Hyperscale Facilities
This brief covers hyperscale cloud and AI data center campuses — the facilities built and operated by hyperscalers and dedicated campus owners such as Microsoft, Amazon, Alphabet, Nebius Group and Digital Realty Trust, alongside the equipment, power and construction enablers, including Vertiv, Modine, Nvidia, AMD, Oracle, Jacobs Solutions, Quanta Services, Celestica and Sanmina, that supply, build and operate the underlying infrastructure.
Active institutional exposure flipped from a net outflow of $38.1 billion in Q1 2026 to a net inflow of $86.3 billion in Q2 2026. 73.5% of companies in this sub-theme saw increased active positions in Q2 2026.
Overview
Hyperscale Facilities refers to the massive, purpose-built cloud and AI data center campuses developed and operated by hyperscale cloud providers and specialized data center owners. Leaders in this sub-theme include hyperscalers such as Microsoft, Amazon and Alphabet, which build and lease capacity directly, alongside dedicated campus owner-operators such as Digital Realty Trust, which develops hyperscale campuses in markets including Northern Virginia, Frankfurt and Paris through joint ventures with partners such as Blackstone and Realty Income, and Equinix, a competing global colocation operator. AI-cloud-focused leaders such as Nebius Group, and HPC/colocation developers such as Core Scientific, Cipher Digital and Applied Digital, build or lease large-scale compute campuses under long-term contracts with hyperscaler and AI-lab tenants. Power-centric developers such as Fermi Inc pair gas turbine and natural gas supply agreements with potential hyperscaler tenancy to build gigawatt-class campuses. Enabling this ecosystem are equipment and construction suppliers: thermal management and power infrastructure providers such as Vertiv and Modine; engineering and construction firms such as Jacobs Solutions, Quanta Services and EMCOR; electronics manufacturing, server and networking suppliers such as Celestica, Sanmina, Foxconn, Arista Networks, Lenovo and TE Connectivity; and semiconductor and systems vendors such as Nvidia, AMD and Oracle, whose compute platforms underpin the campuses. Investment capital flows through diversified asset managers such as Brookfield Asset Management and DigitalBridge, which fund infrastructure, power and data center platforms alongside direct hyperscaler capital expenditure.
Investment Thesis
Investment thesis details are part of the Pro plan.
Bottleneck Tracker
Bottleneck details are part of the Pro plan.
Key Companies
Related Events
Brookfield and its Radiant neocloud platform plan a UK AI campus at Drax Power Station that could exceed £20 billion and 1 GW of capacity. Drax shares rose about 8.8% as investors priced in monetization of its grid-connected site. Phase 1 (up to 100 MW from Q1 2028, £2.5 billion) still needs planning, regulatory and network approvals, and a tenant, and no final investment decision has been made. The project tests whether the UK can host large-scale AI compute despite high power costs and grid connection delays, which pushed Microsoft out of the Nscale deal in October. If delivered, it would add sovereign-style UK compute capacity and demand for power, cooling, chips and construction, though execution risk remains high.
Anthropic's filing shows a compute obligation base of about $895 billion, and $437 billion of that sits in non-cancellable deals with Microsoft, Google, Amazon and Broadcom. Revenue growth is slowing (46% in Q3 versus roughly 2.5x in Q2) while compute costs grew 77%, which raises concern about whether revenue can cover the commitments. Suppliers and hyperscalers gain long-dated contracted demand visibility, but they also take on counterparty concentration risk if Anthropic's growth stalls. About 44% of Anthropic's revenue flows through cloud marketplaces, which ties its fortunes to partners that also compete with it. The IPO could set the valuation and financing benchmark for frontier AI labs, and investors will watch whether capital markets can fund spending on this scale.
Google is turning TPUs into a merchant AI-chip business by distributing them through Oracle's cloud, widening its reach beyond its own data centers. Analysts model TPU sales of $9-10 billion in the second half of 2026 and $62-99 billion in 2027, which would make Google a credible Nvidia challenger. External TPU sales carry a roughly 20% operating margin versus about 35% for overall cloud, so the mix shift could dilute Google Cloud margins even as revenue grows near 90%. Anthropic and OpenAI as customers deepen Google's role as supplier to its model rivals, and Broadcom's $63 billion lease vehicle shows capital markets financing TPU buildouts. Nvidia faces more competitive pressure on price-performance, while Oracle gains a differentiated AI offering.
Microsoft's second voluntary buyout round in 2026 covers about 27% of staff, a sign that AI capex is being financed partly by shrinking headcount. The company projects $145.3 billion in fiscal 2026 spending and reported $41 billion in capex for the June quarter, so labor cost cuts help protect margins. The spokesperson pointed to a 3% to 5% uptake at peer programs, implying roughly 1,800 to 3,000 departures against 60,000 eligible. Follows 4,800 July layoffs, more than 15,000 layoffs in 2025, and a wave of senior executive exits. Earnings are due Wednesday, where investors will look at capex guidance and any restructuring charge; the stock is up about 25% over six months.
Meta is turning to equity markets on top of operating cash flow and investment-grade debt to fund an AI buildout, with capex guidance of $135-145 billion for the year. Shares fell 1.6% on dilution concerns and negative free cash flow. Interest from Nvidia, AMD and Broadcom in buying up to $5 billion of the shares shows suppliers recycling revenue into customer funding. The raise follows large financings by Alphabet and SpaceX and a planned Anthropic IPO, so AI capital demand is crowding the capital markets. It also signals that even the best-funded hyperscalers can no longer self-fund AI capex from cash flow alone.