Aligned Data Centers
Build-and-lease hyperscale capacity: develops powered data-center campuses and leases IT capacity to hyperscalers/AI tenants on long-term (typically 10-15yr) contracts priced by megawatt; capital-intensive, backed by infrastructure equity + project debt. Private, sponsor-owned (not a public REIT).
Only the Oct 2025 ~$40B is a true enterprise-value mark. The Jan 2025 figure is capital raised (equity valuation undisclosed), and the 2018 Macquarie entry price was never publicly disclosed (the 0.5 is a scale placeholder, not a reported number). Aligned is private throughout — no primary/secondary equity price has ever been public.
Earnings, margins, COGS & capex
Private, sponsor-backed platform; no public financials. The economic story is capacity growth, not reported earnings: ~85MW across 2 sites (Dallas, Phoenix) at Macquarie's 2018 entry to ~5GW operational + planned across ~50 campuses by 2025. Value is underwritten on contracted lease MW and a powered-land development pipeline, monetized via the Oct 2025 ~$40B EV transaction. Revenue, margins, and COGS are undisclosed.
Revenue trend
Margins
Structurally: stabilized operating campuses throw off high-margin recurring lease revenue; blended near-term FCF pressured by aggressive build capex.
COGS structure
Not disclosed. Cost base is dominated by power (electricity procurement + power availability), depreciation on built shells/fit-out, cooling/mechanical-electrical opex, and land. Speed-to-power and power-cost management are the core operating levers.
Capex
Growth-capex led and large: >$7B debt + >$5B primary equity committed Jan 2025 to fund the pipeline. Each new hyperscale campus runs into the hundreds of millions to multiple billions; the ~5GW planned base implies a multi-year, multi-tens-of-billions build program.
Latest earnings
n/a
n/a
- Operational + planned capacity
- ~5GW across ~50 campuses
- 2025 capital raised
- >$12B (>$5B equity + >$7B debt)
- Enterprise value (Oct 2025 deal)
- ~$40B
- Cooling patents
- 50+ issued / 500+ claims
Growth drivers
- Hyperscale + AI training/inference demand driving record leasing of large contiguous power blocks
- Speed-to-power — ability to deliver powered capacity fast in supply-constrained markets (Northern Virginia, Phoenix, Dallas, Chicago, Salt Lake City, Ohio)
- High-density AI capability via DeltaFlow~ liquid cooling (up to ~300kW/rack) and Delta3 adaptive air cooling (scale-in-place to ~50kW/rack)
- Latin America expansion (Brazil, Mexico, Chile, Colombia) into underserved cloud markets
- Deep-pocketed new ownership (BlackRock/GIP, MGX/Mubadala, Microsoft, Nvidia via AIP) funding and demand-anchoring the build
Bull & bear
Aligned is a scarce, at-scale owner of the one input the AI buildout cannot manufacture fast enough — powered, coolable, contracted data-center capacity — now capitalized and demand-anchored by the deepest-pocketed consortium in infrastructure. The ~$40B mark is a floor set by strategic buyers, not a peak.
- ~5GW pipeline in the tightest US power markets is effectively irreplaceable on any near-term timeline; scarcity value compounds as interconnection queues lengthen
- New owners (BlackRock/GIP, MGX, Microsoft, Nvidia via AIP) bring both capital and offtake — the platform is funded and its capacity is partly pre-sold to the very firms buying it
- Cooling IP (DeltaFlow~ to ~300kW/rack) makes it a preferred home for the densest GPU training clusters, the highest-value workloads
- Long-duration, contracted hyperscale leases produce bond-like recurring cash flow once campuses stabilize — the reason infrastructure capital pays record multiples
- Record $40B EV validates the platform against the prior $16.6B AirTrunk benchmark; strategic ownership de-risks the funding of the remaining build
A ~$40B private-market mark on a levered, cash-burning build-phase platform prices in near-flawless execution of a multi-GW pipeline and uninterrupted AI demand — either a power bottleneck, a hyperscaler capex pause, or a financing shock impairs the equity.
- Valuation is a peak-cycle strategic price; if AI infrastructure demand digests, re-leasing and mark-to-market on the ~$40B could reverse sharply
- The equity is heavily levered (>$7B debt raised in 2025 alone) against long-dated assets — rate and refinancing risk is real if cash flows lag the build
- Power, not capital, is now the binding constraint; grid interconnection and electricity-cost inflation can strand development capital and delay revenue
- Customer concentration among a few hyperscalers means a single tenant's capex cut or non-renewal disproportionately hits utilization
- Related-party dynamics (owners are also tenants/suppliers) can flatter demand optics but complicate arms-length pricing and true external demand read
- Intensifying competition for land/power/construction erodes returns on the next tranche of capacity
What it is worth
Last priced transaction (Oct 2025 take-private) cross-checked against public data-center comps and a per-MW / per-GW sanity read. No public equity price exists.
If AI leasing digests, power costs rise, or financing tightens, the ~$40B mark is a cycle peak on a levered build-phase asset and equity value could compress materially on delayed or unfilled capacity.
~$40B EV is a fair strategic clearing price given scarcity of powered capacity and consortium demand/funding; value accretes with each stabilized campus but is gated by power + execution.
Justified-to-higher: if the ~5GW pipeline is fully contracted and powered and AI demand persists, stabilized EBITDA supports the ~$40B EV and strategic owners hold/expand — floor, not ceiling.
The AIP/MGX/BlackRock-GIP deal sets EV at ~$40B on ~5GW operational + planned — roughly ~$8B per GW of blended (built + pipeline) capacity, a premium that capitalizes scarce powered land + contracted AI demand, not just in-place cash flow. Public comps DLR and EQIX trade at high-teens-to-20x+ EV/EBITDA; a private hyperscale platform with a heavier development pipeline commands a scarcity premium but carries more execution + funding risk. The mark reflects strategic-buyer + peak-AI-cycle pricing, and is the largest data-center acquisition on record (vs the prior $16.6B AirTrunk deal).
SWOT
Strengths
- Scarce, contracted powered capacity at scale (~5GW) in the most supply-constrained US markets — the single most valuable asset class in the AI cycle
- Differentiated cooling IP (Delta3 adaptive air, DeltaFlow~ liquid to ~300kW/rack; 50+ patents) enabling high-density AI deployments and scale-in-place without stranding capacity
- Now owned/backed by a demand-anchoring consortium (Microsoft as offtaker, Nvidia as silicon partner, BlackRock/GIP + MGX capital) that shortens the leasing and funding path
- 7-year track record scaling ~60x (85MW to ~5GW) under Macquarie ownership
Weaknesses
- Extremely capital-intensive and levered — growth outruns internally generated cash, dependent on continuous equity + debt access
- Customer + revenue concentration opaque — hyperscale leasing is concentrated among a handful of tenants, raising counterparty and re-leasing risk
- No public disclosure — limited external transparency into margins, contract terms, and utilization
- Power procurement and interconnection timelines are outside its control and increasingly the binding constraint
Opportunities
- AI capex supercycle — hyperscaler + neocloud demand for multi-hundred-MW campuses far exceeds available powered supply
- Liquid-cooling leadership positions it for the densest GPU clusters (training superpods)
- LatAm and new US metros (Ohio, Maryland) as greenfield growth
- Owner-affiliated demand (Microsoft/AIP) could convert into anchor leases across the pipeline
Threats
- Power scarcity, grid interconnection queues, and rising electricity costs capping how fast capacity comes online
- Potential AI-demand air-pocket / hyperscaler capex digestion that could soften leasing and pressure the ~$40B valuation
- Aggressive competitor build-out (Vantage, QTS/Blackstone, CoreWeave, CyrusOne, Crusoe, Switch) bidding up land, power, and construction inputs
- Rate/financing risk on a heavily levered, long-duration asset base
- Community/permitting/water-and-power backlash slowing new campuses
Moats, dependencies & bottlenecks
Moats
Scarce powered land + secured grid interconnection in supply-constrained metros The true bottleneck of the AI cycle; multi-year interconnection queues make an assembled ~5GW position very hard to replicate.
Switching cost and lease duration create bond-like, sticky recurring revenue once stabilized.
50+ patents) Enables higher density + scale-in-place economics; a real edge but competitors (and OEMs like Vertiv/Schneider) are converging on liquid cooling.
Capital scale + owner-affiliated demand (AIP/BlackRock/MGX/Microsoft/Nvidia) Funding depth and anchor-tenant relationships are a cycle advantage, but are sponsor-dependent rather than intrinsic to the asset.
Dependencies
Capacity delivery is gated by power availability and interconnection timelines; the primary constraint on growth.
Microsoft, other cloud + neocloud operators) Revenue depends on continued large-block leasing; concentrated counterparty base.
AI compute demand — and thus high-density leasing — tracks the GPU/accelerator buildout; Nvidia is also an AIP consortium partner.
Continuous access to project debt and equity is required to fund the multi-GW pipeline; rate-sensitive.
Eaton, Schneider, Caterpillar, generators, transformers, chillers) Long-lead-time gear (transformers, gensets, switchgear) can delay commissioning.
Speed-to-power depends on construction throughput amid industry-wide labor and materials competition.
Advantages
- Assembled ~5GW of scarce, powered capacity across ~50 campuses in premium US markets
- Liquid-cooling capability (~300kW/rack) suited to the densest AI training clusters
- Scale-in-place design (Delta3) reduces stranded capacity and improves capital efficiency
- Deep-pocketed strategic ownership that both funds and consumes capacity
- Established multi-region footprint including LatAm growth optionality
Weaknesses
- High leverage and negative near-term free cash flow
- Opaque financials + concentrated, undisclosed customer base
- Growth fully dependent on external power and capital, both outside its control
- Valuation set at a cycle peak, exposed to any AI-demand or rate shock
Bottlenecks
- Power availability and grid interconnection timelines
- Long-lead electrical gear (transformers, switchgear, generators)
- Construction labor + EPC capacity in hot metros
- Continuous access to low-cost capital for a levered build program
- Water/permitting and community approvals for new campuses
Top signals & trends
Top signals
Deal close validates the mark and unlocks consortium funding + offtake; watch for any regulatory/CFIUS delay given Gulf sovereign (MGX/Mubadala, KIA) participation.
Confirms the pipeline is contracted, not speculative.
Secured power is the gating item for revenue timing.
A capex pause among the megacap buyers is the clearest early warning for data-center leasing.
Bearish if widening · Levered, long-duration assets are sensitive to financing conditions.
Trends
Primary demand tailwind; drives record leasing of large power blocks.
Plays directly to DeltaFlow~; raises the bar for air-only competitors.
Shifts industry value to whoever controls powered land — favors incumbents like Aligned but caps growth pace.
BlackRock/GIP, Blackstone, KKR, MGX, DigitalBridge competing to own scaled platforms; supports valuations but bids up inputs.
Permitting friction and PUE/water efficiency increasingly gate new builds.
Ecosystem & competitor graph
Suppliers feed the company; customers pull from it. Line thickness shows the strength of each tie (supply-chain dependency, customer earnings contribution). Hover to isolate a tie.
GPU supplier whose accelerators drive high-density demand; also an AIP consortium/ownership partner.
Power + thermal management (UPS, cooling, busway) for data-center fit-out.
Electrical distribution, switchgear, and power infrastructure.
Power management + cooling systems; converging on liquid cooling.
Backup generators / gensets for critical power.
Power, cooling, and integrated infrastructure for AI data centers (Taiwan-listed).
Power + grid interconnection — the primary input and gating constraint.
Hyperscaler tenant and now an AIP consortium owner — anchor-demand relationship.
Google, Meta, Oracle) Large-block leasing counterparties for AI/cloud capacity (Aligned does not disclose named tenants).
OpenAI, xAI, CoreWeave) Fast-growing demand for dense GPU capacity; xAI is itself an AIP consortium member.
Largest public data-center REIT; overlapping hyperscale + colocation footprint and a direct public comp for valuation.
Global interconnection + colocation leader; more retail/interconnection-weighted but competes for hyperscale campuses.
Private (DigitalBridge-backed) hyperscale campus specialist; closest strategic peer in large-block AI capacity.
Private, Blackstone-owned; aggressive hyperscale build-out and a direct competitor for the same tenants and power.
Private (KKR + GIP-owned) hyperscale operator; overlapping US + LatAm ambitions.
AI neocloud building/leasing dense GPU capacity; both a competitor for power/sites and a demand indicator.
Private (DigitalBridge-owned) hyperscale/renewable-focused operator.
Private hyperscale developers competing for the same powered land and hyperscale leases.
AI-native infrastructure builders converting power into GPU capacity; emerging competitors for sites and demand.