From Cloud to Orbit: What Companies Should Ask Before Their Data Leaves Earth (Part Two)

What Changes When Compute Leaves Earth — Risk, Governance, and Procurement Realities

In Part One, we explored how the rapid growth of AI is colliding with the physical limits of Earth‑based infrastructure. As organizations begin to confront these constraints, the question is no longer whether change is coming, but how leaders should evaluate emerging alternatives. This second part focuses on the critical questions enterprises should ask before considering any scenario where data extends beyond Earth.

Solving Physical Constraints Creates New Forms of Risk

Even if orbital compute addresses energy and cooling challenges, it introduces a different category of complexity. When compute leaves Earth, risk does not disappear—it shifts. The most important changes are not technical. They are organizational, regulatory, and contractual.

Jurisdiction Becomes Unclear by Default

Most data governance frameworks assume infrastructure exists within a defined national or legal boundary. Space does not fit cleanly into that model. When processing occurs in orbit, questions arise that lack established precedent:

• • Where is the data legally considered to reside?
  • Which country’s laws apply to processing and storage?
  • How are audits conducted when infrastructure cannot be physically accessed

This ambiguity does not make compliance impossible, but it raises governance expectations significantly. Vendors must explain jurisdictional models clearly and defensibly. Boards and regulators will expect answers long before any workload is deployed.

Security Shifts to the Transmission Layer

In orbital environments, physical security becomes secondary. Secure transmission becomes foundational.

Every workload depends on encrypted, resilient Earth-to-orbit communication links. If those links fail—or are compromised—compute capacity is irrelevant. This makes encryption, key management, monitoring, and incident response core capabilities rather than supporting functions.

For procurement teams, evaluating security maturity in these areas becomes non-negotiable.

Resilience Must Be Engineered, Not Repaired

Traditional data centers benefit from human intervention. Hardware can be replaced. Facilities can be serviced. Orbital infrastructure does not have that luxury.

Resilience must be designed into the architecture itself through redundancy, automated workload shifting, and seamless fallback to terrestrial systems. Failure should trigger rerouting—not disruption.

This represents a fundamental mindset shift: resilience is no longer something you restore; it is something you pre-build.

Physical Threats Are a Continuity Question, Not a Defense One

Orbital debris and geopolitical risks exist. But from a business perspective, the threat itself is not the issue. The issue is continuity.

Executives should focus on whether providers can maintain service if orbital assets must maneuver, pause operations, or temporarily exit certain regions of space. The technical details matter less than the business outcome.

Procurement and Contracts Become the Control Surface

Because orbital infrastructure cannot be physically accessed, contracts carry more weight than ever. Traditional SLAs assume repairability and inspection. Orbital compute breaks those assumptions.

Key Strategic Questions Include: Clear definitions of where processing, control, and storage occur Explicit governance and audit mechanisms Documented fallback paths to Earth-based systems Incident response expectations and communication timelines Risk allocation and force-majeure clarity

In this model, procurement is not a downstream function. It becomes central to risk management and operational feasibility.

A Broader Infrastructure Trend

Space-based data centers represent an extreme case of a broader shift. Compute is becoming more distributed, more energy-intensive, and less physically accessible. Edge environments, specialized accelerators, hyperscale cloud, and eventually orbital systems will coexist.

Organizations that can govern infrastructure they cannot touch will be better positioned than those that rely on physical proximity for control.

Part Two Takeaway

Space-based data centers are not a prediction—they are a forcing function. They push leaders to confront unclear jurisdiction, absolute reliance on secure transmission, resilience that must be engineered rather than repaired, and the need for contracts and governance strong enough to manage inaccessible infrastructure. Even if orbital compute remains niche, the disciplines required to evaluate it will define how organizations scale AI responsibly in the years ahead.

What Leaders Should Do Now

For most organizations, space‑based data centers are not an immediate destination—but they are an important signal. Leaders should focus now on understanding where infrastructure constraints are already shaping strategy, stress‑testing long‑term AI roadmaps against power and cooling realities, and ensuring governance models can adapt as new deployment options emerge. The priority is not migration, but preparation.

By Matthew Lovelace, Senior Manager