Digital Infrastructure Real Estate

State Policy Controls (September 2026)

California's September 21 package of seven signed bills addresses large-load electricity rates and infrastructure costs, water information and supply, land-use review, and community disclosure. Texas's September 21 gubernatorial letter instead directs TCEQ to pause data-center permits pending ERCOT review and says other state approvals should wait for information sought by PUCT, ERCOT, and TWDB. These are different policy mechanisms and legal instruments: signed state laws in California and an executive permitting directive in Texas. The available records establish policy actions, not implementation outcomes, project cancellations, or market-capacity changes. Keep the Texas audit scope tied to ERCOT's interconnection process; do not read the California clean-energy summary as a universal on-site-generation mandate. See Source: Texas and California Data Center Policy Actions September 2026.

Loudoun County adds a local entitlement example: as of September 24, the Board was considering a resolution to pause final votes on legislative data-center and substation applications for up to a year, while allowing applications to be filed and continue through Planning Commission review. The October 20 resolution was pending at that date; this does not establish a countywide application moratorium or project cancellations. See Source: Loudoun County Data Center and Substation Application Pause Proposal 2026.

Source: Knight Frank Corporate Real Estate Balancing Act 2026 Article adds an occupier-side location screen: AI, defence, and green-industry demand can raise the premium for energy-secure, resilient sites, with on-site generation and storage becoming more relevant. Treat grid capacity, reliability, renewable procurement, and storage as property-level diligence gates; Knight Frank's “grid over rent” framing is a conditional strategic thesis, not proof of power availability or value.

Source: Colliers AI Infrastructure Spending on Track to Exceed $800 Billion provides a current corporate-demand signal: the public Colliers article cites Morgan Stanley's projection of more than $800 billion of 2026 investment by major cloud and technology providers and Bloomberg analysis of more than $850 billion in future cloud-provider lease obligations. Keep the claim at the cited corporate level; for asset underwriting, translate it into separate diligence on power, land, fiber, specialized facilities, tenant credit, and executed lease / delivery status.

Source: Green Street Data Centers and the AI Surge 2026 reinforces the demand-versus-deliverability split: AI-driven leasing can strengthen landlord leverage, but power constraints push activity into secondary and tertiary markets and make site, interconnection, supply, and execution evidence decisive. The webinar page supplies no capacity, rent, or valuation series.

JLL Power-Gated Supply Overlay

Source: JLL Global Real Estate Outlook Mid-Year 2026 reinforces that power availability, interconnection timing, grid congestion, and energy cost are gating variables for data centers, advanced manufacturing, and EV charging. Underwrite powered-land claims through utility capacity, interconnection, onsite generation/storage, and delivery sequencing; a broad AI-demand narrative is not enough to establish a viable digital-infrastructure site.

Source: Cushman & Wakefield Americas Data Center Update H1 2026 quantifies the regional demand-and-deliverability split: 50.3 GW operational, 37.7 GW under construction with 91.7% precommitted, and 285 GW in the planned pipeline across the Americas. C&W also reports 3.8% region-wide vacancy. Keep the definitions, Americas scope, and H1 2026 date attached; the planned pipeline is not committed supply, and this landing page does not publish current metro-level metrics for Houston, Austin/San Antonio, or West Texas. See Source: Connect CRE Cushman & Wakefield Power and Data Center Map 2026 for secondary coverage of the same report, not an independent dataset.

This concept captures the recurring pattern, now visible across both the investor and Texas geography branches, that some of the most important modern real estate is neither classic office nor generic industrial but the physical infrastructure that supports cloud computing, AI workloads, data transmission, and wireless connectivity.

Brookings' 2026 financing paper adds a capital-stack control to that physical screen. Its $10.3T AI infrastructure scenario includes IT systems, networking, buildings, and power, with roughly two-thirds of a representative campus's cost assigned to installed IT. More leases, JVs, project debt, private credit, and securitization can expand available capital while shifting leverage and contingent obligations beyond hyperscaler balance sheets. This raises diligence needs around tenant concentration, equipment obsolescence, cash flows, and financing visibility; it does not establish that current data-center exposure is already systemically risky or that a correction is imminent. See Source: Brookings Financing the AI Buildout 2026.

Source-Derived Pattern

Hut 8's Beacon Point campus in Nueces County shows the power-first greenfield model at campus scale: the sponsor reports 1,000 MW of utility capacity, 704 MW of contracted IT capacity, and two 15-year leases. Keep those measures and contractual status separate. They indicate a large, contracted development pipeline, not delivered or operating capacity. See Beacon Point Data Center Campus and Source: Hut 8 Beacon Point Data Center Campus 2026.

Across the current canonical sources, the same logic keeps repeating:

• the largest public-market winners are increasingly specialized infrastructure providers rather than generalist landlords
• data centers, interconnection hubs, towers, and related facilities form the physical layer of the digital world
• AI infrastructure demand increases the value of powered, fiber-connected, entitlement-ready sites
• fiber connectivity inside multifamily is now a core utility expectation, not just an amenity decision
• Texas corridors benefit when digital infrastructure demand overlaps with logistics, semiconductors, and advanced manufacturing rather than competing with them

MSCI's 2026 Real Assets in Focus paper adds an important caution to that pattern. The report says AI training workloads are now pushing densities above 100 kW per rack, while inference workloads often need 20 to 50 kW per rack, and that these higher specifications are increasing both development cost and obsolescence risk. The lesson is that data-center demand is not a free pass. Viability increasingly depends on micro-location, power, cooling, connectivity, and whether older assets can be upgraded without effectively rebuilding the shell.

Source: CBRE North America Data Center Trends H1 2025 adds the first-half 2025 market-wide baseline. CBRE's applied rows show 8,155 MW of primary-market supply, 1.6% vacancy, 5,242.5 MW under construction, and 74.3% of under-construction capacity committed, while pricing split by requirement size: 250-to-500-kW requirements rose 2.5% and large 10 MW-plus deployments rose up to 19%. The concept-level lesson is that digital infrastructure demand was already being priced through contiguous power scarcity before the H2 tightening.

Source: CBRE North America Data Center Trends H2 2025 adds the current market-level version of the same constraint. CBRE's applied rows show 1.4% primary-market vacancy, 2,497.6 MW of 2025 net absorption, 36% supply growth to 9,432 MW, and under-construction capacity falling to 5,994.4 MW at year-end 2025. That strengthens the concept's core rule: digital infrastructure real estate is valuable where power, fiber, cooling, permitting, capital, and customer demand clear together.

Source: CBRE Houston Data Center Market H1 2025 adds a Houston-specific energy-market example. The profile supports a digital-infrastructure lane tied to oil-and-gas occupier demand, behind-the-meter power studies, legacy-space reuse, and enterprise acquisitions for large-scale colocation development, but its 19.7% vacancy row also shows why digital infrastructure cannot be reduced to a universal scarcity thesis.

Source: CBRE Denver Data Center Market H1 2025 adds a smaller Front Range example. Denver's 20.1 MW across 27 online facilities, paired with only two facilities offering large amounts of power and space, shows how digital-infrastructure demand can be real but still highly facility-specific. The investable read is suburban utility access, retrofit feasibility, and policy support rather than broad metro-wide capacity depth.

Source: CBRE Austin-San Antonio Data Center Market H1 2025 adds the Central Texas corridor version: low 2.3% vacancy, 97% preleased under-construction capacity, and 462.0 MW under construction. The digital-infrastructure read is service-territory and power-delivery competition across Oncor-served northeast Austin suburbs and activity moving south of Austin, not a conventional metro-average real estate call.

Source: CBRE Atlanta Data Center Market H1 2025 adds the utility-screening version. Atlanta's H1 profile pairs neo-cloud / AI demand with stricter Georgia Power requirements for end-user commitments, zoning confirmation, and collateral, plus over 8 GW of generation capacity scheduled after 2030 and over 300 MW of hyperscale preleasing. Digital infrastructure here is valuable only where customer commitment and power procurement survive the utility screen.

Source: CBRE Chicago Data Center Market H1 2025 adds the mature-market leasing and approval-pressure version. The source reports a 72 MW AI-related lease at Prime Data Centers and Yorkville projects seeking approval for 38 buildings and 3+ GW of power, while also emphasizing ComEd delays and credit support. Digital infrastructure in Chicago is therefore a tenant-demand story only when utility timing and balance-sheet support are solved.

Source: CBRE Dallas-Ft. Worth Data Center Market H1 2025 adds the Texas scale-market setup. DFW's H1 profile shows 869.5 MW of existing inventory, 425.1 MW under construction, and 78% preleasing on that pipeline, while also tying demand to hyperscalers / AI providers, Oncor transmission planning, rising site pricing, utility requests, and behind-the-meter power studies. Digital infrastructure in DFW is therefore a power-delivery and preleasing problem, not just a land-availability story.

Source: CBRE Seattle Data Center Market H1 2025 adds the urban network / dense-power version. Seattle's profile emphasizes AI, enterprise, and hyperscaler colocation demand, fast-track occupancy, liquid cooling, generator options, SEA10 demand of up to 1 MW, and Sabey's scheduled 6 MW SDC Building 5 expansion. Digital infrastructure here should be read as power density and connectivity inside Puget Sound, not as a substitute for Central Washington powered-land scale.

Source: CBRE Central Washington Data Center Market H1 2025 adds the powered-land scale complement. The profile's planned Tri-Cities hyperscaler campus rows ($4.8B, 16 data centers, 215,000 SF each) show why Central Washington should be read as a power, land, and private-energy execution market. The digital-infrastructure question is whether PUD upgrades, private power, alternative generation, microgrids, and interim generators can convert planned campus scale into deliverable capacity.

Source: CBRE Charlotte-Raleigh Data Center Market H1 2025 adds the Carolinas service-territory example. The source pairs a small 1 MW Raleigh delivery with much larger announced / conversion scale in Madison, Gaffney, and Hamlet, while Duke Energy requirements push developers toward tenant commitments and higher capital contributions. Digital infrastructure here is valuable only where power-delivery timing, customer commitment, and utility-territory edge conditions clear together.

Source: CBRE Southern California Data Center Market H1 2025 adds the Los Angeles carrier-hotel example. One Wilshire's over-10 MW available shell-capacity signal shows that digital infrastructure value can sit in interconnection-rich infill buildings, not only in greenfield hyperscale campuses. The market still needs power-cost, vacancy, AI-demand, and Vernon-development diligence before availability becomes underwritable capacity.

Source: CBRE Montreal Data Center Market H1 2025 adds the Quebec hydro / power-allocation example. QScale and Enovum show AI-ready capacity can be delivered where operators already have power access, but CBRE's Hydro-Quebec restriction and contiguous-block scarcity language keeps the lesson disciplined: low-cost renewable power is not the same as freely available large-load capacity.

Source: CBRE Toronto Data Center Market H1 2025 adds the GTA power-and-security example. CoreWeave's preleasing, Microsoft's campus-expansion load, Hive's conditional repurposing capacity, and the 2027-2028 greenfield range show digital infrastructure shifting from latency-only interconnection logic toward power availability, site security, and utility-study execution.

Source: CBRE Hillsboro Data Center Market H1 2025 adds the Pacific Northwest expansion-with-scarcity example. NTT's 216 MW additional planned capacity toward 354 MW total planned capacity shows platform-scale growth, while CBRE's sub-500 kW / larger-requirement vacancy split keeps the digital-infrastructure thesis grounded in actual operator availability and utility cost allocation.

Source: CBRE New York Tri-State Data Center Market H1 2025 adds the low-latency legacy-market example. No greater-than-5 MW capacity was available in CBRE's H1 text, AI occupiers were preleasing in 5-to-10-MW increments where available, and Equinix's NY3 project represented its 10th Tri-State data center. Digital infrastructure here is a power and contiguous-capacity constraint wrapped around valuable financial-services latency demand and existing operator campuses.

Source: CBRE Northern Virginia Data Center Market H1 2025 adds the mature-core scarcity baseline. The market had only 0.72% colocation vacancy in H1 2025, with 15-to-20-MW requirements difficult to secure and preleasing extending into 2027-plus capacity. Digital infrastructure here is not demand validation; it is the discipline of converting land, power delivery, entitlement, and preleased capacity into actual usable megawatts.

Source: CBRE Silicon Valley Data Center Market H1 2025 adds the Bay Area innovation-cluster scarcity example. Local technology demand, 4.5% vacancy, over-50 MW available-or-under-construction capacity, and AI-company / Vantage preleasing show why Silicon Valley remains strategically valuable, but CBRE's power-constraint language means the digital-infrastructure thesis must be filtered through utility capacity rather than innovation adjacency alone.

Source: JLL 2026 Global Data Center Outlook adds the global capacity and cost frame for that rule. JLL forecasts 97 GW of capacity additions between 2026 and 2030 and says global capacity could reach 200 GW by 2030, while also reporting 97% global occupancy, a 33-week average equipment lead time, and $11.3M/MW forecast shell-and-core construction cost for 2026. The source strengthens digital infrastructure as a durable capital lane, but it also makes execution risk explicit: power, cooling, equipment, and construction delivery are now part of the asset thesis rather than back-office engineering details.

Source: Avison Young Q1 2026 U.S. Data Center Market Overview adds a current U.S. colocation operating check. Avison Young reports Q1 2026 inventory at 29.0 GW, net absorption at 5.29 GW, and vacancy at 1.2%, with most new supply pre-leased ahead of delivery. The implication for this concept is that digital infrastructure demand is translating into physical-capacity scarcity even while supply grows rapidly; the bottleneck is deliverable powered capacity, not simply tenant interest.

Source: Newmark 2026 U.S. Data Center Market Outlook adds a Newmark headline check on the same capacity-scarcity theme. Its public page reports 2% U.S. data-center vacancy and a 160 GW pipeline of facilities under construction or announced. That combination reinforces the concept's main rule: digital-infrastructure demand is not capacity by itself. The underwritable asset is deliverable powered capacity with a viable grid, entitlement, water/cooling, financing, and tenant-credit path.

Source: Cushman & Wakefield Data Centres: EMEA Growth Is Robust But Being Reshaped adds the regional divergence version of that rule. C&W says EMEA demand remains strong, but its ability to convert demand into data-centre capacity is restricted by power availability and pricing, permitting, supply chain, and regulatory complexity. The practical read for digital infrastructure is that AI demand is global, while real-estate investability is local and execution-bound.

Source: Cushman & Wakefield APAC Data Centre Update H2 2025 adds the APAC scale counterpart. C&W reports 13,763 MW of APAC operational capacity and a 19,371 MW pipeline as of H2 2025, with Malaysia and India accounting for 58% of new operational capacity. The digital-infrastructure read is that APAC has moved into large-scale execution, but the investable version still depends on power availability, regulation, geopolitics, and whether planned capacity can become live supply.

Source: Cushman & Wakefield From Megawatts to Meaning May 2026 adds the KPI version of that same execution rule. C&W's APAC data-centre insights page says AI data centres are increasingly viewed as AI factories that convert energy into intelligence, with cost per token emerging alongside megawatts and cost per kilowatt. For digital-infrastructure real estate, that means the asset thesis has to connect the physical stack to workload output; generic power capacity is not enough if hardware, cooling, utilization, or workload economics underperform.

Source: Cushman & Wakefield Artificial Intelligence and the Next Chapter for Australian Commercial Real Estate adds a scenario layer beyond physical data-centre demand. C&W models four AI diffusion paths across 19 industries and Australian cities, finding that logistics and industrial have the highest downside floor while office outcomes diverge sharply by labor substitution, productivity, development timing, and quality. Use the source to widen city-, sector-, and asset-grade sensitivity cases; it is not a point forecast or a substitute for observed leasing, employment, power, or capacity evidence.

Source: CBRE Q2 2026 U.S. Distributed Generation Opportunities adds a utility-program layer to the digital-infrastructure stack. CBRE frames distributed batteries and fuel cells as assets whose economics can come from state incentives, capacity contracts, utility rebates, performance payments, and emissions-offset rules. That makes digital infrastructure partly a policy-and-tariff diligence problem: the same physical battery can be backup power, an FTM lease asset, a BTM bill-savings tool, or an offsite emissions-offset instrument depending on jurisdiction and contract structure.

Source: JLL Get set for the 5th Industrial Revolution 2026 adds the AI-location version of the same infrastructure rule. JLL separates large data-center facilities that follow space and power from edge facilities that need proximity to users for low-latency applications, and it elevates power availability, water, clean energy, power security, climate resilience, and government support as site-selection variables for AI, chip manufacturing, and advanced-computing facilities. Use it as qualitative demand-shape support, not as capacity or vacancy evidence.

Source: Green Street Launches Green Street Infrastructure and Source: Green Street Infrastructure Research Now Available add provider-market validation that digital infrastructure is now tracked as part of a broader infrastructure research lane, not only as a CRE niche. The public launch release says Green Street Infrastructure combines IJGlobal infrastructure data / news with Green Street research and analytics, and the later research release says the suite covers data centers, towers, airports, railroads, toll roads, health care, senior housing, student housing, cold storage, contractors, and infrastructure debt analytics. Use them as coverage-map sources, not as capacity, price, lending-spread, or site-level data.

The Fisk University Quantum Leap announcement is a boundary example: a campus "data and technology center" may matter for institutional real estate and education infrastructure, but it should not be treated as commercial data-center capacity without clearer technical and operating evidence. See Source: $1B Fisk Upgrade Includes On-Campus Data Center.

Prime Data Centers' SMF02 announcement is the opposite kind of boundary item: it is a commercial data-center milestone, but it remains announcement-stage until power, permits, and delivery evidence are preserved. See Source: Prime Data Centers Breaks Ground on Second Sacramento Facility.

The Cox & Castle legal Q&A adds a state-and-local friction layer to the digital-infrastructure thesis: backup generation, environmental review, interconnection upgrades, cooling-water design, and community pushback can matter as much as fiber and land control. This is especially important in California, where the article frames generator and CEQA review as potential gating items. See Source: The Legal Side of Data Center Development - Q&A with Cox & Castle's Margo Bradish and Arielle Harris.

Patmos' former Kansas City Star production-facility retrofit is a mixed-use digital-infrastructure case. January-reported C-PACE finance and unnamed leases do not establish all-project funding, tenant credit or phase commissioning. Keep the next-phase power forecast separate from earlier operator capacity claims through Energy Infrastructure and CRE and Data Center Underwriting and Powered Land. Use Adaptive Reuse of Obsolete Office as an adjacent comparison method and CRE Capital Markets for the actual debt/assessment stack. See Source: Patmos Receives $100M C-PACE Loan for Redevelopment of Kansas City Star Building into Data Center.

Corning's Hickory optical-cable expansion tied to Meta's data-center buildout is a supply-chain example rather than a data-center asset. It shows how AI infrastructure demand can create manufacturing and fiber-connectivity real estate demand in inland industrial markets. See Source: Corning and Meta Announce Hickory Optical Cable Manufacturing Expansion.

Nvidia / Corning's May 2026 three-plant announcement broadens that same supply-chain pattern beyond Hickory: data-center demand now pulls optical connectivity and fiber-production capacity into U.S. manufacturing decisions. The source does not identify exact facility locations, so treat it as national supply-chain evidence rather than a site-level real estate comp. See Source: Nvidia Corning US Data Center Manufacturing Plants 2026.

Bisnow's May 2026 industrial supply-chain synthesis reinforces that the digital-infrastructure branch is creating non-data-center industrial demand as well: switchgear, cooling systems, optical connectivity, prefabricated components, and ongoing data-center operations all require manufacturing and logistics space. Treat the broker-cited leasing figures as source-reported until the underlying datasets are preserved. See Source: Data Centers Are Becoming Industrial's Growth Engine 2026.

Celestica's AllianceTexas campus is the local DFW version of that supply-chain pattern: a planned advanced manufacturing and engineering campus tied to data-center infrastructure products, not a data-center facility itself. See Source: Celestica AllianceTexas Manufacturing Campus 2026.

The parallel power-shortage source adds the constraint side of the same pattern. Industrial automation, EV charging, robotics, high-power manufacturing, and data centers are all competing for grid capacity, making existing heavy power service and credible energization timelines part of digital-infrastructure real estate value. See Source: Industrial Automation Power Shortages Data Center Competition 2026.

The JLL-authored St. Louis data-center commentary supplies a legacy-site reuse hypothesis. Price, transmission, water, fiber and policy arguments do not establish actual service, tenants or investable advantage. Use St Louis Industrial and Logistics Market to keep the cross-state industrial branch separate from Missouri utility scope, Energy Infrastructure and CRE for contracts and rights, and Data Center Underwriting and Powered Land for phase delivery. Keep tariff-summary clarification and historical price uncertainty in Source: Why Data Centers Are Emerging as a Defining Opportunity for the St. Louis Region.

Metrobloks / Lincoln's April 2026 Liberty, Missouri announcement is a Northland campus watchlist case, separate from downtown Patmos reuse. The approximately 30-acre / 568,000 SF plan and initial 150 MW total utility power are announced scopes, not critical IT capacity, commissioned service or proof of tenants. Use KCI Airport and Northland Logistics Corridor, Data Center Underwriting and Powered Land and Energy Infrastructure and CRE for geography, phases and contracts. Keep developer readiness and later city plan scope distinct in Source: Metrobloks and Lincoln Property Co. to Develop 568,000 SF Data Center Campus in Metro Kansas City.

Batch 72 adds a national power-stack layer. DOE nuclear financing, CIM's rural QOZ power-platform launch, and battery-storage operations all point to the same conclusion: digital infrastructure real estate is increasingly underwritten through generation, storage, utility structure, and equipment-level resilience rather than only through the data-center shell. See Source: DOE To Finance Parts for 10 Nuclear Reactors to Power Data Centers, Source: CIM Group Launches Platform To Invest In Rural QOZ Power Projects, and Source: Battery Storage in Data Centers.

The 2027 Opportunity Zone round adds tax-policy exposure to rural data-center underwriting. Enacted rural incentives coexist with Hawley's introduced bill and Wyden's staff discussion proposal to curtail or change data-center tax treatment; neither proposal is enacted. Tract nomination, final designation, qualifying structure, and actual project eligibility remain separate checks. See Source: Data Centers and Opportunity Zone Tax Policy 2026.

DataBank's Red Oak construction loan brings that infrastructure pattern back to DFW: the relevant real estate fact is not only that a campus was announced, but that a leased 180 MW phase reportedly secured construction financing. See Source: DataBank Secures $2B Construction Loan for Red Oak Data Centers.

Gradiant's Series E financing is not a real-estate asset comp, but it adds a supporting infrastructure-services signal: AI infrastructure and semiconductor growth are also financing water-treatment and advanced-water platforms. Cushman & Wakefield's AI summary adds the strategy guardrail that the real estate impact will be uneven by asset quality, sector, and timing rather than a uniform lift. See Source: Gradiant Closes on Series E Financing to Drive Advanced Water Solutions and Source: Digging Beneath the Confusion: The Future of Artificial Intelligence and Commercial Real Estate.

The June 2026 RSS verification pass promoted the May 18-19 digital-infrastructure batch to reviewed. Use it to support the branch's infrastructure-stack thesis, not as a property dataset: generation, storage, utilities, water treatment, data-center finance, and developer pipelines are increasingly part of the same real-estate screen, but exact capacity, cost, tenant, timing, and ranking claims remain source-attributed until primary artifacts are preserved.

Batch 75 adds two opposite ends of the digital-infrastructure spectrum. Google's reported orbital-data-center work with SpaceX is a frontier watchlist item: it is an extreme response to power and cooling constraints, not current terrestrial capacity. Blackstone's BXDC debut is the institutional capital-market expression: a dedicated data-center REIT vehicle seeking newly built hyperscaler-occupied assets and liquidity for developers or operators. Together they show the branch stretching both upstream into physics/power constraints and downstream into specialized capital vehicles. See Source: Google, SpaceX In Talks To Launch Orbital Data Centers As Space Race Accelerates and Source: Blackstone BXDC Data Center REIT Debut Raises $1.75B.

Source: JLL Data Centers in Space 2026 refines the orbital side of that spectrum. JLL's useful CRE point is workload sorting: orbital systems may fit asynchronous, energy-heavy jobs if launch economics, operational validation, debris management, and terrestrial power bottlenecks clear, while real-time inference and transaction processing should stay close to users and networks. That makes orbital compute a digital-infrastructure boundary condition and power-scarcity signal, not a substitute for terrestrial data-center underwriting.

Source: JLL The Future of Quantum Real Estate 2026 adds the quantum-computing version of the same long-range boundary. JLL frames near-term quantum development around ecosystems with academic institutions, existing quantum facilities, government support, and private-sector depth, with later movement toward pilot deployments and hybrid quantum-classical data centers. Treat this as a first-mover diligence map for specialized facilities, cryogenic cooling, electromagnetic shielding, and quantum-as-a-service adoption, not as present-day capacity or leasing evidence.

Helix Digital Infrastructure’s capital-formation record now includes the June launch commitments of more than $10B and Samsung’s later additional $1B commitment. Samsung reports six participating affiliates, including a $500M Samsung Electronics commitment; KKR describes the funding as building on the launch capital. Treat the resulting more-than-$11B platform commitment as issuer-reported capital, distinct from deployment, sites, power, or operating capacity. See Source: KKR NVIDIA Kuwait Helix AI Infrastructure Platform 2026 and Source: Samsung $1B Commitment to Helix Digital Infrastructure 2026. The wider capital spectrum also includes DigitalBridge / ArcLight platform formation, Switch’s expanded credit facilities, and hyperscaler site control in San Jose and Goodyear. See Source: DigitalBridge ArcLight Power Infrastructure Acquisition 2026, Source: Switch Credit Facilities Nearly 10B 2026, Source: Microsoft North San Jose Data Center Groundbreak 2026, and Source: Microsoft Goodyear Data Center Land 2026.

Broadcom's AI XPV platform with Apollo and Blackstone adds a hardware / credit-capital version of the same infrastructure-stack thesis. The source is not a delivered data-center record, but it shows compute-capacity finance forming around XPUs, networking, frontier-lab demand, and private-credit / insurance capital. See Source: Broadcom Apollo Blackstone AI Infrastructure Platform 2026.

Source: Cushman & Wakefield AI, Bonds & CRE 2026 adds the public-bond-market version of that capital stack. C&W reports 2025 hyperscaler / AI infrastructure corporate issuance of $121B and projects 2026 AI-related investment-grade issuance of $200B to $400B, with cumulative hyperscaler bond supply potentially reaching $3.5T by 2030. That reinforces digital infrastructure as a capital-intensive real-assets lane, but it also adds a constraint: AI infrastructure can compete with CRE credit for the same fixed-income buyer base and long-duration capital.

Waymo's Wittmann test-track acquisition extends the concept beyond data centers and telecom. Autonomous-vehicle systems need controlled proving grounds, simulation-linked physical testing, sensor calibration, safety validation, and operating infrastructure. Treat the Arizona site as AI / mobility infrastructure evidence, not as a data-center or warehouse comp. See Source: Waymo Wittmann Test Track 2026.

Robotaxi operating depots extend that physical layer from specialized proving grounds into infill industrial facilities. Bisnow reports fleet-service and staging projects that need space for vehicle storage, charging, maintenance, and quick turnarounds, including a Lyft / Flexdrive 80,000-SF Nashville depot corroborated by Lyft. The wider 2026 leasing total is attributed to Cushman & Wakefield data but lacks a retained dataset; it is an emerging tenant-use signal, not a market-wide industrial demand forecast. See Source: Robotaxi Firms Grow Industrial Footprint 2026.

Why It Matters

This concept connects:

• REIT Landscape because the investor source explicitly frames digital infrastructure as one of the main specialized public-market real estate winners
• American Tower and Equinix because they are the clearest benchmark entities for the wireless and data-center sides of the theme
• Powered Land and Grid Advantage because digital infrastructure only works where power, fiber, and site readiness line up
• Conviction Theme Investing because capital increasingly rotates toward infrastructure-like sectors with secular demand rather than broad commodity property exposure
• Texas because the state thesis explicitly treats AI infrastructure and data-center demand as major drivers of future real estate advantage
• Multifamily Hub and Multifamily Bundled Internet and Managed Wi-Fi because fiber and resident connectivity have become operating and investment variables in housing, not just digital-infrastructure add-ons
• Multifamily Ancillary Income Programs because smart access, package lockers, EV charging, and other resident-facing technology can be ancillary-income programs only when their operating costs and fee-disclosure risks are underwritten
• Industrial Hub because the industrial branch now repeatedly overlaps with semiconductors, AI infrastructure, and infrastructure-led site selection
• Alliance and North Fort Worth, East Austin Tesla and Airport Corridor, Williamson County Semiconductor Corridor, and Plano Richardson Telecom Corridor because each corridor expresses a different version of the same digital-physical demand pattern
• Northern Virginia Digital Infrastructure Corridor because the Washington-facing contractor and cloud-services layer connects federal demand directly to the largest U.S. data-center corridor

It also matters because the digital-infrastructure branch now looks more like a capital-intensity and depreciation problem than a generic "tech demand" story. If newer AI-ready facilities require materially heavier electrical and cooling systems, owners may face the same kind of obsolescence-and-recapitalization cycle that commodity office owners faced after tenant expectations changed.

Related Pages

• Energy Infrastructure and CRE

• Investor Concepts Hub

• Investors and Platforms Hub

• CRE Investor Landscape

• REIT Landscape

• American Tower

• Equinix

• Data Center Underwriting and Powered Land

• Powered Land and Grid Advantage

• Conviction Theme Investing

• Texas

• Multifamily Hub

• Multifamily Ancillary Income Programs

• Multifamily Bundled Internet and Managed Wi-Fi

• Industrial Hub

• Alliance and North Fort Worth

• East Austin Tesla and Airport Corridor

• Williamson County Semiconductor Corridor

• Plano Richardson Telecom Corridor

• Texas AI and Industrial Infrastructure Opportunity Map

• Northern Virginia and Washington DC

• Northern Virginia Digital Infrastructure Corridor

Source: Knight Frank Data Centres Global Forecast Report 2026 frames the 2026 digital infrastructure cycle as both a demand expansion and a delivery-selection problem. Its 33 GW two-year global capacity forecast and regional growth table support screening for power-rich secondary markets, while its discussion of AI density, 1–10 MW inference facilities, neo-cloud credit, obsolescence, and platform financing keeps the investment thesis conditional on technical adaptability and capital execution.

AI and Office Space Intersection

Newmark’s April 2026 report on AI and office demand confirms that AI is reshaping the physical-space conversation beyond data centers and industrial. AI is also an office-demand variable: 88% of firms report using AI, but the impact on space is nuanced. Near-term displacement concentrates in automatable back-office roles, while quality office near agglomeration economies becomes a "fortress." This connects digital infrastructure to the office branch — AI demand is not confined to industrial shells and power infrastructure but extends to the quality and location of knowledge-work space.

Source: Cushman & Wakefield AI Impact Barometer supplies a reusable cross-sector monitoring frame: C&W describes positive AI momentum in data centers and highly automatable industrial facilities, while office, retail, and multifamily remain mixed and differentiated by quality, location, and technology exposure. For digital-infrastructure underwriting, that means pairing compute and power demand with the labor, tenant, and location signals that determine whether AI-linked demand reaches adjacent real estate. The barometer is directional methodology, not a power-availability, lease-up, or valuation dataset.

Source: Cushman & Wakefield AI Impact on Commercial Real Estate adds timing and spillover discipline. C&W's framework treats data-center and automated-industrial demand as part of a broader AI transmission chain, but reserves direct labor-displacement effects mainly for office and expects industrial, retail, and living impacts to arrive through productivity, income, operating efficiency, and capital investment. Model compute demand, power delivery, automation capex, and tenant economics as linked but separate underwriting cases.

Source: MSCI State of Private Markets 2026 supplies a capital-formation caveat for the AI infrastructure thesis: MSCI reports AI-related assets at roughly 16% of global private equity and says the physical infrastructure build-out is still developing, while highlighting the need to measure portfolio exposure and liquidity. Treat private-capital appetite as a funding-context signal; validate actual data-center basis, power, tenant, construction, and exit evidence separately.

Government AI Infrastructure Expansion (2026)

Two Amazon announcements confirm that AI infrastructure demand is not confined to commercial hyperscalers — it extends deeply into the government cloud and creates a distinct real estate demand segment with different site-selection criteria (security clearance access, SCIF capability, proximity to federal agencies).

Amazon $50B for U.S. Government AI — November 2025 announcement: The retained report describes up to $50B of planned investment and 1.3 GW of planned capacity, with groundbreaking intended to begin in 2026. Its 11,000-plus agencies are AWS's existing government customer base, not new federal leases or commitments to this capacity. Government classification and procurement requirements are diligence questions; the report does not establish delivered supply or parcel locations. Use National Digital Infrastructure Capital Deployment 2026 for program scope and Data Center Underwriting and Powered Land for customer, financing and power-delivery gates. Northern Virginia Digital Infrastructure Corridor provides comparison context, not a confirmed program allocation. See Source: Amazon to Invest Up to $50B to Expand AI Infrastructure for U.S. Government.

Amazon $15B / 2.4 GW Indiana with NIPSCO Utility Integration (2026): The retained November 26, 2025 report describes a planned $15B expansion and an Amazon/NIPSCO power arrangement. It does not establish Amazon ownership of GenCo, completed power delivery or measured ratepayer protection. Keep the reported investment and capacity plans separate from asset ownership and service performance. Use Energy Infrastructure and CRE to verify the actual counterparties, corporate structure, contracts, approvals, upgrade costs and service dates. The arrangement is a project-specific diligence case, not proof that all hyperscalers own utility infrastructure or that this structure matches Google's Texas model. See Source: Amazon Plans to Invest $15B to Build New Data Center Campuses in Northern Indiana.

Adaptive Reuse for AI-Ready Data Centers (2026)

The retained architect commentary describes potential reuse of vacant commercial buildings and retrofit of older data centers. It supports building-specific questions about structural capacity, layout, cooling, fiber and power, not a universal entitlement shortcut or AI specification. The inherited ~30–50 kW/rack range is unverified in this source review; existing commercial use alone does not establish permission for the proposed data-center use. See Source: What Developers Should Know About Repositioning Commercial Properties into AI-Ready Data Centers.

Use Adaptive Reuse of Obsolete Office to compare reuse, replacement and other disposition paths. Data Center Underwriting and Powered Land supplies the workload, capacity and financial questions; Energy Infrastructure and CRE separates physical capability, service rights, continuity and economics. Verify actual use rights, design, utility work, tenant needs, costs and delivery dates. Cost, timing and embodied-carbon benefits remain hypotheses until matched alternatives and the property record support them.

AI-Driven Office Demand: NYC Leasing Evidence (Q1 2026)

Connect CRE's April 10, 2026 article attributes faster NYC AI leasing and an average-size increase from 16,600 to 34,500 SF to JLL. Broker commentary describes leasing ahead of expected hiring and demand for flexible, reconfigurable space. Nscale's reported $320/SF lease and a separate family-office lease at $327.50/SF are individual transactions; the family office is not identified as an AI tenant. Interpretation: office leasing is a cohort-specific channel alongside compute infrastructure, but the article does not establish market-wide premium demand, realized employment, comparable rent economics or effects on commodity stock. See Source: AI Tenants Double Their NYC Leasing Activity Year Over Year.

Hyperscaler Texas Commitments (2025–2026)

Three announcements confirm that institutional capital is now treating Texas-based powered land as a priority infrastructure asset at historic scale, and that digital infrastructure asset management is itself a consolidation target:

SoftBank / DigitalBridge agreement and completion: The December 29, 2025 agreement reported approximately $4B enterprise value and $16/share cash consideration, with a 50% premium to the unaffected 52-week average closing price as of December 4. SoftBank’s September 30, 2026 release reports completion of the common-stock acquisition for approximately $3.1B, with DigitalBridge a controlled subsidiary and separately managed under Marc Ganzi. Keep enterprise value, equity consideration, historical share-price premium and completion status distinct. This is platform ownership evidence, not a property rent/valuation benchmark, realized return or delivered AI capacity. See Source: SoftBank Agrees to Acquire DigitalBridge in $4B Deal and Source: SoftBank DigitalBridge Acquisition Completion 2026.

Google $40B Texas Data Centers (Nov 2025): Google announced ~$40 billion in Texas data center investment through 2027, including three new campuses in Armstrong and Haskell counties in West Texas (Haskell County co-located alongside a new solar and battery storage plant) and an expansion of existing Ellis County facilities south of Dallas. Google also committed $30M to an Energy Impact Fund and pledged to train 1,700+ electrical apprentices in Texas by 2030. The announcement identifies one Haskell campus alongside solar and storage; it does not establish generation ownership across the campuses or metro-grid congestion as the cause of site selection. See Source: Google to Invest $40B in Texas Data Centers Through 2027. Use Energy Infrastructure and CRE to distinguish planned co-location, ownership, contracted resources, parcel service and operating continuity. Assess actual delivery and cost obligations through Data Center Underwriting and Powered Land before applying a site or investment advantage.

Google / DTE Michigan (Mar 2026): Google's first-party Michigan announcement shows the same energy-first pattern through a utility agreement rather than a disclosed campus investment total. Google says 2.7 GW of new solar, advanced storage, and demand-flexibility resources will serve a proposed operation, alongside a $10M Energy Impact Fund and a commitment to cover electricity and infrastructure needs. The Van Buren Township site remained under evaluation, so the source belongs in the proposal-stage digital-infrastructure lane. The 2.7 GW figure is a resource portfolio, not stated IT load. See Michigan Data Centers and Powered Land and Source: Google Michigan Clean Energy and Reliability Announcement 2026.

BlackRock / Aligned Data Centers $40B Acquisition (Oct 2025): A consortium led by BlackRock’s Global Infrastructure Partners, Abu Dhabi’s MGX, and the Artificial Intelligence Infrastructure Partnership agreed to acquire Dallas-based Aligned Data Centers from Macquarie Asset Management for approximately $40 billion — among the largest data center transactions in history. Aligned had expanded from two original campuses in Dallas and Phoenix to 50+ campuses across the US, Mexico, Brazil, Chile, and Colombia, representing 5+ GW of operational and planned capacity. The deal’s scale confirms that data centers have crossed from real estate into infrastructure asset class territory, commanding infrastructure-fund capital at par with toll roads and utilities. See Source: BlackRock-Led Consortium Agrees to Acquire Aligned Data Centers for $40B.

MacroValor Mount Hydrogen: 3 GW AI + Robotics Mega Campus in Austin (April 2026)

Connect CRE's April 1, 2026 report describes MacroValor and Favis announcing an Austin AI, robotics and semiconductor campus with a proposed 3,000 MW scale. It reports initial site planning, partner evaluations and infrastructure modeling, without a parcel or groundbreaking date. Proposed natural-hydrogen supply and continuous, zero-emission operations are attributed announcement claims, not verified performance or delivered IT capacity. See Source: MacroValor, Favis Will Build Hydrogen-Powered AI Mega Campus.

Use the watch-list lane in National Digital Infrastructure Capital Deployment 2026. The inherited Texas-largest ranking and Google's ~1 GW comparison are unverified here; the article gives no matched capacity universe or comparative cost evidence. Use Energy Infrastructure and CRE for the production-to-operations interfaces and Powered Land and Grid Advantage for actual parcel, service and entitlement evidence. Test supply, delivery, storage, conversion, reliability, emissions scope, approvals, funding and tenant obligations before adopting a power or investment conclusion. The dated planning report does not establish the project's current status or capital commitment.

Modular AI Factory Manufacturing: Crusoe Spark Factory (March 2026)

The retained March 13 report describes Crusoe's planned 352,000 SF Brighton manufacturing space and a more-than-$200M commitment across leasing, build-out and an initial Spark module fleet. This is upstream industrial manufacturing, not operating data-hall capacity or a standalone construction price. First factory-produced modules were expected in Q3 2026; actual delivery is not established here. Healthcare and manufacturing applications are intended product uses, not confirmed orders.

Use Industrial Location and Tenant Demand Underwriting for manufacturing tenancy, premises and local demand. Use Energy Infrastructure and CRE and Data Center Underwriting and Powered Land for the units' deployment-site service and workload requirements. The announcement does not establish freight access or land availability as the reason for Brighton's selection, or a broad Denver industrial advantage. See Source: Crusoe Building $200M Denver-Area AI Factory.

West Virginia Powered-Land Watchlist

Penzance's March 4 report, captured May 18, describes a planned Bedington, Berkeley County campus: $4 billion, 600 MW, 548 acres and 1.9 million square feet at full buildout, with first capacity targeted for late 2028. Source: Penzance to Develop $4B Data Center Campus in Bedington, West Virginia records the separate dated state designation account without attesting complete legal terms, permits or current capacity. Use Data Center Underwriting and Powered Land, Energy Infrastructure and CRE, CRE Tax and After-Tax Underwriting and CRE Capital Markets for load boundaries, phase delivery, service obligations, tax eligibility and funding. Keep this West Virginia project outside the core Northern Virginia market and the separate Chantilly project.

Amazon's March 4 reported purchase of GW’s Ashburn campus is an acquisition and planned institutional-campus redevelopment case, not delivered data-center supply or measured market demand. Source: George Washington University Sells Virginia Science and Technology Campus to Amazon for $427M separates the dated seller accounts, reported price/acreage and continued-program option from project approvals and service. Use Northern Virginia Digital Infrastructure Corridor, Powered Land and Grid Advantage, Data Center Underwriting and Powered Land, Energy Infrastructure and CRE, Adaptive Reuse of Obsolete Office, CRE Capital Markets and CRE Valuation Methodologies for site, transition, reuse, funding and value comparisons.

Amazon’s November 11, 2025 Bristow acquisition report, captured May 18, 2026, is a Prince William undeveloped-land case, distinct from GW’s Ashburn institutional campus and the AWS government program. Source: Amazon Acquires Data Center Site in Northern Virginia for $700M preserves the secondary price, acreage and legal chain without inferring site-choice causality, current approval or capacity. Use Northern Virginia Digital Infrastructure Corridor, Powered Land and Grid Advantage, Data Center Underwriting and Powered Land, Energy Infrastructure and CRE, CRE Capital Markets and CRE Valuation Methodologies for rights, actual service, funding and sale scope.

The PW Digital Gateway appeal sequence adds a governance boundary case: QTS appealed to preserve its Prince William data-center plan while Compass chose to step away. The broader lesson is that data-center platforms are underwriting local legal process and community acceptance as real development inputs, not side issues. See Source: QTS Files Appeal To Save PW Digital Gateway 2026.

The NoVA-vs-Texas investor source adds a capital-market distinction: future AI-campus growth can shift toward power-rich markets like Texas while Northern Virginia remains the preferred institutional market for existing data-center exposure because of scarcity, interconnection, liquidity, and rent depth. See Source: Texas Might Soon Have More Data Centers, But Investors Still Prefer Northern Virginia.

Beale Infrastructure's April 2026 Project Pilot report is a De Soto / western Johnson County phased-campus case. The $3B plan and utility/incentive/unnamed-tenant claims are different evidence from spending, actual agreements and commissioned service. Use Olathe and Western Johnson County Growth Corridor, Data Center Underwriting and Powered Land, Energy Infrastructure and CRE, CRE Tax and After-Tax Underwriting and CRE Capital Markets for Kansas-side geography, current plan versions, phase obligations and funding. Keep sponsor "first hyperscale" framing in Source: Beale Infrastructure Unveils Plans for $3B Data Center Campus in De Soto, Kansas.

Vantage's November 2025 VA4 announcement is a Stafford / Fredericksburg project case outside the mature Northern Virginia core. The May 2026 capture is custody, not a new announcement. Keep campus capacity, first-building timing and cooling design separate from usable service, operating performance and committed tenancy. Use Data Center Underwriting and Powered Land and Energy Infrastructure and CRE for the delivery and design boundaries.

AI's CRE Application: CBRE Institutional Framing (April 2026)

CBRE’s February 2, 2026 episode with Sandy Pentland and Sandeep Davé discusses data quality, business-focused measurement and human judgment in AI use. Davé recommends relevant business KPIs and notes that personal-productivity gains are harder to measure; Pentland stresses context and imperfect human/AI decisions. These are attributed views, not measured asset-value gains, proof of large-platform superiority or an industry-wide consensus. See Source: Digital Love: AI & the Future of CRE (CBRE Weekly Take).

Sources

• Legacy CRE Investor Knowledgebase
• Legacy Texas Market Thesis
• Source: Everything Is Bigger in Texas, So Why Isn’t Our Fiber Internet?
• Source: Understanding the AI-Office Space Connection
• Source: AI Tenants Double Their NYC Leasing Activity Year Over Year
• Source: Google to Invest $40B in Texas Data Centers Through 2027
• Source: BlackRock-Led Consortium Agrees to Acquire Aligned Data Centers for $40B
• Source: SoftBank Agrees to Acquire DigitalBridge in $4B Deal
• Source: Amazon to Invest Up to $50B to Expand AI Infrastructure for U.S. Government
• Source: Amazon Plans to Invest $15B to Build New Data Center Campuses in Northern Indiana
• Source: What Developers Should Know About Repositioning Commercial Properties into AI-Ready Data Centers
• Source: MacroValor, Favis Will Build Hydrogen-Powered AI Mega Campus
• Source: Crusoe Building $200M Denver-Area AI Factory
• Source: Digital Love: AI & the Future of CRE (CBRE Weekly Take)
• Source: MSCI Real Assets in Focus Trends to Watch for 2026
• Source: Beale Infrastructure Unveils Plans for $3B Data Center Campus in De Soto, Kansas
• Source: Vantage Data Centers to Invest $2B for New Campus in Northern Virginia
• Source: DOE To Finance Parts for 10 Nuclear Reactors to Power Data Centers
• Source: CIM Group Launches Platform To Invest In Rural QOZ Power Projects
• Source: Battery Storage in Data Centers
• Source: JLL Get set for the 5th Industrial Revolution 2026
• Source: CBRE Seattle Data Center Market H1 2025
• Source: CBRE Central Washington Data Center Market H1 2025
• Source: CBRE Charlotte-Raleigh Data Center Market H1 2025
• Source: CBRE Southern California Data Center Market H1 2025
• Source: CBRE Montreal Data Center Market H1 2025
• Source: CBRE Toronto Data Center Market H1 2025
• Source: CBRE Hillsboro Data Center Market H1 2025
• Source: CBRE Northern Virginia Data Center Market H1 2025
• Source: CBRE Silicon Valley Data Center Market H1 2025

Source: CBRE Fortune Will Favor the Bold adds a forecast-level power constraint: CBRE expected record U.S. data-center leasing and mid-70% preleasing for new construction, while noting that rapid delivery depends on available power and may favor less-regulated energy markets. Treat this as a dated scenario claim, not a project-level absorption or interconnection assumption.

Source: Green Street News Infrastructure H1 2026 Funds and Investor Reports Release identifies an infrastructure-finance evidence lane spanning adviser league tables, fundraising, transactions, and regional rankings. Its reported platform scale is a custody and discovery fact, not evidence of project economics; request the underlying report or independent deal data before changing a power, capital-stack, or data-center underwriting assumption.

Source: Green Street H1 2026 Funds and Investors Reports provides the companion public landing page for the gated report products. It separates a fundraising/transaction report from a global and regional league-table report. Use that separation when building the infrastructure evidence stack: fundraising, transaction volume, adviser or investor rankings, and project economics are distinct inputs and must not be substituted for one another.

Source: Knight Frank Active Capital Top Themes 2026 places AI among the survey's leading investment themes (30%) and links it to data centres, energy infrastructure, and operational real estate. This is a capital-intent signal that supports screening for power, interconnection, tenant demand, and operating-platform capability; it does not establish data-centre demand, pricing, or a viable site without property-level evidence.

Source: The Real Deal Data Center Brokers and Power Real Estate February 2026 reinforces that digital-infrastructure real estate is an execution stack, not just a powered-land thesis. Screen power, fiber, substations, water, floodplain, zoning, financing tranches, prelease conditions, construction delivery, and community acceptance together; a site can have AI demand and still fail because one physical or political dependency is missing.

Source: Cushman & Wakefield EMEA and France Data Center Market Study 2026 adds a hub-comparison lens: AI/cloud demand, energy availability, and regulation interact across Paris, Frankfurt, London, Dublin, and Milan. The public synopsis does not support a ranking; use the hubs to structure comparable diligence, not to assume that European demand equals permitted, powered, financed, or leased supply.

Source: Cushman & Wakefield France AI Data Centers From Algorithm to Infrastructure 2026 adds a workload-to-building translation: generative AI changes not only demand volume but also density, power, cooling, and architectural requirements. Model these as linked technical dependencies; a favorable energy-policy narrative cannot cure an unverified grid connection, liquid-cooling design, heat-rejection path, capex plan, tenant workload, or operating standard.

Loudoun Proposal — September 24 Evidence Boundary

Full retained county and reporting-source review confirms that September 15 advanced a proposed final-Board-action pause. The October 20 resolution was pending as of September 24. This record does not establish an enacted moratorium, an October 6 decision, project cancellations or loss of powered capacity. Source: Source: Loudoun County Data Center and Substation Application Pause Proposal 2026.

Complete held source review, October 11, 2026

The complete September 16 Connect article reports Cushman & Wakefield's Americas data-center construction at a record 37.7 GW. It does not provide a named facility, power deliverability, commissioning date, market allocation or dataset methodology. This is secondary coverage of the same primary research, not an independent normalized capacity series. John McWilliams, head of data-center insights at C&W, describes power availability, grid infrastructure and regulatory certainty as competitive advantages and demand as insatiable; these are attributed commentary, not measured unlimited demand or investor return. Established Northern Virginia, Atlanta, Chicago and Dallas are named for connectivity, customer concentration and mature ecosystems. West Texas, Pennsylvania, Alberta and Cheyenne are named as emerging areas with some large proposed developments. Proposed projects are not delivered, financed, permitted or energizable capacity; no individual allocation, ranking, price, available MW, utility commitment or parcel is supplied. Primary operational, precommitment, planned-pipeline and vacancy numbers discussed elsewhere in the source note are not additional values in this recap. All three complete faithful panels and linked rack illustration were inspected; panels are identical because the article fits in one viewport. The unlabeled rack image does not establish a real named facility, electrical load, cooling specification, rack count, physical condition or operating capacity. Portrait, biography, CTA and navigation are excluded from asset facts. Use held primary research for its own supported aggregates and keep missing embedded maps/datasets excluded. This article supports a source-specific siting hypothesis, not an investment conclusion without investor price, objectives, horizon, execution ability and power/site diligence.

• Source: Connect CRE Cushman & Wakefield Power and Data Center Map 2026.

Complete held robotaxi source review, October 11, 2026

Complete September 28, 2026 Bisnow article, 7:34pm ET, reviewed across five contiguous panels and three original images. Cushman & Wakefield data supplied to Bisnow reportedly show Lyft, Uber, Waymo and Zoox nearly 1 million SF industrial leases during 2026 versus 830,000 SF combined 2022–2025. Underlying dataset, cutoff, geographies and lease definitions unheld; unequal periods cannot establish an annual growth rate or normalized market series. Reed Vestal's doubling interpretation and tens of millions of square feet within a decade are attributed speculation, not calibrated forecast or inventory. Facilities store, clean, charge, repair and stage fleets; signed space does not prove current operations. Article says 26 states allowed robotaxis as of June, but governing rules and later status unheld; no present legal permission is asserted. Waymo, Zoox, Lyft, Uber and Tesla did not respond to requests for real-estate comment.

Site scopes: Flexdrive subsidiary of Lyft reportedly began construction in April at planned 80,000-SF Nashville Waymo-service depot; held April 15 primary says construction already underway and fall opening intended, not proof construction began that day or later delivery. Uber reportedly secured 50,000-SF Houston depot/charging space in June, intended operations in 2027, citing Axios. Zoox May lease 34,000 SF Hayward logistics/staging hub is distinct from 219,000-SF manufacturing facility reportedly opened 2025 nearby, citing The Real Deal; exact addresses and executed terms unheld. Waymo 32,000-SF Northeast DC leased facility is one of three developing maintenance/charging facilities, WBJ August attribution; do not multiply 32,000 by three. July spokesperson Ethan Teicher to Axios describes tens of millions committed in Wards 5/7, no precise amount, project allocation or proof paid. Waymo jobs-hundreds assurance and no-loss assertion are company expectations, not demonstrated net employment effects. Tesla Lars Moravy January analyst-call statement credits existing service/charging network and claims unique ability to scale; transcript and quantitative capacity unheld, competitive advantage not independently established.

Regulation and opposition: Article attributes Upstate proposal withdrawal by Governor Hochul following People First Transportation Alliance opposition; NYC trial testing in Downtown Brooklyn/Lower Manhattan and March permit expiry without renewal; current legal texts/permits unheld, not verified present authorization. DC Charles Allen May limited-test bill, cyclist/disability transparency concerns, Teamsters/32BJ/ATU689 livelihood objections and planned October 6 religious/labor rally are dated proposal/positions, not enacted terms or proof rally occurred. Ryan Nelson of Turnbridge reports typical 50,000–100,000-SF space searches near Manhattan in Long Island City/South Bronx; no signed individual NYC deals inferred. Stephen Millington of Rosenberg & Estis links prelegislation commitments to permitting and infrastructure lead time; Nelson says ConEd hookup can take years and requires intended-use/capacity review. These are practitioner observations, not measured utility service standard, all-site guarantee or statutory requirement. Vestal attributes confidentiality to rent bargaining, tech backlash and worker/electricity concerns; motive is hypothesis, not demonstrated causal evidence.

Safety and perception: Almost 60% April Gallup respondents favour human-driver road safety is secondary survey attribution with sample/method unheld. Waymo-reported 95% fewer serious-injury crashes differs in outcome and denominator from article's IIHS-attributed 68% fewer accidents across San Francisco, Phoenix, Los Angeles and Austin; studies, exposure baselines and methods unheld. Do not combine these into independent safety certification or market-wide causal conclusions. Aseon Labs cofounder Dan Keene predicts public trust acceptance and Nelson anticipates market entry; interested-party hypotheses, not inevitable approvals or investments. Original Waymo street photograph, NYC night traffic and Cybercab showroom image show vehicles/context; none establishes lease area, fleet utilization, actual autonomy, legal approval or injury-rate effects. Cybercab photograph cannot establish separate claimed NYC sightings. Contact/navigation excluded; complete held public authored scope reviewed.

Complete held April 15, 2026 Lyft announcement by Flexdrive CEO John Parks reviewed through two contiguous panels and five original Chad Ziemendorf photographs. Says construction already well underway, crews rewiring planned 80,000-SF Nashville depot, intended fall 2026 opening in Waymo partnership for service/charging/maintenance. Charging capacity for hundreds of vehicles and future satellite cleaning/charging sites are proposed operational programme, not delivered capacity, precise stall count, uptime or current occupancy. No street address, title, lease execution, cost, rent or independent commissioning supplied. First Atlanta fleet facility 2016; Flexdrive says 24 North American sites manage roughly 15,000 vehicles as of announcement. Different network universe from Nashville depot, not a count of autonomous vehicles or occupied Nashville capacity.

Company plans more than 70 full-time Nashville-area jobs in 2026, technician/operations/fleet/maintenance roles; 50% of full-time roles filled so far for depot were current/former Lyft drivers, but filled count and survey date not disclosed. More than one-third of national employees were drivers is a different denominator and should not become 50% of all proposed 70 jobs. Former/current drivers receive priority for next interview while roles open; not guarantee of employment. Jonathan, first Nashville newly hired employee and Operations Lead, had more than ten years driving on Lyft; individual profile not representative causal employment evidence. Waymo/Flexdrive administer technical/training requirements without formal curriculum supplied. Hybrid driver/autonomous-service network, high utilization/quick ETAs, improved access/reduced car ownership/reliability are company intentions and performance expectations, not measured outcomes or proof no displacement.

Forward-looking statement expressly includes Nashville facility construction, employment/training, partnership/services, other markets, autonomous strategy and benefits, with estimates/assumptions/uncertainty and risks referenced to SEC filings not newly acquired here. No later completion, legal operating clearance, net job effects or technical safety conclusion established. Images show conduit installation, interior charger equipment, hard-hat visitors, a worker portrait and exterior loading docks with stored conduit; they corroborate pictured fitout context, not energized equipment, operational charging power, measured area, worker counts, full fleet utilization or completion certificate. Photograph of group is not headcount of all employees. Primary supports its own announcement; source bytes remain preserved in original supporting package and separately routed identity without changing original custody.

Underwriting interpretation

Power access, approvals and usable site configuration can govern depot delivery. Reported leases and space searches support an emerging demand hypothesis. Unequal aggregate reporting periods, unknown executed terms, operating-permission risk and uncertain public acceptance limit a market-rent or vacancy conclusion. Test that hypothesis against dated lease-level evidence, utility capacity, approval conditions and actual utilization before committing capital.

• Source: Robotaxi Firms Grow Industrial Footprint 2026.
• Source: Lyft Flexdrive Nashville Complete Announcement Review.

Complete reproduced source review, October 11, 2026

Complete September 30 reporting preserves one combined $1 billion Samsung commitment alongside more than $10 billion of founding platform commitments. These are announced capital commitments, with funding/deployment, project allocations and realized returns unknown. Worldwide data-center/power/connectivity strategy and expected Samsung benefits are participant plans, not delivered capacity or signed contracts. Existing separate issuer affiliate allocations retain attribution. The unidentified hero image establishes no bounded property, workplace or Helix ownership; original historical image identity remains unresolved.

• Source: Connect CRE Samsung Helix Complete Commitment Review September 30 2026.

Saline application — financing, construction and operating supply

Source: Related Digital Secures Financing for $16B Oracle Data Center Project in Michigan provides a dated capital-stack case for Michigan Data Centers and Powered Land. Keep the April closing conditions, later construction account and Source: MPSC Saline Data Center Contract Admission 2025-26 separate from full drawdown, energized phases, tenant acceptance and rent commencement. Use CRE Capital Markets, Data Center Underwriting and Powered Land and Energy Infrastructure and CRE for actual funding and service. Do not combine this campus’s announced capacity with wider Stargate commitments or Google’s separate Van Buren proposal.

Castroville application — accessibility review and delivered supply

Source: Reports: Microsoft Plans $400M Data Center in Castroville, Texas is a January-reported SAT82 proposal in Medina County. Keep the retained report’s estimates and additional-project attribution separate from later-read registry fields, actual construction and tenant acceptance. Accessibility review does not establish energized IT capacity or operating inventory. Austin-San Antonio Data Centers and Powered Land Market gives the regional boundary; Data Center Underwriting and Powered Land, Energy Infrastructure and CRE and CRE Capital Markets route phase delivery, utility obligations and actual funding. Do not use this project to change regional warehouse metrics or infer a site-selection advantage.