the old playbook; TCOC (Total Cost of Compute) favors the new one.

The arithmetic makes the point. On paper, the Playbook's blended on-site power portfolio runs ~$80–100/MWh compared to ~$60–80/MWh for grid power and >$130/MWh for on-site gas baseload. The premium over grid power totals to roughly $175-350K per MW-year, a small fraction of annual revenue. In practice, the Playbook drives lower total cost through:

  • Faster time-to-revenue: GPU utilization starts months earlier, each month recovered is a month of revenue at whichever layer is held 
  • Higher effective uptime: diversified power portfolio reduces downtime risk, downtime is stranded revenue against the same layer economics
  • Avoided stranded costs: no multi-year capital commitments to grid interconnection that may slip or fail
  • Orchestrated efficiency: AI-driven controls maximize compute output per watt of constrained power input

The 2026 benchmarks reinforce the comparison from both directions: Lazard's latest LCOE analysis shows costs rising across every generation technology, with gas new-build at a fifteen-year high while storage costs actually reversed downward. Further, hyperscalers are validating clean-firm offtake as first-of-a-kind buyers in the $107–150/MWh band.

The commercial structures that underpin clean flexible power co-development further de-risk economics. Developers have validated delivery of solar+storage through ~15-year PPA structures at ~$100/MWh with 90% firm power guarantee, meaning zero upfront CapEx to the data center operator for a meaningful portion of the power portfolio. The energy developer partners finance, build, own, and operate the on-site power infrastructure. The data center operator pays for power delivered, rather than infrastructure built.

Financing follows the same logic as development, advanced in stages. No single capital check underwrites a multi-GW campus — projects advance in tranches backed by creditworthy offtakers and long-term PPAs. Fifteen-year terms under tariff or energy service agreements are the baseline; longer contracted cash flows unlock better financing terms, aligning developer and operator incentives around duration beyond price.

4

Policy & Public Perception

Under the old playbook, data centers are perceived as grid burdens, subject to regulatory “sticks” including take-or-pay tariffs, additionality mandates, curtailment requirements, and worst case, interconnection moratoria and de-growth legislation. The Playbook shifts the narrative. Data centers that bring additionality and embed load flexibility are positioned as grid assets, earning regulatory “carrots” including accelerated “fast-track” grid interconnection (as much as 3–5 years faster) and favorable utility tariff treatment by delivering rate suppression as a public benefit for all ratepayers. This converts policy risk into a competitive advantage.

The same logic applies to community acceptance. Data centers paired with on-site clean generation are consistently better received than standalone facilities, converting a permitting liability into a local economic asset. As Christian Belady, one of the pioneers of hyperscale data center development, put it: “All stakeholders should thrive — you can't have one-dimensional thriving” in regard to how data centers can earn community acceptance and “the license to scale”. The permitting delay math makes this commercially rational - the cost of skipping community engagement dwarfs the investment required to earn it.

5

Scalability

The old playbook scaled through grid interconnection capacity upgrades, meaning each increment of new capacity requires a new interconnection study, new approvals, and new grid infrastructure builds, each gated by multi-year timelines. The Playbook scales modularly through on-site power expansion complemented by grid capacity over time, with the flexibility to integrate additional power resources and emerging technologies (SMRs, geothermal, advanced storage) as they become commercially available and viable — all without needing to re-platform the power supply architecture.

The workload reality demands this modularity: even inference-only deployments now require hundreds of megawatts at minimum, while training clusters operate at gigawatt scale. A well-designed co-development site can scale generation and load together across this full range without redesigning the interconnection. Storage is scaling to meet this architectural requirement. BloombergNEF projects US energy storage capacity rising from ~40GW today to ~240GW by 2035 — a 6x expansion driven substantially by data center co-development and grid flexibility requirements.

6

Execution Risk

The old playbook is binary go/no-go: a single interconnection approval or equipment delivery determines whether the entire project proceeds. One delay stops everything.

The Playbook mitigates this through staged, phased execution with multiple parallel pathways. If grid interconnection slips, on-site generation continues delivering power. If one equipment vendor delays, modular procurement across multiple suppliers limits exposure. If a specific technology underperforms, the blended portfolio absorbs it without full-site impact. Revenue generation begins in Phase 2 (months from NTP) rather than depending on completion of the entire power stack. Each phase de-risks the next — and critically, no single delay creates a total project stoppage.

The supply chain diversity reinforces this. Solar panels, battery storage, fuel cells, and grid-enhancing technologies each draw from independent manufacturing bases and procurement channels. The concentrated equipment risk that defines gas turbine procurement (three OEMs, six-year backlogs, 10% upfront deposits) does not apply to a diversified portfolio.

Old Playbook vs. New Playbook Comparison

Access

Old Playbook

Single path to power — fully dependent on utility

New Playbook

Multiple paths: on-site self-supply + grid

Speed

Old Playbook

Multi-year (median 5+ yrs)

New Playbook

Staged phases: 3 months / 12 months / 12–18 months

Reliability

Old Playbook

Grid-dependent + backup power

New Playbook

Modular firm on-site power portfolio

Economics

Old Playbook

Lower LCOE (~$60–80/MWh), higher TCOC

New Playbook

Higher blended LCOE (~$80–100/MWh), lower TCOC

Policy

Old Playbook

Grid burden — public perception & policy risk

New Playbook

Grid asset — driving utility rate suppression

Scalability

Old Playbook

Scaled solely through grid upgrades

New Playbook

Modular on-site expansion + grid in parallel

Execution

Old Playbook

Binary go/no-go risk on single interconnection

New Playbook

Staged/phased portfolio risk management

The policy landscape is rapidly developing in a direction that directly validates the clean flexible power co-development opportunity. Across federal regulators, regional grid operators, and state legislatures, the signal is consistent: large loads that bring additionality and embed flexibility will receive preferential treatment.

The posture has shifted materially in the past twelve months. Data center operators are increasingly pursuing on-site additionality and load flexibility, driven by a straightforward recognition: this unlocks faster power access through both self-supply and grid interconnection, which maximizes compute output and revenue through faster energization.

Federal & RTO-Level Reform

In June 2026, FERC sharpened the federal signal, making flexibility the throughline. In a unanimous 5-0 vote, FERC issued show-cause orders requiring all six FERC-jurisdictional RTOs (PJM, SPP, NYISO, MISO, CAISO, ISO-NE — ERCOT sits outside FERC’s reach) to justify or revise their large-load tariffs — including a directive to create new transmission service products designed specifically for flexible loads. After extensions, the full six RTO response wave is being filed against a mid-November deadline, as the docket remains live with direction unchanged.

FERC's premise is explicit: a data center that can flex its consumption should connect in months rather than waiting years for network upgrades. Prior to its show-cause order, FERC approved PJM's new Expedited Interconnection Track, a fast lane for projects of 250MW or more to reach the grid within roughly three years. Separately, FERC's co-location proceeding pushed PJM toward firm and curtailable tiered large-load service — with bring-your-own-power data centers expected to interconnect 3–5 years faster than the standard queue. PJM governors from PA, MD, NJ, and VA have all jointly backed interconnection priority for self-supplying loads. 

SPP's FERC-approved HILL program enables interconnection timelines as short as 150 days through paired generation-load studies; its new Consolidated Planning Process merges transmission planning and generator interconnection into a single framework, cutting study timelines from 400+ days to roughly seven months; and its Conditional High-Impact Large Load Service (CHILLS) grants curtailable grid access on an accelerated study timeline. 

ERCOT's large-load queue, meanwhile, has surged to ~474GW (~90% data centers), more than five times the grid's all-time peak (ERCOT, July 2026). The queue is aspiration, not load with only ~9GW holding approval to energize, and observed large-load consumption remaining under 4GW. That gap forced a "Batch Zero" review to screen which requests are real — the review itself now paused pending the state's large-load audit, with eligibility verification for 326 applicants (~205GW) due to the PUCT by December. Within Batch Zero, Texas now offers a Provisional Controllable Load Resource (PCLR) fast-track that lets flexible data centers draw power ahead of the standard queue in exchange for being controllable.

As former U.S. Energy Secretary Jennifer Granholm framed FERC's action, the emerging philosophy is to "get more out of the infrastructure we already have before building more of what we don't" — flexibility itself is becoming an essential pillar of infrastructure development. 

The flexible-interconnection construct now has a name and multiple precedent examples. In ERCOT, the adopted Provisional Controllable Load Resource (PCLR) framework operationalizes what Arushi Sharma Frank (Luminary Strategies) articulates in "Connect & Manage for Loads": dispatchability (flexibility with obligation and evidence) as a transmission-access product, built on a firm floor with a conditional band above it, and moved from concept to adopted grid code in nine months (now implementation phased). 

This construct is not specific to Texas. Silicon Valley Power's connect-and-manage approach energizes large loads for "8760-minus-X" hours, with the curtailment bounds written into the interconnection agreement and the data center firming its own variable band. SPP's aforementioned HILL and CHILLS programs run the same trade. Internationally, Ireland has traveled the full arc — from a connection moratorium to a 2026 policy requiring dispatchable on-site or proximate generation with flexible, non-firm connection paths (Ireland CRU). And PJM's filed framework counts storage, demand response, and uprates as qualifying capacity. As Arushi Sharma Frank puts it: "the grid provides the firm floor, while the private sector underwrites and firms the rest."

What the data center actually does when called is an engineering choice, composed from a stack: switch power sources (grid draw to dispatchable on-site generation and storage), procure third-party flexible resources (aggregated storage and demand response), shift workloads across time and sites. Orchestration adds a distinct lever: maximizing token output within the available envelope, so flexibility and output compound rather than trade off.

Across these designs a common standard is emerging for what counts: flexibility that is quantified (megawatts and curtailable hours), capped (bounded exposure that can be underwritten), contracted (written into the interconnection agreement), and verifiable (metered and auditable). These four standards are what makes flexibility a bankable transmission-access product and codifying them is the live work of the moment. 

Federal reform is also live but unsettled: the Bipartisan American Affordability and Jobs Act (BAAJA) permitting package is working through the Senate currently through the remaining 2026 lame duck session, carrying data-center cost-causation rules alongside broader transmission and siting reform, though its path to passage remains unclear. Our expectation is unchanged either way: most of this gets written at the state and commission level through 2027, not through a single federal act.

State-Level Action

The federal signal is only half the story. Having pressed the RTOs to act rather than rule through mandate, FERC effectively hands the binding questions of how large-load curtailment is defined and enforced to the states. Across all three state venues where November elections concentrated, the same structured welcome is on the record. 

Governors are welcoming load, conditioned on standards.

  • Pennsylvania: Gov. Shapiro has already signed his GRID (Governor's Responsible Infrastructure Development) standards by EO, conditioning expedited interconnection on self-funded power, local hiring, transparency, and water commitments. Compliance earns the easier permitting track embodying standards as the speed lane
  • Michigan: Gov. Whitmer launched a responsible-growth pledge requiring data centers to ensure working families “do not pay a single penny” for their development, protect natural resources, and create local jobs — six companies (including Google and Oracle) signed within a day. Her successor Gov-elect Benson has proposed hardening the same bargain into a prerequisite via a pause until guardrails are established (Dive 2D), making the pledge terms the likely floor of what become mandated next.
  • Virginia: Gov. Spanberger has been consistent that data centers must “pay their fair share” along with an accountability framework covers transparency, noise, water, and clean energy. Her 2026 Energy Plan writes the structure into policy, requiring large data centers seeking interconnection to demonstrate they can reduce demand during grid emergencies, with data-center-driven grid costs assigned to those customers, and treating demand as an instrument of the state’s carbon-free buildout. Load is managed as a tool rather than a problem.
  • New Mexico: Gov-elect Haaland has outlined an eight-point plan that is effectively a conditions checklist for approval including: build your own renewables, closed-loop water, zero utility rate increases, transparent community benefits, local labor, public reporting, no NDAs, and environmental protections — planning to implement a pause in place until the protections are. The conditions are being stated as the the exit ramp in advance.

Statehouses are building the enablement layer through legislation. Laws are already being implemented to remove different legal barriers to energy co-development: the right to build on-site power, the right to buy it from third parties, and the terms on which flexible load is welcomed. Enacted legislature maps directly onto the Playbook’s architecture:

  • Oklahoma SB 480 removes the regulated monopoly bottleneck enabling a third-party to build, own, and sell behind-the-meter power to a large customer without being regulated as a public utility (Onsite Power)
  • Ohio HB 15 opens competitive self-supplypathways and improves tax treatment for new on-site generation (Onsite Power)
  • New Jersey A796 directs the BPU commission to set large-data-center rates to avoid burdening other ratepayers and lets demand-reduction obligations be met through behind-the-meter storage (Ratepayer Protection + BTM Flexibility)
  • Utah SB 132 gives large-load customers of 100MW or more get three statutory paths to power, not drawn as mutually exclusive: bespoke utility special contract, outside suppliers if utility negotiation exceeds 90 days, or closed private generation exempt from utility regulation (Accelerated Interconnection, Onsite Power)
  • Illinois POWER Act would provide a fast-track interconnection in exchange for new clean capacity and grid-upgrade payments (Accelerated Interconnection, Additionality) remains in negotiations for 2027 legislative session

Utility commissions are writing the structured welcome into rate design.

  • Wisconsin: The PSC cited “Existing Wisconsin customers should not pay a single cent to subsidize the service of data centers” (Apr 2026) in a decision that enabled growth on condition: requiring large customers fund 100% of their buildout (raised from the proposed 75%), while lowering the tariff threshold from 500MW to 100MW with 15-year minimum terms. Microsoft responded supporting “a rate class that ensures large users pay their own way” publically accepting the bargain. (Ratepayer Protection, Demand-Side Buy-In)
  • Georgia: With the November election flipping the PSC to an affordability bills-first majority, Commissioner Hubbard framed the mandate as “prioritizing lowering power bills” and “ramping up the deployment of clean energy technologies.” Rather than opposing load, regulators are emphasizing flexible, ratepayer-protective large-load design. (Affordability, Clean Additionality)
  • Virginia: The SCC’s large-load order (effective Jan 2027) sets a ≥25MW class with 14-year contracts and minimum demand charges the commission frames as insulating ratepayers from buildout costs, with a follow-on directing direct assignment of large-load transmission costs. This is the enforcement half of the Governor’s Energy Plan. (Ratepayer Protection)
  • Illinois: While the POWER Act is still in negotiation, ICC has approved ComEd’s revised large-load tariff (Mar 2026, ≥50MW, escalating deposits) as a “first step toward investigating how Illinois’ largest electric utility can serve large load customers while preventing undue cost shifts.” The commission has already built the floor while the statehouse debates the fast lane. (Ratepayer Protection)
  • Minnesota: The PUC approved Xcel’s large-load tariff in May 2026 and directed Xcel to file a separate clean-energy-and-capacity tariff by December: a voluntary path for very large users to support the state’s 100%-by-2040 goal, building the on-ramp for load to accelerate state energy transition priorities. (Ratepayer Protection, Clean Additionality)
  • Oregon: The OPUC adopted PGE’s new data-center rate class (May 2026, ≥20MW) with interconnection tied to compliance with the state’s decarbonization targets, clean additionality as a condition of access. (Clean Additionality)

2027 window makes this the moment. Legislative sessions open in January with data center bills pre-filed in restriction-minded and framework-minded states alike. Even as moratoria likely expands state-wide, as existing moratoria expire into mid-2027 expect states and counties alike to follow the template of the frameworks being written now.

The underlying question of how flexibility and additionality must be structured to count is being solved. The four standards — quantified, capped, contracted, verifiable — are emerging as the answer. Operators who bring commitments in that form into 2027 proceedings will not just clear the bar; but also help write it. This rapid, bipartisan diffusion across governors, statehouses, and commissions is the broadest recognition yet of co-developed additionality and structured flexibility as the path to speed and local economic value.

Industry Alignment

The demand side is also not waiting to be regulated into this posture and is actively organizing for it. The Data Center Coalition is broadly advocating expedited interconnection for curtailable and BYOP-compliant loads. EPRI’s Flex MOSAIC framework, endorsed by 65+ utilities, ISOs, regulators, and hyperscalers, standardizes how data centers describe flexibility — ramp-down speed, duration, and frequency — with an open letter signed by 30+ organizations including Google, Meta, Nvidia, Southern Company, Exelon, and Constellation calling for shared flexibility standards.

The industry is also focusing on market structure. The AI-Energy Markets Alliance (AEMA), launched in September by Emerald AI, Google, and NVIDIA with more than twenty partners across the value chain, is working to standardize how flexible load is valued and monetized in wholesale markets. This is an essential step to turn flexibility into a bankable interconnection product.

The codification is the live work of the moment across every venue in this dive: PCLR is in its implementation phase, the six RTO responses to FERC’s flexibility directive are being filed now, EPRI’s framework is standardizing the vocabulary, and AEMA is building the market structures to value it. How these definitions get codified over the coming year will set the terms for grid access for the decade to come. 

The direction across all levels — federal, regional, state, industry — is clear: the regulatory framework is being built around additionality and flexibility as the conditions for accelerated interconnection and favorable utility treatment. The Playbook is built around the framework regulators are already rewarding. The demand side has demonstrated it is willing to meet the higher bar. What still remains is for the bar to be set — and 2027 is where much of that writing happens.

A core foundation underpinning Section 3 is a “grand bargain” - stated simply “be a good grid citizen and a good community neighbor, get faster interconnection and permitting”. What makes the bargain hold is being made enforceable and bankable, so commitments made can be planned on by utilities, underwritten by capital providers, and trusted by communities.

On the grid-citizen half, a data center that brings qualified additionality and embeds quantified, capped, contracted, verifiable flexibility (Dive 3B) becomes a grid asset that utilities can dispatch against instead of a risk that utilities must hedge against. Qualified additionality means net-new capacity, accredited by the utility/RTO for what the grid can count on, and deliverable at interconnection on the local grid system. Quantified means flexibility bounds a counterparty can agree to (specific megawatts, hours per year, notice periods). Capped makes the commitment financeable, lenders cannot underwrite uncapped operational exposure. Contracted means the obligation is written into the interconnection agreement and large-load tariff. Verifiable means metered, certified delivery providing the M&V layer that lets a utility count on the commitment in its resource plan.

The community-neighbor half runs on the same architecture. The gold standard emerging in practice is a development agreement enforceable in both directions: commitments written in, independently verified, delivery gated to permits and occupancy, ratcheting as the campus grows. A community-benefit permitting framework is emerging: developers who commit to a defined checklist — local hiring, water replenishment, grid flexibility sharing, environmental mitigation — can target six-month permitting timelines versus eighteen months through traditional channels. The framework is being developed by a coalition of large campus developers and represents a 3x acceleration for projects that invest in social license upfront.

The strongest community commitments go beyond mitigation to replenishment. As an example for water resource, DigitalBridge's water replenishment model illustrates what this looks like in practice: funding agricultural irrigation upgrades — converting flood to drip systems — that generate two times the water credits consumed by the data center, making the facility a ‘water-positive’ net water contributor to the community. The same replenishment logic is also being applied to power itself: New Jersey's first-of-its-kind Fair Share Act grants interconnection queue priority to data centers that fund household electrification upgrades (heat pumps, rooftop solar, home batteries) crediting the capacity freed elsewhere on the system to the facility that paid for it. Replenish the resources you draw on and the community math flips from extraction to contribution.

Enforceability is the hinge of both halves, because only enforceable commitments can be planned on. That is what makes the bargain bankable in both directions. Utilities can build enforceable flexibility into its resource plan. Capital providers can underwrite enforceable terms. Communities can trust enforceable benefits. Enforceable flexibility converts a data center into a grid asset; and enforceable benefits convert a community from an opponent into a counterparty.

The grand bargain is no longer rhetoric, but actively being written into large-load tariffs, development agreements, and legislation.

These aren’t hypotheticals. The core attributes of our Playbook’s published clean flexible power framework — additionality, on-site clean firm power, load flexibility, ratepayer protection, community benefits, and accelerated interconnection — are already being demonstrated in signed deals and operational projects. 

The common thread across the examples below is clear: data centers that proactively bring additionality, embed load flexibility, and commit to community benefits earn accelerated interconnection, favorable utility treatment, and community support. The strategies described in this Playbook are already being executed by leading companies. What’s missing is the shared framework to scale them as a new default.

Additionality — New Clean Power Capacity (Google/DTE Michigan)

Google’s partnership with DTE Energy in Michigan represents among the most comprehensive additionality examples to date. In March 2026, Google confirmed a 1GW data center campus in Van Buren Township near Detroit, structured through DTE’s Clean Transition Tariff. The agreement commits 2.7GW of new clean power resources to the local grid — 1.6GW solar, 400MW four-hour storage, 50MW long-duration storage, 300MW additional clean resources, and 350MW demand response. Google covers all electricity costs and infrastructure needs, with $1.7B in affordability benefits to ratepayers over the life of the contract. Critically, every megawatt is new-build, directly filling the generation gap left by coal retirements rather than reshuffling existing clean energy claims. The deal touches every pillar of this framework simultaneously: additionality (2.7GW new clean generation), embedded load flexibility (350MW demand response), ratepayer protection (no cost pass-through plus active affordability investment), and accelerated grid access.

Load Flexibility — Data Centers as Grid Assets (Google FIT)

Google’s Flexible Interconnection Tariff (FIT) blueprint proposes the first tariff framework designed specifically for flexible large loads, centered on a core trade: data centers receive a mix of firm and conditional grid service while bringing enough accredited capacity through on-site power, PPAs, or VPPs to cover incremental capacity added to the grid. In exchange, data centers receive accelerated grid interconnection — as much as 3-5 years faster than standard queue timelines.

This isn’t theoretical. Google has integrated 1GW of demand response into utility contracts across the South and Midwest, scaling beyond initial I&M and TVA pilots to five to six data centers. The operational model routes compute to match available power in real time — a proof point that flexibility is a competitive advantage for interconnection speed, not a concession. Google’s conditions for participating in demand flexibility are instructive for program designers and regulators: adequate financial incentives, clear guardrails on disruption frequency and duration, and integration of flexible resources into utility planning processes.

Load Flexibility — Record Scale Priced by Utility (OpenAI/Georgia Power)

OpenAI committed up to 1GW of flexible demand, the largest disclosed demand-response commitment on record, at its 3.2GW Effingham County campus under an approved 25-year Georgia Power contract. The utility directly cited the commitment as enabling Georgia Power to “invest in fewer new generation resources to serve future growth.” With the latest approval of the flexible load contract within the utility’s portfolio of large-load agreements, Georgia Power projects that the incremental revenue will deliver approximately $950 million per year in customer savings beginning in 2029, supporting at least $15 per month for the typical residential customer using 1,000 kWh-month, raising projected savings for typical residential customer from $102 to $180 per year. This stacks on top of GA PSC approved rate-freeze in July 2025 and overall rate reduction approved in May 2026. Flexibility, priced by the utility, flowing to ratepayers tangibly demonstrates the grand bargain. The project represents the first within OpenAI’s Grid Protectors Program, serving as a national framework for its compute campuses to reduce power draw when grids are strained, with a UK trial demonstrating roughly 40% reduction within seconds.

Community-Aligned Additionality at Record Scale (Google Steel River Arkansas)

Google's Steel River Energy Center with Cypress Creek broke ground in July 2026 and will be the largest solar+storage project in the US as planned once complete: planned 2.5GW of solar and 2.9GWh of storage, 1.6GW solar and 1.9GWh storage in phases 1–2, with modular phases energizing by mid-2028 and final completion in 2029. Google serves as the anchor investor and offtaker. The additionality structure: a twenty-year anchor commitment renders the project bankable, so the capacity exists because the deal does. The storage firms solar into evening peaks, displacing marginal gas rather than midday surplus. The supply chain is entirely domestic: First Solar US-made modules, 400,000+ steel piles fabricated in Blytheville, Arkansas from US Steel coils, North American battery cells. And the community terms are concrete: ~700 construction jobs per phase, ~$300M in local tax revenue over project life, and $8M in community funds including energy-affordability support. VPPAs decarbonize grids on multi-year horizons while co-located clean firm energizes sites in 12–18 months — both are complementary instruments that Google is visibly running.

Accelerated Interconnection — Storage as a Grid On-Ramp

Aligned Data Centers’ partnership with Calibrant Energy demonstrates how on-site storage can unlock grid access where transmission constraints would otherwise impose multi-year delays. Aligned funded a 31MW/62MWh on-site battery system developed by Calibrant at its PDX-01 campus in Hillsboro, Oregon, after the local utility required demonstrated capacity additionality as a condition for interconnection approval. The utility studied the grid impact and determined that without new on-site capacity, the interconnection would strain local infrastructure. This is a critical precedent: additionality requirements are not just regulatory theory -- they are already being applied as binding conditions for interconnection.

The deployment — executed under Portland General Electric's flexible-interconnection framework with GridCARE modeling — is now cited by industry analysts as "a template other utilities can copy," with potential US storage mandates sized as a $37B battery-industry opportunity (Currence). Grid operators are moving the same direction: PJM is running a second surplus-interconnection effort while MISO (~15GW) and SPP (~14GW) study the same unlock. Storage has become an interconnection product, not a backup product — and the model converts what would be a queue bottleneck into a solvable capital problem.

Flexibility — BYOC Additionality via VPPs (Google/Voltus)

In June 2026 Google became the first hyperscaler to directly fund a virtual power plant (VPP) — a 100MW "Bring Your Own Capacity" commitment in PJM via Voltus. Google has since expanded the partnership by an order of magnitude, now building towards a total ~1.17GW of VPP capacity in PJM. 

The model has expanded on other axis too: Sunrun has partnered with Voltus in Illinois to aggregate thousands of existing residential batteries across PJM and MISO territory; Sunrun, Renew Home, and Tesla have pooled 16.8GW of distributed flexible capacity (millions of home batteries and smart thermostats across the US) and are marketing it directly to hyperscalers as a procurable distributed, dispatchable grid capacity resource that earns faster interconnection. 

The regulatory track is validating the structure with FERC ruling in Voltus's favor on statistical sampling in PJM — a decision that unlocks roughly 5GW of VPP and demand response capacity.

Execution Credibility — Phased Delivery (Microsoft Wisconsin)

Microsoft's Fairwater Phase 1 in Wisconsin went from announcement to operation (including pre-NTP development work) in 39 months against its region's 54-month average interconnection timeline — grid-connected, phased, delivered by an experienced developer. In a quarter where announced capacity vastly outran energized capacity, phased execution by proven operators provides the delivery-side proof of the execution-credibility filter described in Section 1.

Section 4 · The Framework

The Framework

How to Execute

The Data Center Power Playbook (DCPP) provides a replicable four-phase execution template for powering AI compute at scale through clean flexible power co-development.

Four-Phase Execution Template

1
Pre-NTP

Power-First Site Selection

Site selection process led by power potential and community fit. Evaluate existing transmission capacity and interconnection infrastructure, available excess generation, and land for on-site energy infrastructure development. Begin community engagement from day one: social license is earned in Phase 1, not retrofitted later. Pre-file for grid interconnection and engage partners early to coordinate accelerated interconnection through flexibility and additionality.

2
0–3 months (From NTP)

Initial Power Deployment

Deploy turnkey modular firm-power units, such as fuel cells and microgrid systems, to deliver initial MW capacity within months. Enable near-term compute revenue ahead of larger infrastructure development.

3
3–12 months (From NTP)

Modular BYOP Build-Out

Develop optimal on-site power portfolio: solar, energy storage, and complementary dispatchable generation scaling to hundreds of MWs within a year. Energy storage supports energy shifting and load curtailment. Flexible power electronics support modular scaling of power sources over time.

4
12–18 months (From NTP)

Grid Integration, Flexibility, & Optimization

Finalize accelerated grid interconnection, with on-site power enabling both additionality and load flexibility. Proactively fund targeted grid upgrades to expand interconnection capacity. Implement AI-driven controls to maximize compute (token output) per watt by orchestrating power, cooling, and compute workloads.

Acknowledging that active clean flexible power co-development is more complex relative to passive utility grid procurement or even on-site gas baseload deployment, the DCPP is executed through a live coalition of best-in-class, vetted partners across the energy ecosystem — spanning three interdependent pillars of on-site power, load flexibility, and grid interconnection — working in parallel across all phases to deliver power with unparalleled speed, reliability, and scalability. See Data Center Power Coalition via website Coalition tab.

Deeper Dives

The Playbook integrates on-site power, load flexibility, and accelerated grid power into a unified framework. Rather than being treated as procurement, power is treated as infrastructure co-development, pairing data center deployment with proactive energy infrastructure planning and execution. The solution is a dual on-site BYOP + accelerated grid interconnection strategy, prioritizing co-development of energy infrastructure (including co-located energy campuses) to achieve energy velocity and de-risk energy execution.

Four strategic development objectives drive the framework:

  1. Develop Large-Scale On-Site Power & Storage In Parallel: Power and compute scale together. Partner with energy developers to build large-scale portfolio of on-site generation and energy storage within 12 months, scaled modularly across phases near-term and over-time, while simultaneously pursuing grid interconnection.
  2. Design Power for Speed, Reliability, Scale, and Flexibility: Each power source serves a specific role in the timeline and grid coordination strategy. Ensure flexibility in design to accommodate modular deployment of power infrastructure to scale alongside compute infrastructure (SSTs), deploy turnkey firm-power modules (fuel cells, microgrid systems) to provide initial power for near-term uptime, establish EMS controls system to manage on-site energy assets with responsive capabilities to support load flexibility/curtailment signals from grid.
  3. Accelerate and Expand Grid Access Proactively: Use BYOP additionality and load flexibility to earn fast-track queue positioning. Assess grid capacity within site selection through power-first approach, proactively engage utility / grid partners early to embed load flexibility and fund targeted grid upgrades to earn “fast-track” and expanded grid interconnection.
  4. Maximize Data Center Operations Efficiency: Maximize compute revenue / token output relative to constrained energy input. Orchestrate inside-the-fence controls of power, cooling, compute management, and other levers to maximize compute per watt.

The framework addresses both immediate power needs and long-term reliability/profit optimization by enabling data centers to self-supply while coordinating accelerated grid interconnection. This shifts power from a constraint and risk to a competitive advantage — and the development itself from community liability to community asset — delivering energy velocity to accelerate speed to power and revenue.

Phase 1 begins with initial data center development site selection. Traditional data center site selection prioritized fiber proximity and metro-adjacent colocation. AI shifts data center development to where power can scale. AI training workloads are inherently latency tolerant and the majority of AI inference workloads have demonstrated higher latency tolerance from users. As OpenAI CEO Sam Altman noted, users are “surprisingly willing to wait for a great answer.” Post-training and test-time compute require scale and concentration.

This drives the shift towards power-first site selection to rural, power-rich locations with favorable land, power, and community development economics. Power-first site selection is not driven by fiber routing but rather land abundance, access to transmission capacity, and increasingly by community fit. A site that clears the power screen but fails the community screen does not achieve energy velocity but rather faces cancellation risk.

The Evaluation Framework

◆Existing transmission and interconnection infrastructure: What's the current capacity? What's the queue position for interconnection? Are there utility partnerships or policy tailwinds that could accelerate access?
◆Flexible-interconnection readiness: Does the utility or RTO framework support an accelerated or flexible interconnection pathway — can a defined demand-flexibility envelope be embedded in the interconnection agreement and traded for connection speed?
◆Available excess generation nearby: Are there stranded wind, solar, or other generation assets that could power on-site operations or feed the grid efficiently?
◆Land availability at scale: Large data centers require significant space for solar, battery storage, turbines, and modular power systems. Proximate clusters are acceptable; continuous land is not required, but density enables efficiency.
◆Construction and operations workforce: Securing construction crews now requires 2–3 years of advance planning, with EPC labor having surpassed equipment availability as the primary execution constraint. Screen beyond the build: operations staffing pool, technical trades, training-pipeline partners, and realistic drive-shed plus airport access for both phases.
◆Water strategy and community water context: Understand community water concerns (depletion, cost, quality) for site, especially within water-scarce regions. Closed-loop, zero-evaporation cooling designs largely decouple siting from water stress, but water shapes social license regardless of design. Plan for zero- or low-water cooling, account for the shared basin rather than the parcel, and communicate with communities early, as water is consistently a top driver of local opposition.
◆State policy and regulatory posture: Does the state have development frameworks in place to mitigate regulatory risk? Does the utility have a large-load tariff developed with the commission? Are there enablement pathways — expedited siting regimes, designated energy zones, single-point permitting?
◆Community fit and engagement readiness: What has the community already said — through measures passed or pending, campaigns organized, and priorities voiced in public processes? Understand what commitments matter to the community as priorities and whether there is a credible path to build support. Ensure commitments are written into the development agreement and are independently verified, to document community support and provide transparent accountability.
◆Natural gas pipeline infrastructure: Relevant for the dispatchable complement of the portfolio — fuel cells and modular firm units sized at a fraction of campus load, not gas baseload (Dive 2A). The screen is existing pipeline access with available capacity: proximity to in-service infrastructure keeps the complement on the campus timeline, while a site that would require new pipeline construction inherits the gas system's own queue and fails the timeline test.
◆Air permitting pathway: Any on-site combustion — backup generation included — routes through air permitting. Screen attainment status and likely permitting class, headroom under minor-source thresholds, and offset availability and cost in nonattainment areas. Keeping combustion to a modest dispatchable complement keeps projects below major-source thresholds, where permitting timelines are typically shorter.

The Execution Approach

◆Pre-file for grid interconnection early: Engage utility partners before or immediately after land acquisition. Communicate plans to meet additionality and flexibility requirements under established large load tariffs. Position for fast-track queue placement.
◆Embed grid capacity analysis software within initial site selection: Optimize for higher starter capacity and faster interconnection timelines. This drives site ranking.
◆Engage developer partners early to plan and execute the BYOP on-site portfolio strategy: Align timing expectations. Coordinate land use for both compute infrastructure and energy infrastructure. Lock in EPC capacity and labor commitments alongside equipment procurement — labor availability is now a critical path dependency.
◆Engage local partners to streamline permitting approval process: Build relationships with local government, community stakeholders, and permitting authorities early in the site selection process. Proactive community engagement, local partnership, and communication of public benefits accelerates approvals and reduces the regulatory friction that delays NTP.
◆The industry's own project data now makes the sequencing explicit: community and political opposition drives the majority of cancellations, and the operative lesson as Currence expresses is to "win over the county before you win the queue." The bargain that works is enforceable measurable community benefits delivered and full costs covered.
◆Set enforceable water position: Commit to quantified water parameters in development agreements — metered usage, established baselines, third-party measurement and verification. Leading practice extends to watershed-positive commitments, replenishing more than the facility consumes

The industry's own project data now makes the sequencing explicit: community and political opposition drives the majority of cancellations, and the operative lesson as Currence expresses is to "win over the county before you win the queue." The bargain that works is enforceable, measurable community benefits, with full costs covered.

Phase 2 delivers the first critical milestone: firm MW capacity within 90 days of NTP. This proves "power in months not years" and de-risks the immediate revenue timeline for compute operations.

Install turnkey, modular, firm-power units (ex: fuel cells, microgrid systems) to provide immediate power, favoring modular fuel-efficient combustion-free designs that stay clear of the air-permitting friction screened in Phase 1. The solution toolkit is widening including oxy-combustion capture integrated modular turbines among other technologies. Recent precedent shows design responsiveness alone does not secure social license; Phase 1's community groundwork remains the gating work. 

Initial turnkey modular firm-power units cost >$100/MWh, higher than long-term on-site generation. But this is the premium paid for speed. The value is time-to-revenue. GPUs begin generating revenue within three months while larger scale on-site power development and grid interconnection advance in parallel.

This phase achieves multiple objectives:

◆Near-term uptime: Operations begin with reliable power available immediately.
◆Front-loaded access and speed to power: Revenue-generating capacity is live while Phase 3 and Phase 4 infrastructure develops in parallel.
◆Operational proof-of-concept: The site demonstrates power stability, equipment cooling, and grid communication capabilities at scale before major capital deployment.

These initial turnkey modular firm-power units remain operational as part of the site's power portfolio, providing continued firm capacity as Phase 3 and Phase 4 assets come online and the overall power mix scales.

Phase 3 deploys the optimal on-site power portfolio, solar, energy storage, dispatchable gas capacity, scaling to hundreds of MWs within the 12-month window, in parallel to Phase 2's initial power operation. The layered approach includes:

  • Large-scale battery storage at the data center perimeter serves multiple functions: energy shifting of excess intermittent generation (solar production peaks during the day; compute demand may peak at night), dispatchable power for peak demand periods, and load curtailment support. Shifting power consumption to battery and on-site generation reduces peak grid demand and improves efficiency.
  • Flexible power electronics (Solid State Transformers / SSTs) enable modular scaling of power infrastructure over time. As compute density increases or new workloads are added, power distribution systems adapt without major re-engineering. SSTs also accommodate integration of emerging power technologies as they become commercially available.
  • Sophisticated Energy Management System (EMS) controls manage on-site generation, storage, and demand in real-time. The EMS receives signals from the grid operator (curtailment requests, frequency regulation needs) and responds automatically, adjusting on-site power dispatch, battery charging/discharging, and flexible load management — the delivery mechanism for contracted flexibility.
  • Rack-level supercapacitors inside-the-fence absorb the volatile load oscillations from GPU clusters. Compute loads swing from 10% to 100% utilization with variations at millisecond, second, and minute frequencies. Supercapacitors and battery storage provide frequency regulation, protecting both compute hardware and grid connection, and ensuring compute reliability even under extreme load volatility.

As a procurement reality check: the power electronics layer is now a supply-chain constraint in its own right with UPS and power-conversion equipment (required for all developments) carrying lead times approaching 200 weeks. The President’s August 2026 Executive Order 14421 restricting grid equipment from 24 countries (transformers, inverters, battery systems, UPS) adds an additional federal overlay across the electrical stack. This makes it all the more important to order equipment early; the modularity advantage is real but not exempt from queues. A hybrid multi-source power strategy buys optionality as well as resilience (portfolio with grid and multiple on-site paths has no single point of power failure). Staged delivery and disciplined multi-phased execution is what converts the strategy to speed.

By the end of Phase 3, the data center operates with diverse, modular on-site power assets — solar, storage, a minimal complement of dispatchable capacity, and supercapacitor buffers — with the EMS control systems to optimize them. The site is substantially self-sufficient and positioned for grid integration.

Phase 4 transitions the site from self-powered operation to grid-integrated optimization. The data center becomes a dynamic grid asset, providing flexibility to the utility while further optimizing its own operations. The layered approach includes:

  • Finalize fast-track grid interconnection with utility partners: On-site BYOP provides additionality (the grid doesn't pay for self-supply, so additional on-site generation doesn't displace grid responsibility). The system is designed to meet curtailable load flexibility requirements, leveraging on-site energy storage and other levers to reduce or shift grid load consumption in response to grid stress events. This additionality and flexibility earns accelerated interconnection. The market now formally scores it: PJM's filed framework counts storage, demand response, and uprates as qualifying capacity for large-load entry — making flexibility capacity additionality by the market's own accounting.
  • Fund targeted grid upgrades to expand capacity: Grid Enhancing Technologies (GETs) — such as Dynamic Line Rating (DLR), conductor coating, and reconductoring — can enhance existing transmission capacity and reduce power losses. Work with utilities to identify and implement the grid infrastructure upgrades that clear interconnection bottlenecks.
  • Implement AI-driven orchestration controls to manage inside-the-fence power, cooling, and compute systems simultaneously: The goal is to maximize token output per watt. Given constrained power input (defined by on-site generation plus grid interconnection capacity), the data center should operate to produce maximum revenue and token output from every megawatt available. AI models predict workload patterns, optimize cooling efficiency in real time, and distribute compute loads across hardware to minimize power waste.
  • Full load flexibility and curtailment capability become operational: The data center integrates with grid interface platforms to respond to grid signals in real-time, in compliance with its flexible interconnection agreement. When the grid experiences stress (peak demand, generation shortfall, frequency drop), the data center reduces non-critical workloads or shifts consumption to on-site energy storage resources. This flexibility earns accelerated interconnection and preferential utility tariff agreements, creating a virtuous cycle: flexibility → faster interconnection → lower costs → higher margins.
  • Storage-led flexibility bridges community and grid risk: The requirement has sharpened from both directions in 2026: when grid emergencies force curtailment onto backup diesel — as DOE emergency orders have repeatedly done — communities bear the air-quality cost. Storage-heavy campus design is the bridge that spares both sides. Mass disconnection is also now a documented grid risk, with enforceable ride-through standards on the calendar. Storage at the point of interconnection buffers both ends — riding through grid faults rather than dropping load en masse — as smarter power electronics make mandated ride-through native.

The contractual path: the energization ladder. From first approach to full power, the commercial path now has a recognizable shape: an Early Load Operations Agreement (study-stage commitments), maturing to a Conditional Load Operations Agreement (conditions and curtailment terms attached), maturing to a full Energy Services Agreement. Each rung trades certainty for commitment and is how willingness becomes obligation in practice.

By Phase 4 completion, the site is fully grid-integrated, operationally efficient, and financially optimized.

Energy co-development is more complex than traditional utility procurement. The Playbook is executed through partnerships with companies across the energy ecosystem, organized across three interdependent pillars to achieve energy velocity at scale.

The Sijbrandij Foundation's Data Center Energy Initiative serves as an agnostic coordination layer -- convening partners across all three pillars, architecting the framework, and aligning around shared timelines and energy velocity objectives. The Data Center Power Coalition is live at ++datacenterpower.ai++ (see Coalition tab) a curated set of vetted partners mapped to the categories below, currently spanning 15 members across both original and newly added categories — including Measurement & Sensing and Power-First Site Selection.

On-Site Power

Deploy rapidly across multiple phases, developed in parallel to compute, for greater speed and reliability of power

Load Flexibility

Enable data center to be a dynamic grid asset to earn faster grid interconnection and deliver rate suppression aligned with policy priorities

Grid Interconnection

Proactive utility planning starting with power-first site selection; focus on expanding grid capacity faster for accelerated access and scalability

Successful execution of the Playbook depends on coordinated partnerships across these three pillars, aligned around shared timelines and energy velocity objectives.

Pillar 1: On-Site Power Infrastructure Development

Develop the on-site energy campus in parallel to compute. The objective is to deliver reliable power faster and more flexibly, providing self-supply in complement to grid interconnection.

Energy Infrastructure Developers

  • Strategic Purpose: Foundational partner and executor for the development of the large-scale on-site BYOP portfolio, managing the complex integration of solar, storage, complementary dispatchable power, and other resources behind-the-meter.
  • Key Criteria: 1) Leading utility-scale power developer with demonstrated track record of developing GW-scale power generation; 2) On-site power integration expertise — capability to manage complex integration of large-scale power on-site for BYOP; 3) Financing capability, likely structured as longer-tenor PPA (15-years+) with embedded buy-out option.

Initial Turnkey Modular Firm-Power Equipment Providers

  • Strategic Purpose: Deliver the first firm MW capacity within 90 days of NTP through modular, rapidly deployable power units (fuel cells, microgrid systems) to enable immediate compute operations while permanent on-site generation develops in parallel.
  • Key Criteria: 1) Speed of deployment — proven ability to deliver 50+MW within 90 days of site readiness; 2) Modularity and scalability — units that can be incrementally added as demand grows; 3) Equipment availability — manufacturing capacity available to deliver on orders; 4) Fuel efficiency and avoidance of permitting risks — combustion-free power to steer clear of local air pollution and air permit risks.

Power Electronics (SSTs)

  • Strategic Purpose: Enable modular scaling of power infrastructure over time alongside compute, provide flexible improved power conversion and distribution system design (natively supporting 800V DC power distribution architecture), and accommodate integration of emerging power technologies as they become commercially available.
  • Key Criteria: 1) Proven commercial readiness — demonstrated technical capability prior to deployment; 2) Equipment availability — manufacturing capacity available to deliver on orders.

Energy Storage

  • Strategic Purpose: Provide energy shifting of excess intermittent generation, dispatchable power for peak demand, load curtailment support, and frequency regulation for the grid connection. Large-scale battery storage at the data center perimeter for energy shifting and peak demand.
  • Key Criteria: 1) Equipment availability — manufacturing capacity to deliver on orders at required scale; 2) Technology readiness — proven commercial deployment (not pilot/demonstration stage); 3) Integration readiness — commercially ready for behind-the-meter integration with EMS and on-site power portfolio.

GPU Pulse Load Mitigation

  • Strategic Purpose: Absorb highly volatile sub-minute load oscillations from GPU clusters, protecting upstream generation equipment and grid connection from power-quality degradation. Without pulse-load mitigation, GPU power swings stress on-site generation and the grid system. Two primary architectures: perimeter-level E-STATCOM (supercapacitor rooms at the site boundary) and rack-level supercapacitor sidecars (solving at the source of oscillation to avoid transient and harmonic challenges in the medium-voltage network).
  • Key Criteria: 1) Sub-minute response capability for GPU power oscillations at rack or row level; 2) Cycle life sufficient for continuous high-frequency discharge/recharge at GPU duty cycles (millions of cycles per month); 3) Equipment availability — manufacturing capacity to deliver on orders at data center scale; 4) Integration readiness — commercially ready for deployment inside-the-fence alongside compute racks, power electronics, and EMS/DCIM systems.

Pillar 2: Load Flexibility Integration

Enable the data center to operate as a dynamic grid asset — incorporating sophisticated EMS controls to manage on-site energy assets with responsive capabilities to support load flexibility and curtailment signals from the grid. Orchestrate power consumption, cooling, and compute workloads to respond to grid conditions and maximize compute output relative to constrained power input. Flexibility provides policy currency to enable fast-track interconnection, tariff benefits, and rate suppression.

Energy Management Systems (EMS)

  • Strategic Purpose: Central control layer for the on-site power portfolio — manages real-time orchestration of on-site generation, storage, and demand to serve data center load demand. Meet load flexibility and curtailment requirements by receiving dispatch signals from grid interface partners and determining how to meet curtailment across controllable on-site assets.
  • Key Criteria: 1) Capability to manage portfolio of multiple on-site power resources simultaneously in real-time; 2) Integration with grid interface partners to receive and act on dispatch signals; 3) Integration with data center operational orchestration layer for coordinated power-compute optimization.

Grid Interface Layer

  • Strategic Purpose: Interface between the grid operator/utility and the data center's EMS — receiving dispatch signals (curtailment requests, frequency regulation, demand response events) from grid operators and translating them for the EMS to execute against.
  • Key Criteria: 1) Established relationships and integration with RTOs/ISOs and utility dispatch systems; 2) Real-time signal processing and communication capability; 3) Track record operating in demand response and curtailment markets.

Data Center Operational Orchestration

  • Strategic Purpose: Maximize compute per watt by orchestrating inside-the-fence power, cooling, and compute systems simultaneously — AI-driven controls that predict workload patterns, optimize cooling efficiency in real-time, and distribute compute loads across hardware to minimize power waste.
  • Key Criteria: 1) Data center specialized (vs. broader building management systems); 2) AI/ML-driven optimization of compute workload scheduling relative to power availability; 3) Ability to shift non-critical workloads or redistribute inference tasks to battery-backed resources during grid stress events; 4) Integration with EMS for coordinated power-compute optimization.

Measurement & Sensing

  • Strategic Purpose: Provide the high-resolution electrical measurement layer beneath the flexibility stack — capturing the sub-second power physics (GPU load transients, equipment degradation signatures, actual capacity utilization) that standard monitoring systems cannot see, and delivering certified, decision-grade telemetry that EMS, orchestration, and grid-facing layers act on. Serves as the independent measurement-and-verification substrate for flexible interconnection — certifying dispatchable capacity and verifying flexibility delivery to utility counterparties. Measurement is the product in this category: providers whose primary function is workload or power control are mapped to the orchestration and EMS categories, which they serve as consumers of this layer.
  • Key Criteria: 1) Sub-second sampling across the full site power chain — from the point of interconnection through distribution to rack level — with resolution sufficient to characterize GPU-scale load transients; partial-chain or device-side-only telemetry does not meet the substrate standard; 2) Certified, attested data quality — telemetry trustworthy enough to serve as evidence for capacity decisions, SLAs, and utility flexibility commitments; 3) Open integration surface — telemetry consumable by EMS, orchestration, and DCIM platforms via open industry standards; 4) Production track record in operating data center environments.

Pillar 3: Proactive Grid Planning & Interconnection Acceleration

Integrate power-first site selection with proactive utility engagement — embedding grid capacity analysis within initial site evaluation, proactively engaging utility and grid partners to embed load flexibility and fund targeted grid upgrades to earn “fast-track” or expanded grid interconnection.

Power-First Site Selection (Software / Platform)

  • Strategic Purpose: Operationalizes Phase 1, embedding grid capacity analysis within the initial site selection process to find and qualify sites where power can scale, optimizing for higher starter capacity and faster interconnection timelines. Because the Playbook develops energy alongside compute, this category's scope spans both pillars it serves (Pillar 1 & Pillar 3): screening sites for grid interconnection potential and for on-site energy co-development — land, buildability, and co-located generation feasibility evaluated in one workflow. Site selection is the product in this category: providers whose primary function is accelerating capacity at an already-chosen site map to the flexible interconnection categories this layer hands off to.
  • Key Criteria: 1) Specialized grid capacity analysis for large loads — ISO/RTO-aligned power-flow modeling with comprehensive transmission and interconnection data, and queue-positioning and timeline scenario modeling across target geographies; 2) Co-development screening — capability to model load and generation, including co-located configurations, at a point of interconnection — not load alone; 3) Integrated pre-NTP workflow — land, buildability, zoning and permitting, and community-context screening alongside grid analysis; 4) Production adoption by data center developers and operators.

Utility & Transmission Partners

  • Strategic Purpose: Coordinate proactive grid interconnection planning starting before or immediately after land acquisition, communicate BYOP additionality and flexibility plans to position for fast-track queue placement, and identify and execute targeted grid infrastructure upgrades (Dynamic Line Rating, conductor coating, reconductoring) that clear interconnection bottlenecks.
  • Key Criteria: 1) Willingness to engage in proactive interconnection planning ahead of standard queue timelines; 2) Openness to co-funded grid upgrade investment as part of interconnection acceleration; 3) Alignment on large load tariff structures that reward additionality and flexibility.

Policy & Regulatory Partners

  • Strategic Purpose: Collaborate at state and federal levels to establish and refine regulatory structures that provide the framework for data centers to deliver capacity additionality and load flexibility in exchange for accelerated interconnection and favorable tariff treatment.
  • Key Criteria: 1) Demonstrated engagement with FERC, RTOs/ISOs, and state-level regulatory proceedings; 2) Understanding of evolving large load tariff landscape; 3) Ability to represent data center energy interests in policy development aligned with ratepayer protection and grid reliability.

Emerging Energy Technology Radar

The Playbook is designed for technology flexibility. The modular architecture supports the integration of new power sources and storage technologies as they reach commercial viability. The following technologies are on active radar, with limited commercial availability for data center scale deployment today due to location-specific constraints or technological readiness:

  • Enhanced/Advanced Geothermal: Growing interest and improving technology. Limited deployed capacity at data center scale today, but potential for high-capacity, low-footprint baseload power at specific geothermal-rich site locations (geographically constrained).
  • Small Modular Reactors (SMRs): Promise firm baseload power with minimal land footprint. Earliest commercial deployments expected early 2030s. Potential to provide 24/7 firm carbon-free power at data center scale without combustion or intermittency constraints.
  • Advanced Energy Storage (iron-air, flow batteries, compressed air, gravity): Could complement lithium-ion for longer-duration energy shifting and seasonal storage. Commercial viability for data center scale expected within 2–5 years.
  • Fusion: Long-term potential. No near-term commercial pathway for data center deployment.

The Playbook's modular architecture does not depend on any of these technologies reaching commercial readiness on a specific timeline. But it is designed to integrate them as they do — without re-platforming the power supply stack. This is deliberate: the framework that works with today's commercially available technologies should also work with tomorrow's.

Section 5 · The Call

The Call

How to Get Involved

aric@sijbrandijfoundation.org →

Industry energy defaults are forming now. Policy frameworks are actively being codified. 2027 is when the writing happens across statehouses, utility commissions, and market design. The companies and frameworks that define development standards now will determine how the next 50GW of US data center capacity gets powered, and the window to help set them well is open now.

Following extensive research, industry engagement, and direct conversations with operators, developers, regulators, and technologists across the AI data center energy ecosystem, the DCPP represents our best current understanding of how to solve this challenge. Execution is underway: the coalition is live and growing, and the framework is being applied in practice.

The Sijbrandij Foundation's Data Center Energy Initiative was created to solve the data center energy challenge to accelerate the AI transition. We are building an agnostic operational platform to convene and coordinate a coalition of energy partners to execute clean flexible power co-development for the data center industry.

The DCPP is an open living framework. We encourage independent adoption, and will also facilitate select reference project deployments with partners to demonstrate the model in practice. The Playbook will be publicly iterated regularly as we receive new information including as reference project deployments deliver real-world data, as the regulatory and market landscape evolves, and as the partner coalition expands.

The coalition is organized around specific partner categories including:

  • On-site power developers (solar, storage, complementary dispatchable generation)
  • Initial turnkey modular firm-power equipment providers (fuel cells, microgrid systems)
  • Flexible power electronics (solid-state transformers, power conversion systems)
  • Energy storage and GPU pulse-load mitigation
  • Measurement & sensing (decision-grade telemetry and M&V)
  • Energy management and controls systems (EMS, grid interface platforms, operational orchestration)
  • Power-first site selection (grid analytics and development software)

The Foundation is facilitating select reference project deployments to validate the model in practice — directly coordinating energy partners with AI infrastructure partners to demonstrate the Playbook framework in the field. Partners who engage early help define the standard that the broader market will adopt.

We evaluate partner capabilities, track record, commercial readiness, and alignment with the Playbook framework. This is not a vendor directory — it is a coordinated coalition built around shared execution standards and timelines.

Next step: Email aric@sijbrandijfoundation.org with an overview of your solution (technology, deployment track record, commercial readiness, and relevant capabilities).

Email aric@sijbrandijfoundation.org →

Engagement starts with a scoping call. We assess whether opportunities within your development pipeline fit the Playbook framework and determine where the Foundation's partner network adds the most value. The Foundation's role is agnostic — we are not selling a specific technology or partner, but coordinating across the coalition to match the right capabilities to each project's needs.

The Foundation is facilitating select reference project deployments alongside early participants. These engagements validate the model in practice, de-risk commercial execution with energy providers, and generate shared learnings that strengthen the framework for all participants. You retain control of your energy strategy; the Foundation accelerates the convening, vetting, and partner coordination.

This engagement is designed for companies building GW-scale AI infrastructure pipelines that recognize power as their primary bottleneck and want to move faster than assembling the energy ecosystem alone. The Foundation brings the coalition, framework, and deployment coordination.

Next step: Schedule a scoping call to discuss your pipeline. Email aric@sijbrandijfoundation.org with an overview of your development plans (target geographies, timeline for power decisions, where coordination support would be most valuable).

Schedule a scoping call →

The Data Center Power Playbook provides a framework through which data centers deliver measurable public benefit: local energy affordability improvement, grid reliability strengthening, economic development, and avoidance of negative environmental impacts including local air pollution and water resource strain.

The Foundation's position is direct: data centers that bring additionality and embed flexibility should earn accelerated interconnection and favorable tariff treatment. This alignment between data center development priorities and public interest is embedded in the Playbook's execution model.

Co-development models are not yet widely represented in regulatory proceedings as an alternative to conventional grid service or on-site gas baseload procurement. We encourage commissions to invite framework-level input early — before interconnection filings arrive — and we encourage the developers we work with to engage host communities and regulators proactively rather than at the permit stage. Early engagement on both sides is what shapes workable rules.

Large-load tariffs and interconnection standards are actively being developed across utilities, states, RTOs, and federal regulators. The Foundation is engaged in this landscape and available as a resource — not as an advocate for a specific company or technology, but as a source of framework-level analysis and evidence-based input on topics including tariff design, interconnection policy, grid impact modeling, and rate suppression potential.

If this framework aligns with your policy objectives, we welcome the opportunity to work together to ensure these standards incentivize the best outcomes for both industry and the public.

Next step: Email aric@sijbrandijfoundation.org to discuss how the Playbook framework intersects with your regulatory developments.

Engage the Foundation →

Stay Updated

The Playbook is a living document.

We iterate publicly as reference projects deliver real-world data, the regulatory landscape evolves, and the partner coalition expands. Get updates when we publish.