Cloud is no longer enough
For two decades, moving to the cloud was synonymous with modernization. Today, organizations face a different reality—one that no hyperscaler can solve alone. Sovereignty regulations require local control of data, AI models demand low latency and privacy, cloud costs are rising faster than revenue, and operational complexity is overwhelming teams.
The future is not “more cloud,” but the right cloud, in the right place, operated in the right way. Future Cloud Infrastructure represents a distributed infrastructure ecosystem that integrates on-premise, edge and public cloud into a unified, autonomous and sustainable platform.
The shift from multicloud to hybrid cloud
The dominant multicloud narrative—distributing workloads across multiple providers to avoid vendor lock-in—is losing momentum in 2026. Private cloud has returned to the strategic agenda, driven by three forces:
- Sovereignty and regulation: GDPR, the EU AI Act and sector-specific regulations require certain data and AI models to remain under direct organizational control.
- Cost and predictability: predictable, high-volume workloads are more cost-efficient on private infrastructure than in public cloud.
- On-premise AI: training and running proprietary models requires a level of control that public cloud does not always guarantee.
Forrester predicts that at least 15% of organizations will deploy private AI on private clouds by 2026. Broadcom warns that cloud repatriation is not a tactical adjustment, but a strategic move to regain control. The winning model will be hybrid cloud: an architecture that combines the best of multicloud and private cloud under unified governance.
360-degree sovereignty across data, infrastructure and AI
Sovereignty goes beyond where data resides. The real question is whether control, governance and auditability travel with the data as it moves across AI pipelines, inference endpoints and multicloud environments.
Three key dimensions of sovereignty must be addressed:
- Data sovereignty: jurisdictional control over where information is stored, processed and transferred, with policies that travel with the data.
- Infrastructure sovereignty: the ability to decide what runs in public cloud, on-premise and edge, without dependence on a single provider.
- AI sovereignty: the ability to train, fine-tune and run AI models in controlled environments, ensuring that intellectual property and training data remain within a defined perimeter.
Data residency without governance is a sovereignty failure. Organizations need a unified data layer with a global namespace, policy-based governance and consistent lineage across all environments.
Cloud, edge and on-premise as a distributed platform
Future Cloud Infrastructure operates as a continuum, with infrastructure deployed where it delivers the greatest value:
- On-premise / data center: regulated workloads, sovereign AI, sensitive data and ultra-low latency applications.
- Edge: real-time processing in factories, retail environments, hospitals and service points, where decisions cannot wait for a cloud round trip.
- Public cloud: elasticity, rapid innovation, development and variable workloads that require on-demand scalability.
All of this operates under consistent security, governance, observability and automation policies. This is what differentiates a fragmented architecture from a modular, composable infrastructure platform.
Modular and composable infrastructure: the future consumption model
The infrastructure of the future is no longer acquired in monolithic blocks. It is assembled by combining compute, storage, networking and acceleration resources, such as GPUs, into modular building blocks consumed through APIs and policies.
- Software-defined everything: networking, storage and compute defined by software and decoupled from hardware.
- Platform engineering: platform teams provide developers with infrastructure as an internal service, with built-in security and compliance controls.
- Infrastructure as Code (IaC): infrastructure is declared, versioned and reproducible, eliminating manual configuration and drift.
This model enables organizations to respond to new business needs in minutes, creating complete environments for AI, development or production without waiting weeks for manual provisioning.
Autonomous operations: from reactive management to intelligent infrastructure
Operations are shifting to an autonomous model, with AI taking on a significant share of the workload required to manage hybrid, distributed and composable environments:
- Intelligent observability: AIOps platforms correlate events in real time across on-premise, edge and cloud environments and detect anomalies before they impact the business.
- Automated remediation: systems that not only alert but also execute corrective actions within policies defined by humans (human-in-the-loop).
- Continuous optimization: AI that adjusts workload placement, resource sizing and costs based on real usage patterns.
- Automated Zero Trust security: policies that are applied, monitored and continuously adapted across all environments.
Autonomous operations reduce incidents, optimize licensing and align every technology investment with business outcomes.
Resilience, security and sustainability as non-negotiable pillars
Resilience by design: Forrester warns that data center upgrades for AI will cause at least two major multi-day outages among hyperscalers in 2026. Resilience is achieved through distributed architectures, automated failover and redundancy across on-premise and cloud.
Distributed Zero Trust security: In a hybrid ecosystem, security must be an inherent property of every layer, including microsegmentation, confidential computing, end-to-end encryption, identity as the perimeter and continuous monitoring.
Sustainability as a regulatory and competitive requirement: The European CSRD directive, green taxonomy and investor pressure are turning data center energy efficiency into a business KPI. Future Cloud Infrastructure incorporates carbon metrics into workload placement decisions, cooling optimization and the selection of locations powered by renewable energy.
NTT DATA: Future Cloud Infrastructure in action
NTT DATA does not view Future Cloud Infrastructure as a concept. It executes it:
- Strategic partnerships with leading hyperscalers and technology vendors, with a presence in 34 countries and more than 14,000 technical experts.
- AI-powered SDI services already in market, including intelligent automation, predictive license management and reliability driven by AI agents.
- Hybrid cloud management capabilities that unify the management of public and private clouds, servers and network devices within a single platform.
- Proven experience in autonomous operations, with a clear roadmap toward self-managed infrastructure.
At NTT DATA, we don’t predict the future of infrastructure. We deliver it today.