Sovereign AI in Europe expands beyond data localization

Sovereign AI in Europe is moving beyond just keeping data local, according to Philippe Lecanthe of Dell Technologies. It now focuses on creating secure, observable, and resilient AI environments that can evolve at scale for future AI systems.

Sovereign AI faces new challenges

The evolution of sovereign AI is unavoidable, as it now requires organizations to access computing power, maintain control over strategic assets, govern AI usage, and accelerate innovation while ensuring security, compliance, and trust.

European AI investment reaches 200 billion euros

The European Union announced in October 2025 its goal to invest 200 billion euros in AI by the end of the decade. This includes the "AI Factories" initiative, which combines supercomputing, data, and talent to develop trustworthy generative AI.

Investing in AI requires coordinated efforts

For these investments to create sustainable value, companies must link sovereign computing capabilities with reliable data control, consistent governance, and true resilience in AI operations.

New trust architecture is essential for AI

As AI becomes central to daily activities, protecting data is no longer enough. Organizations must show where data is hosted, how models are developed, who can access infrastructure, and how services remain secure and resilient at scale.

AI is moving toward agent systems

AI is evolving toward agent systems that can plan, coordinate, and execute tasks across environments. These systems will continuously interact with applications and contribute to complex decision-making, requiring AI infrastructure to support continuous and observable operations.

Operational priorities for sovereign AI

CIOs and technology leaders in the public sector must translate sovereign AI ambitions into operational capabilities that produce measurable results. Five priorities are essential to build secure, controlled, and scalable AI environments.

Deploy AI where impact is strongest

Sovereign AI should be deployed where it can generate measurable value quickly, such as in public services, industrial processes, or research. It must start with a clear objective, like improving national data use or increasing competitiveness.

Adapt infrastructure to data sensitivity

Sovereign AI does not mean all workloads must be handled the same way. Some may require national control, while others might need private infrastructure, edge computing, or hybrid models, depending on data sensitivity and performance needs.

Integrate governance into AI operations

Governance is as crucial as computing power. Sovereign AI must ensure transparency and responsibility in data use, model governance, and AI decision-making. This includes clear roles, audit trails, and training for AI-related responsibilities.

Prepare for the rise of agent systems

The next phase of AI involves agent systems capable of continuous, distributed operations. Traditional environments may not suffice, requiring infrastructures that handle permanent workloads across on-site, edge, and cloud environments while maintaining governance and human oversight.

Resilience is a key part of sovereignty

Sovereignty is not just about data or infrastructure control. It must also ensure operational resilience, allowing systems to be restored quickly and maintain continuity during incidents, cyber threats, or traffic peaks.

Future of sovereign AI depends on integration

The future of sovereign AI in Europe will depend on combining infrastructure, governance, and resilience into a coherent strategy. Organizations that move beyond data residency to build genuine operational trust will lead the way in deploying AI responsibly and securely at scale.