Analysis by
Founder, Lex Wire Journal • Technology, Governance & Sovereignty Strategist
Digital Sovereignty Begins With the Ability to Act
Digital systems give individuals and institutions extraordinary access to information, infrastructure, communication, markets, software, and increasingly intelligence. But access alone does not tell us how much control the user actually possesses.
A person can have access to a platform while remaining dependent on the platform operator for identity, permissions, functionality, pricing, continuity, and the rules governing participation. An organization can have access to its data while relying on another party’s infrastructure to make that information useful. A law firm can have access to powerful artificial intelligence while remaining responsible for decisions made through systems it does not fully control.
That distinction moves the sovereignty inquiry beyond ownership and toward a more fundamental question: can the individual or institution meaningfully act when permission, infrastructure, or cooperation from another party is no longer assumed?
The Bottom Line
Digital sovereignty is not simply access to technology, ownership of data, or freedom from outside providers. It is the preservation of meaningful agency within technological relationships. The central question is whether individuals and institutions retain enough control, portability, verification, choice, and ability to exit to act on their own behalf when circumstances change.
This is not a settled legal definition of digital sovereignty. The term is used differently across cybersecurity, public policy, national strategy, data governance, and technology debates. Within the Sovereignty & Law framework, the concept is narrower and more practical: sovereignty is examined through the degree of agency retained within relationships of technological dependence.
“Access tells you what a system currently allows you to do. Agency tells you what you remain capable of doing when the conditions change.”
Jeff Howell, Sovereignty & Law
Access and Agency Are Not the Same Thing
Access is often conditional.
A platform can grant access through an account. A software company can grant access through a license. A cloud provider can grant access through a subscription. An AI company can provide access to a model through an interface or application programming interface.
In each case, access can be valuable without necessarily conferring meaningful control over the underlying system.
Terms can change. Prices can change. Features can disappear. Accounts can be limited. Integrations can break. Systems can be discontinued. Providers can be acquired. Business priorities can shift. Regulation can alter availability. The user may retain contractual remedies, but the technological capability itself can still remain dependent on another party’s continued operation of the system.
Agency asks a different question: what can the user still do if access changes?
Can the information be moved? Can another provider be substituted? Can critical processes continue? Can the system’s claims be independently verified? Can the user retain identity, records, institutional knowledge, or operational capability without remaining permanently tied to one intermediary?
From Ownership and Control to Agency
The first three analyses in Sovereignty & Law approached this problem from different directions.
The Age of Digital Dependence examined what happens when technological reliance also transfers meaningful control to the systems and providers on which individuals and institutions depend.
Who Controls the Intelligence? extended that question into enterprise AI by distinguishing control over information from control over the systems that transform information into institutional intelligence.
Your Data Is Not Your Intelligence then examined how formal ownership can diverge from the practical technological capabilities required to make information useful.
Agency connects those ideas.
Ownership matters because rights affect what an individual or institution is entitled to possess, use, transfer, or control. Technological control matters because architecture determines what actions are possible in practice. Agency describes the resulting capacity to make meaningful choices and act upon them.
Ownership establishes rights. Control establishes capability. Agency determines whether those rights and capabilities can actually be exercised as meaningful choice.
Agency Does Not Require Complete Independence
A useful sovereignty framework cannot depend on the idea that every individual or institution should become technologically self-sufficient.
Modern economic and professional life is built on specialization, shared infrastructure, contractual relationships, networks, institutions, vendors, and intermediaries. Eliminating dependence entirely would often be impossible, inefficient, or undesirable.
Sovereignty therefore does not require rejecting centralized systems, cloud computing, commercial software, or third-party AI. Those architectures can provide significant benefits in performance, security, expertise, coordination, scalability, and cost.
The issue is whether dependence has become so structurally important that meaningful choice disappears.
“Sovereignty is not the elimination of dependence. It is the preservation of agency within dependence.”
Jeff Howell, Sovereignty & Law
Human Agency Is Already a Recognized AI Governance Principle
The importance of agency is not limited to the sovereignty framework developed here. Major AI governance frameworks already recognize the need to preserve meaningful human involvement as artificial intelligence becomes more capable.
The OECD AI Principles, first adopted in 2019 and updated in 2024, identify human rights and democratic values as core principles for trustworthy AI. The framework specifically includes the autonomy of individuals and calls for safeguards that preserve human agency and oversight.
NIST approaches the problem from a risk-management perspective. Its AI Risk Management Framework organizes AI governance around the functions Govern, Map, Measure, and Manage, with governance operating across the AI lifecycle. NIST emphasizes that organizations must continually evaluate whether AI systems meet their intended purposes and how identified risks should be managed.
These frameworks do not define technological sovereignty in the way this series does. But they reinforce a related principle: deploying an AI system does not eliminate the need for human judgment, organizational accountability, or the ability to intervene when technology produces unacceptable outcomes.
Lawyers Provide a Clear Example of Why Agency Matters
The legal profession makes the distinction between technological capability and human agency particularly clear.
Artificial intelligence can assist with research, drafting, document analysis, knowledge retrieval, and other legal tasks. But professional responsibility remains with the lawyer.
ABA Formal Opinion 512 states that lawyers using generative AI remain subject to existing ethical duties, including competence, protection of client information, communication, supervision, candor, and reasonable fees.
The ABA’s treatment of generative AI illustrates a broader principle. Technology can extend a professional’s capabilities without transferring the professional’s ultimate responsibility. The lawyer must still understand enough about the system, its risks, and its limitations to exercise independent judgment.
That is agency in practice. The lawyer can use technology without surrendering the obligation or capacity to make the professional decision.
“The purpose of powerful technology should be to expand human capability without quietly replacing human agency.”
Jeff Howell, Sovereignty & Law
The Architecture of a System Determines Where Agency Resides
Agency is not determined only by legal rights or user intentions. It can also be shaped by system architecture.
Consider two systems that offer similar functionality. One allows a user to export information in usable formats, connect alternative tools, preserve identity independently, verify important records, and migrate to another provider. The other provides equivalent functionality but makes each capability dependent on continued participation in a single closed environment.
Both systems may provide access. They do not necessarily provide the same degree of agency.
This is why technological sovereignty is ultimately an architectural question as much as a philosophical one. Permissions, interfaces, standards, portability, interoperability, identity systems, custody arrangements, verification mechanisms, and exit options can all affect the practical ability of users to act independently.
Choice
Can the user meaningfully choose among alternatives rather than merely accept the available terms?
Control
Which important actions can the user initiate, authorize, limit, or stop?
Portability
Can information, identity, configurations, or capabilities move to another environment?
Verification
Can important information, transactions, or system behavior be independently confirmed?
Continuity
Can essential activity continue if a particular provider, platform, or relationship changes?
Exit
Can the user leave without surrendering information, identity, accumulated capability, or meaningful participation?
None of these characteristics produces absolute sovereignty. Together, they help reveal where meaningful agency resides within a technological relationship.
Permission Can Be Valuable Without Being Sovereign
Permission-based systems are not inherently problematic.
Institutions routinely control access for legitimate reasons involving security, safety, confidentiality, quality, fraud prevention, regulatory compliance, and contractual responsibility. A system that permits every user to do anything without restriction would often be unusable or dangerous.
Sovereignty therefore should not be confused with unrestricted action.
The relevant distinction is between legitimate constraints within a relationship and structural dependency that leaves the user without meaningful alternatives.
A system can impose rules while preserving agency. An institution can rely on vendors while retaining meaningful control. A person can participate in networks governed by others while maintaining ownership of important assets or identity. Sovereignty exists by degree rather than as a binary condition.
Digital Sovereignty Is Better Understood as a Continuum
Thinking about sovereignty as a continuum avoids two extremes.
At one extreme is complete technological dependence, where important information, identity, capabilities, and choices are controlled by systems the user cannot meaningfully inspect, replace, or leave.
At the other is complete technological self-sufficiency, an ideal that is neither realistic nor necessarily desirable for most people or organizations.
Between those extremes are countless architectures that distribute control differently.
A sovereign approach asks which capabilities are important enough to preserve, which dependencies are acceptable, which risks can be delegated, which claims should be independently verifiable, and where the ability to exit must remain real.
Sovereignty is not a destination where dependence disappears. It is a design principle for deciding where agency must remain.
The more consequential a technological dependency becomes, the more important it is to understand what choices remain available when the interests of the user and the system operator no longer align.
Agency Leads Directly to the Question of Exit
Agency becomes easiest to see when a relationship changes.
When a system works as expected and the interests of the user and provider remain aligned, dependency can be almost invisible. The structural question emerges when prices rise, policies change, access is restricted, functionality disappears, a provider fails, or the user simply decides that another system better serves its needs.
At that moment, sovereignty becomes measurable in a practical way: can the user leave?
That question makes exit more than a commercial convenience. It becomes one of the clearest tests of whether technological choice remains meaningful.
“Agency is revealed not only by what a system allows you to do while you remain inside it, but by what you retain the power to do when you choose to leave.”
Jeff Howell, Sovereignty & Law
The next Sovereignty & Law analysis will examine the structural side of that question: The Architecture of Dependence: How Technology Quietly Transfers Control .
This article is part of Sovereignty & Law, a Lex Wire Journal editorial initiative examining how technology is changing the relationship between law, ownership, trust, agency, and power.
