Global cloud infrastructure spending crossed $800 billion in 2025. By the end of 2026, it will exceed $940 billion. By 2030, it will pass $1.5 trillion. These are not speculative projections. They are the compound arithmetic of enterprise digital transformation, AI compute demand, and the structural shift from capital expenditure to operating expenditure in IT budgets.

Within this expansion, the fastest-growing segments are not the ones generating headlines. Sovereign cloud, privacy-enhanced computing, and confidential infrastructure — markets that barely registered five years ago — are growing at two to three times the rate of the overall cloud market. Understanding where the growth is concentrated reveals where the next generation of cloud infrastructure companies will emerge.

Total Cloud Market: 2024-2030

YearTotal Cloud Spending (Global)YoY Growth
2024$679B21.5%
2025$807B18.8%
2026E$942B16.7%
2027E$1,085B15.2%
2028E$1,228B13.2%
2029E$1,368B11.4%
2030E$1,505B10.0%

Sources: Synthesized from Gartner, IDC, Synergy Research Group, and Canalys estimates. All figures in USD.

Growth is decelerating in percentage terms but accelerating in absolute dollars. The $135 billion added between 2025 and 2026 exceeds the total size of the cloud market in 2017. Scale creates its own gravity.

The deceleration reflects market maturation in traditional IaaS and SaaS segments, partially offset by acceleration in AI infrastructure, edge computing, and sovereignty-driven deployments. The cloud market is not slowing — it is recomposing around new demand vectors.

Market Segmentation: IaaS, PaaS, SaaS

The cloud market comprises three primary service models with distinct growth profiles:

Infrastructure as a Service (IaaS)

YearIaaS Market SizeYoY Growth% of Total Cloud
2024$182B26.3%26.8%
2025$228B25.3%28.3%
2026E$279B22.4%29.6%
2030E$485B32.2%

IaaS is the fastest-growing major segment, driven by AI training workloads (which require massive GPU clusters), sovereign cloud mandates (which force infrastructure replication across jurisdictions), and the migration of legacy on-premises workloads. IaaS’s share of total cloud spending is increasing — a structural shift that reflects compute-intensive AI workloads consuming a larger proportion of enterprise IT budgets.

Platform as a Service (PaaS)

YearPaaS Market SizeYoY Growth% of Total Cloud
2024$167B22.8%24.6%
2025$203B21.6%25.2%
2026E$244B20.2%25.9%
2030E$410B27.2%

PaaS growth is fueled by the platform engineering movement, Kubernetes adoption, database-as-a-service demand, and AI/ML platform services (SageMaker, Vertex AI, Azure AI Studio). PaaS is where cloud providers capture the most value — higher margins than raw IaaS, stronger lock-in than SaaS.

Software as a Service (SaaS)

YearSaaS Market SizeYoY Growth% of Total Cloud
2024$330B17.6%48.6%
2025$376B13.9%46.6%
2026E$419B11.4%44.5%
2030E$610B40.5%

SaaS remains the largest segment in absolute terms but is growing slower than IaaS and PaaS. The market is mature in core categories (CRM, ERP, HCM, collaboration) and incremental growth comes from vertical SaaS, AI-augmented SaaS, and expansion into emerging markets. SaaS’s share of total cloud is declining as IaaS and PaaS grow faster — a rebalancing toward infrastructure driven by AI compute demand.

Provider Market Share

The hyperscale cloud market remains concentrated, but the concentration is evolving:

IaaS + PaaS Market Share (2025)

ProviderRevenue (Est.)Market ShareYoY Growth
AWS$115B31%19%
Microsoft Azure$106B28%27%
Google Cloud$48B13%29%
Alibaba Cloud$16B4%8%
Oracle Cloud$10B3%35%
IBM Cloud$7B2%5%
Others$72B19%

The dominant narrative is Azure closing the gap with AWS. In 2020, AWS held 39% market share to Azure’s 19%. By 2025, the gap has narrowed to 3 percentage points. Azure’s growth is driven by three factors: Microsoft 365 cross-sell, Azure OpenAI Service (the exclusive cloud for OpenAI’s models), and aggressive enterprise licensing that bundles Azure credits with Microsoft EA agreements.

Google Cloud’s 29% growth rate is the highest among the top three, driven by BigQuery, Vertex AI, and increasing enterprise credibility. Oracle Cloud’s 35% growth rate (from a smaller base) reflects database workload migration and Oracle’s unique position offering autonomous database services.

The “Others” category — at 19% — is significant. It includes European sovereign cloud providers (OVHcloud, Scaleway, Deutsche Telekom), Asian cloud providers (Tencent, Huawei, NTT), and specialized infrastructure providers. This segment is growing at approximately 22% annually, faster than AWS, suggesting that the hyperscale oligopoly is not tightening — it is fragmenting at the edges.

The Privacy and Sovereignty Segment

The most strategically relevant segment for Stealth Cloud positioning is privacy-enhanced and sovereign cloud infrastructure. This segment is defined as cloud spending specifically motivated by privacy, sovereignty, or data protection requirements — including sovereign cloud services, confidential computing, privacy-preserving AI infrastructure, and end-to-end encrypted cloud services.

Privacy/Sovereignty Cloud Segment

YearSegment SizeYoY Growth% of Total Cloud
2024$42B28%6.2%
2025$56B33%6.9%
2026E$74B32%7.9%
2027E$97B31%8.9%
2030E$195B13.0%

This segment is growing at roughly double the overall cloud market rate. The drivers are regulatory (GDPR enforcement acceleration, new sovereignty mandates), market (enterprise demand for confidential computing, AI data privacy), and geopolitical (US-China technology decoupling, CLOUD Act concerns in Europe).

Several sub-segments show even faster growth:

Confidential computing infrastructure. $5.8 billion in 2025, growing at 38% CAGR. Driven by Intel TDX and AMD SEV-SNP adoption, with Azure’s confidential computing portfolio generating the largest share. See the confidential computing deep dive for technical detail.

Privacy-preserving AI. $3.2 billion in 2025, growing at 45% CAGR. This includes federated learning platforms, differential privacy tooling, and confidential AI inference. The segment barely existed in 2022 and is experiencing hypergrowth driven by regulatory pressure on AI training data practices.

Sovereign cloud services. $47 billion in 2025, growing at 23% CAGR. The largest sub-segment, driven by government procurement mandates and regulated industry compliance requirements. European sovereign cloud accounts for approximately 58% of this market.

End-to-end encrypted infrastructure. $1.8 billion in 2025, growing at 52% CAGR. The smallest sub-segment but the fastest growing. This includes services like ProtonMail’s infrastructure, Tresorit, and emerging Stealth Cloud architectures where the provider architecturally cannot access customer data. The 52% growth rate reflects a market in early hypergrowth — small in absolute terms but compounding rapidly.

Geographic Distribution

Cloud spending is geographically concentrated but diversifying:

Cloud Spending by Region (2025)

RegionSpending% of GlobalYoY Growth
North America$354B43.8%17%
Europe$185B22.9%19%
Asia-Pacific$201B24.9%22%
Latin America$32B4.0%25%
Middle East & Africa$22B2.7%28%
Rest of World$13B1.6%20%

North America’s share is declining from 48% in 2022 to a projected 40% by 2030 as other regions accelerate. Asia-Pacific’s growth is driven by India’s digital transformation, Southeast Asian cloud adoption, and Japan’s sovereign cloud initiatives. Europe’s growth outpaces North America for the first time, fueled by sovereignty mandates and public sector cloud migration.

The fastest-growing region — Middle East & Africa at 28% — reflects early-stage cloud adoption in markets that are building infrastructure with sovereignty requirements from inception rather than retrofitting.

AI Infrastructure: The Growth Accelerant

AI workloads are reshaping cloud market economics. GPU compute for AI training and inference consumed an estimated $65 billion in cloud spending in 2025 — roughly 8% of total cloud spending — and is projected to reach $180 billion by 2030.

The AI infrastructure market is even more concentrated than general cloud. NVIDIA GPUs power over 90% of AI training workloads, with AI cloud spending concentrated among a small number of providers that can procure NVIDIA hardware at scale: AWS, Azure, GCP, Oracle, CoreWeave, and Lambda Labs.

For the privacy segment, AI creates both threat and opportunity. The threat is that AI training on sensitive data (medical records, financial data, private communications) concentrates previously distributed data into centralized, high-value targets. The opportunity is that privacy-preserving AI techniques — confidential computing for training, differential privacy for model outputs, federated learning for distributed data — create new market demand for privacy infrastructure.

The AI privacy market specifically (privacy tools, confidential AI infrastructure, data anonymization for AI training) is projected to reach $12 billion by 2028, growing at 40%+ CAGR from a 2024 base of approximately $2.8 billion.

The Pricing Landscape

Cloud pricing trends affect market growth and competitive dynamics:

Compute pricing has declined 5-8% annually for standard instances over the past five years. GPU instances have bucked this trend — pricing for A100 and H100 instances has been stable or increasing due to demand exceeding supply. The H100 spot market premium reached 40% over on-demand pricing in 2024 before easing in 2025.

Egress pricing remains the most criticized element of hyperscale cloud economics. AWS charges $0.09/GB for data egress to the internet (declining with volume). Azure and GCP match this pricing. European providers have undercut: OVHcloud offers 1 TB free egress monthly, Scaleway provides 75 GB free. The European Commission’s Data Act (2024) includes provisions targeting cloud egress fees, which may force hyperscaler pricing changes in the EU.

Sovereign cloud premium. Privacy and sovereignty-enhanced cloud services carry a 15-40% premium over standard hyperscale equivalents. Confidential computing VMs carry a 5-15% premium for AMD SEV instances and 10-25% for Intel TDX instances. These premiums are compressing as adoption scales but remain significant.

Zero-knowledge infrastructure pricing is not yet benchmarked by major analyst firms due to the segment’s nascent size. Current providers (Proton, Tresorit, Stealth Cloud-category services) price at 2-5x standard cloud equivalents, reflecting smaller scale and higher per-unit encryption overhead. As the segment scales, pricing compression toward 1.3-1.5x premiums is expected by 2028.

Venture Capital and Private Equity Activity

Investment in privacy-focused cloud and infrastructure companies has accelerated:

  • 2024 total VC/PE investment in privacy tech: $8.7 billion (up from $5.4 billion in 2023)
  • Notable rounds: CoreWeave ($7.5B debt facility, 2024), Wiz ($1B Series E at $12B valuation), Proton ($66M community round, fully subscribed in 8 days)
  • Privacy infrastructure specifically: $2.1 billion in VC funding in 2024, up 67% year-over-year
  • Confidential computing startups funded 2023-2025: Anjuna Security ($60M), Edgeless Systems ($20M), Cape Privacy ($50M), Opaque Systems ($32M)

The Proton precedent is instructive. ProtonMail launched in 2014 as an encrypted email service. By 2025, Proton AG had expanded to VPN, cloud storage, calendar, password manager, and email — generating an estimated $500+ million in annual revenue, all built on zero-knowledge architecture. Proton demonstrated that privacy as a product feature can scale to significant revenue without venture capital dilution (Proton’s 2024 community round was its first external funding).

This precedent has not been lost on investors. The pipeline of privacy infrastructure startups seeking funding in 2026 is the largest the sector has seen, with particular concentration in confidential AI, encrypted collaboration, and software-defined perimeter networking.

Market Forecast: Key Predictions for 2027-2030

Based on current growth trajectories, regulatory momentum, and technology maturation:

  1. The privacy cloud segment will exceed $195 billion by 2030, representing 13% of total cloud spending (up from 6.2% in 2024). This is the fastest structural recomposition in cloud market history.

  2. European sovereign cloud providers will capture 35% of EU public sector cloud spending by 2028, up from approximately 18% in 2024. This shift is mandated by regulation and funded by public investment.

  3. Confidential computing will become default for new enterprise deployments by 2029. Intel and AMD’s roadmaps make TEE capabilities standard in server CPUs by 2027-2028, eliminating the performance premium and making confidential VMs the default rather than an option.

  4. At least one privacy-first cloud company will reach $10 billion valuation by 2028. The candidate pool includes Proton (if it raises growth capital), a Stealth Cloud-category entrant, or a confidential computing platform company.

  5. Egress fees will decline 50%+ in the EU by 2027 due to Data Act enforcement, creating pricing pressure that spreads to other markets and reduces cloud lock-in.

What the Data Tells Us

The cloud market is not a single market. It is at least four overlapping markets with different growth rates, competitive dynamics, and structural drivers:

  • Mature cloud (standard IaaS/SaaS for non-sensitive workloads): Growing 10-15%, dominated by incumbents, commodity pricing
  • AI infrastructure (GPU compute for training/inference): Growing 35-45%, supply-constrained, hyper-concentrated
  • Sovereign cloud (jurisdictionally controlled infrastructure): Growing 20-25%, regulation-driven, geographically fragmented
  • Privacy infrastructure (zero-knowledge, confidential, encrypted): Growing 30-50%, early-stage, venture-funded

The first two markets are well-understood and well-served. The second two represent the frontier — where new categories are forming, new companies are emerging, and the architecture of cloud computing is being fundamentally reconsidered.

The Stealth Cloud Perspective

The market data is unambiguous: privacy-enhanced cloud infrastructure is growing at 2-3x the overall cloud market rate, driven by regulatory acceleration, enterprise demand, and genuine technological maturation. A $195 billion market by 2030 is not a niche. It is a category — and the companies that define its architecture will capture a disproportionate share of the next decade’s cloud value creation.