Managed Security Services: CISO Does Not Bear Sole Liability
Benedikt Langer
8 min. read In many organisations, the CISO is seen as the person who stands accountable for security. ...
6 min read
The EU unveiled its Tech Sovereignty Package on 27 May. It proposes restricting the use of US cloud providers for sensitive government data across all 27 member states. Three US corporations control around 70 percent of the European cloud market. This turns a question that has lingered in IT procurement into a boardroom issue: whose law governs access to our data, and what happens if geopolitics shifts.
Key Takeaways
Related:Why boards demand defensibility over vision boards / How capital markets rate AI governance
What is the CLOUD Act? The CLOUD Act is a US law that can require American companies to surrender data they store, no matter where the servers are located. A provider under US ownership cannot sidestep this obligation with a contractual promise of data residency, because the law overrides such promises.
Cloud selection used to be a procurement decision: which vendor delivers which service at what price. That view is outdated. With the EU package and growing awareness of the CLOUD Act, the question becomes a strategic risk assessment that the board, legal and compliance must tackle together. It revolves around operational resilience, geopolitical exposure and what happens if access is restricted for political reasons.
Recent events make this tangible. Reports emerged of a sovereign Microsoft service in Frankfurt that briefly stopped working, and of exposed cloud keys belonging to a US agency. These incidents are no longer abstract worries; they are concrete proof that dependence on a handful of providers is its own risk class. A board that hasn’t evaluated this has a gap in its risk register.
Sovereignty is not the same as data residency. A data centre in Frankfurt meets residency requirements, yet if the operator is under US ownership, exposure to the CLOUD Act remains. True sovereignty requires that neither operations nor legal access depend on a foreign jurisdiction. This distinction belongs in every boardroom discussion because it determines whether decisions are based on appearance or substance.
The EU has sent its own signal. In April, the Commission awarded a contract worth up to €180 million over six years to four European providers, explicitly including sovereignty criteria in the tender for the first time. This is more than symbolism. It establishes sovereignty as a procurement standard and creates a market for providers that meet these criteria.
Apparent sovereignty
Reliable sovereignty
The first step is an honest inventory of dependencies. Which critical processes run on which provider, where are the data located, and under which jurisdiction does the operator actually fall? Many organisations lack this map because cloud choices were historically decentralised and made by function rather than geopolitical risk.
The second step is not a hasty migration. No one benefits from pulling everything out of the US cloud overnight. A phased approach is more sensible: which workloads are non-critical and can stay where they are, and which are sensitive enough that the sovereignty question becomes mandatory. This prioritisation is a board decision because it weighs risk against cost and effort-and that is not a purely IT matter.
A server in Frankfurt proves nothing if foreign law can reach its data. Sovereignty is decided by jurisdiction, not postal code.
What remains is the classification as a mandatory task, not an optional extra. The direction of regulation is set, market concentration is real, and recent incidents show the risk is not theoretical. A board does not need to finish the assessment today. It does need to show that it has started. Anyone who cannot answer at the next audit or geopolitical upheaval has left the question too long in procurement.
No. A location within the EU meets data residency requirements, but if the operator is owned by a US company, exposure to the CLOUD Act remains. True sovereignty requires that legal access is also bound to EU jurisdiction.
It proposes restricting the use of US cloud providers for sensitive government data across member states. The backdrop is high market concentration: three US providers control roughly 70 percent of the European cloud market.
Not across the board. A risk-based approach makes sense: non-critical workloads can stay, while sensitive ones should be evaluated against sovereignty requirements. A hasty migration introduces more risk than it resolves.
Because it touches on operational resilience, geopolitical exposure, and regulatory continuity-not just performance and price. This trade-off is for the board, legal, and compliance teams to weigh together, not procurement alone.
The market is emerging. In April, the European Commission awarded its first cloud contract with explicit sovereignty criteria to European providers. This sets a standard and creates demand for sovereign offerings.
More from the MBF Media Network
Image source: AI-generated (May 2026), C2PA certificate embedded in image