IBM Sovereign Core targets AI-ready digital sovereignty

IBM has announced IBM Sovereign Core, positioning it as an AI-ready, sovereign-enabled software foundation designed for enterprises, governments, and service providers. The move comes as organisations worldwide face growing regulatory pressure to retain direct control over their technology environments, particularly as AI workloads become central to operations.

The company frames digital sovereignty as a broader challenge than data residency alone. Control over infrastructure, authority over operations, governance of data access, and jurisdiction over AI models are all becoming critical issues as artificial intelligence moves from experimentation into production systems.

Regulatory pressure meets AI adoption

Enterprises and public-sector organisations are increasingly seeking self-managed environments where they retain operational authority. This need is amplified by AI use cases such as large language models and agentic workflows, which often process sensitive data and raise concerns around jurisdiction and accountability.

According to industry expectations cited in the announcement, a majority of enterprises are expected to formalise digital sovereignty strategies within the decade, often centred on sovereign Cloud approaches.

IBM executives argue that without a clear sovereign destination, many organisations struggle to modernise or re-host applications, especially those incorporating AI capabilities, while also maintaining auditable and continuous compliance.

Sovereignty built into software

IBM Sovereign Core is positioned as purpose-built software that embeds sovereignty directly into the technology stack rather than layering controls on top of existing architectures. Built on Red Hat’s open source foundation, the platform is designed to support cloud-native and AI workloads within jurisdictions chosen by the customer.

Key capabilities highlighted include:

• Customer-operated control plane allowing organisations to retain authority over deployment, configuration, and operations without reliance on out-of-region vendors.

• In-boundary identity and key management, ensuring authentication, authorisation, and encryption remain under local control.

• Continuous compliance enablement, with operational data, telemetry, and audit trails generated and stored within sovereign boundaries.

• Governed AI inference, enabling local model deployment, GPU clusters, and AI agent operations without exporting data externally.

• Flexible deployment, supporting on-premises datacentres, in-region Cloud infrastructure, or managed environments via service providers.

The company positions this approach as addressing not only compliance requirements but also the ability to demonstrate control to regulators on an ongoing basis.

Open standards and operational autonomy

IBM emphasises that the platform aligns with its broader open hybrid Cloud strategy. By relying on open-source technologies and open standards, the company aims to reduce dependency on single vendors and allow organisations to participate in wider innovation ecosystems without compromising sovereignty.

This positioning reflects a shift in enterprise thinking, where sovereignty is no longer seen as a trade-off against openness, but as a governance challenge spanning data, operations, and infrastructure.

Deployment through local ecosystems

IBM Sovereign Core can be deployed across customer-owned environments or through in-region IT service providers. Initial collaborations have been announced with service providers in Europe, enabling local operational independence while allowing partners to offer differentiated sovereign services for AI-scale workloads.

Service providers involved highlight the ability to reduce deployment timelines by relying on a pre-architected sovereign software foundation, rather than assembling and validating disparate components over extended periods.

Availability timeline

IBM stated that Sovereign Core will be available in technology preview starting February, with general availability planned for mid-2026. Additional capabilities are expected to be introduced at the time of general release.