AI infrastructure M&A

TOPIC HUB · AI INFRASTRUCTURE M&A

AI infrastructure M&A

The reference on M&A activity in AI infrastructure: sponsor + strategic acquisitions across data centre operators, power semiconductor targets, cooling vendors, and industrial-technology consolidation. Deal case studies, DD methodology, and the capital-source landscape.

01Direct answer

AI infrastructure M&A is the accelerating deal activity in the physical + commercial infrastructure supporting AI: data centre operator acquisitions (Blackstone/QTS, Digital Realty/Teraco), hyperscaler-strategic transactions (Amazon/Talen, Microsoft/Constellation TMI PPA), power-electronics consolidation (CoolIT (Ecolab), Eaton/Boyd, Wolfspeed restructuring), and antitrust-adjacent deals (FTC/NVIDIA-ARM block, CMA UK hyperscaler cloud investigation). PE sponsors are the dominant deployer of capital via permanent-capital vehicles, but strategic acquirers (Eaton, Schneider, Vertiv, Emerson) are consolidating adjacent AI-critical segments. DD methodology for these deals differs from generalist M&A because the power/thermal/DC assets carry specific technical + regulatory + supplier-concentration risks that require practitioner knowledge.

02Key concepts

Take-private LBO

Public-company acquisition financed with debt; Blackstone/QTS is the AI-era reference.

Strategic acquisition

Adjacent industry buyer entering AI-adjacent segment; Eaton/Boyd is a reference.

Antitrust review

FTC/DOJ + CMA + EU + specific-country reviews for AI-adjacent deals.

CFIUS

US inbound foreign-investment review; specific relevance for cross-border AI infra deals.

QoE

Quality-of-earnings; specific line-item validation in AI infra with contract-heavy revenue.

Contract collateralisation

The specific way AI infra contracts create financeable + acquireable value.

Vendor concentration

Supplier-side concentration risk (SST vendors, POL silicon, connectors); specific DD line.

R&W insurance

Reps + warranties insurance; standard on larger AI infra transactions.

04Deep-dive research

AI infrastructure M&A mapThe transaction landscape at high level.
Case Study 01. Blackstone/QTSAI-era LBO reference; 5-year retrospective.
Case Study 02. Microsoft/Constellation TMIHyperscaler nuclear PPA reference.
Case Study 03. Amazon/Talen FERC rejectionBTM nuclear cautionary case.
Case Study 04. CoreWeave post-IPOAI-native GPU cloud public-market reference.
Case Study 05. Iron Mountain DC pivotAdjacent-market entry via 10-year build.
Case Study 06. Digital Realty/TeracoEmerging-market DC platform via M&A.
Case Study 07. FTC/NVIDIA-ARM blockAI-adjacent M&A antitrust reference.
Case Study 08. CMA UK cloud investigationCloud market antitrust precedent.
DD I. The scoping decisionAI infra DD entry point.
DD Supplement A. Question banks200+ practitioner DD questions.

06Commercial + investment implications

For sponsors + strategics

DD methodology for AI infra deals requires the technical layer (power/thermal/vendor) beyond financial + legal + commercial. The 14-essay DD series covers the methodology end-to-end. For sponsor-specific playbook: PE playbook layer by layer. For scoping specific engagements: Advisory.

07Frequently asked

What is different about AI infrastructure M&A DD?

Three specific layers: (1) technical DD on power/thermal/DC architecture (contract-heavy but architecture matters materially); (2) supplier-concentration risk at multiple layers (SST vendors, POL silicon, high-current connectors); (3) regulatory overlay (CFIUS on cross-border, FERC on grid-adjacent, EPA + local zoning on facilities). Generalist DD misses layer 1 and 3.

Why is Blackstone/QTS the reference AI-era LBO?

Timing + scale + strategic thesis. Take-private June 2021 at $21B EV, hyperscaler pre-lease strategy validated 2022-2024, portfolio value inflected upward on AI capacity demand. Reference case for permanent-capital vehicles deploying into DC operators.

Which strategic acquirers are most active?

Vertiv (Strategic Thermal Labs for cold-plate engineering; PurgeRite for fluid management), Eaton (via Resilient Power Systems SST acquisition + Boyd Corporation cooling), Schneider (EcoStruxure expansion), Emerson (industrial automation adjacencies). Chinese strategics (BYD, Sungrow) active domestically but constrained on Western targets.

How does CFIUS affect cross-border AI infra deals?

Meaningfully for Chinese acquirers (near-certain block or forced structure), meaningfully for Middle East + APAC sovereigns (mitigation agreements common), less so for European strategics (standard review). Outbound Investment Rule (launched 2025) adds a separate US-outbound overlay for advanced-computing targets. See Regulatory V.

What is the antitrust outlook for AI-adjacent M&A?

Elevated scrutiny for horizontal deals in cloud/hyperscaler + AI chip segments. FTC/NVIDIA-ARM block is precedent. CMA UK hyperscaler cloud investigation is precedent. Vertical deals in cooling/power/DC infra generally clearing but require careful structuring.

08Further reading

See also: Financing (capital structures), Regulation (CFIUS + antitrust). Full series: Case Studies (8 essays) + DD (14 essays).

Definition AI infrastructure M&A across 2023-2026 covers vendor consolidation across the six layers of the AI power chain and adjacent infrastructure. Notable transactions include Eaton acquiring Boyd Thermal for approximately $9.5B (closed March 2026 per company disclosures), Ecolab acquiring CoolIT from KKR for approximately $4.75B (announced March 2026), Blackstone acquiring QTS at the developer layer, and Vertiv building cooling capability through capability tuck-ins.