When Bitcoin Service Changes Terms Unilaterally

Unilateral Terms Changes by Custody Service Providers

This memo is published by CustodyStress, an independent Bitcoin custody stress test that produces reference documents for individuals, families, and professionals.

The Nature of Unilateral Change

A holder uses a service related to their bitcoin custody. Perhaps it is a wallet application, a hardware provider's companion software, a backup service, or an exchange. The terms under which the service operates were acceptable when the holder started using it. Time passes. The service changes its terms. The holder receives notice, or does not. When a bitcoin service changes terms unilaterally, the holder's options narrow to acceptance or exit. Negotiation does not exist.

This analysis covers the asymmetric nature of service relationships in bitcoin custody. Holders depend on services. Services do not depend on individual holders. This imbalance means terms can shift at the service's discretion. The holder who built their custody around specific service features faces disruption when those features change or disappear. The dependency was always there. The term change makes it visible.

---

The Nature of Unilateral Change

Service terms are contracts, but they are not negotiated contracts. The holder clicks "I agree" to terms drafted entirely by the service. No back-and-forth occurs. No provisions are modified based on the holder's needs. The terms exist as presented or not at all. This is standard for consumer services across industries. Bitcoin-related services are no different.

Terms of service typically include provisions allowing future changes. Buried in the document the holder agreed to is language permitting the service to modify the agreement. The modification mechanism varies. Some services require notice. Some services consider continued use as acceptance. Some services claim the right to change terms at any time for any reason. The holder agreed to this when they started using the service, whether they read the provision or not.

Change arrives as announcement, not request. The service decides what the new terms are. The service communicates them through whatever channel it chooses. The holder learns of the change through an email, a pop-up notification, an updated document on a website, or perhaps not at all. The holder's input into the new terms is zero. The terms were set before the holder knew they were being considered.

This dynamic is not adversarial from the service's perspective. Services change terms for many reasons: regulatory compliance, business model shifts, risk management, feature changes, liability reduction. The reasons may be legitimate. They may even benefit users as a group. But the individual holder has no voice in the process. Their interests are considered only in aggregate, if at all.

---

The Dependency Asymmetry

The holder needs the service more than the service needs the holder. This asymmetry defines the relationship. A service with thousands or millions of users does not depend on any single user's continued participation. A holder whose custody depends on a specific service depends on that service entirely.

Exit costs for the holder may be substantial. Leaving a service may require migrating keys, learning new software, reconfiguring backup procedures, and updating documentation. These costs are paid entirely by the holder. The service incurs no cost from a single user's departure. At scale, even many departures may not significantly affect the service's operations.

The asymmetry increases over time as the holder becomes more embedded. Initial use creates familiarity. Documentation references the service. Processes build around it. Other people involved in the custody setup learn the service's interface. The longer the holder uses the service, the higher the switching cost becomes. The service's leverage grows as the holder's investment deepens.

Services may be aware of this asymmetry and factor it into decisions. They may know that holders are unlikely to leave over moderate term changes. The cost of switching exceeds the cost of accepting unfavorable new terms. This calculation, made implicitly by thousands of users, allows services to shift terms further than they could if switching were frictionless.

---

Categories of Term Change

Term changes affect holders in different ways depending on what changes. Some changes are cosmetic. Others alter fundamental aspects of the service relationship. The impact depends on what the holder relied upon and whether that reliance survives the change.

Fee structures can change. A service that was free may introduce charges. A service with one fee schedule may adopt another. The holder's cost of using the service shifts. If the holder's budget or expectations assumed the original fee structure, the change forces recalculation. Continued use at the new price may not make sense, but leaving incurs its own costs.

Feature availability can change. A backup feature the holder used may be discontinued. A security option may be removed or altered. An interface the holder mastered may be replaced. The holder's workflow assumed certain capabilities. When capabilities change, the workflow breaks. Adaptation requires effort the holder did not plan to spend.

Data handling terms can change. How the service stores, accesses, or shares information may shift. Privacy expectations based on original terms may no longer apply. The holder may have chosen the service partly for its data practices. If those practices change, the original choice no longer reflects current reality.

Jurisdictional or regulatory terms can change. A service may cease operating in certain regions. It may add compliance requirements. It may impose identity verification where none existed before. Holders affected by these changes may find themselves suddenly outside the service's acceptable user base, despite having used it without issue previously.

---

Notice and Its Limits

Services typically provide notice of term changes, but notice does not equal comprehension. An email subject line may not convey the importance of the change. A long document may obscure significant modifications among routine updates. The holder may technically receive notice without actually understanding what changed or why it matters.

Notice periods vary. Some services announce changes weeks in advance. Others provide days or less. The holder's ability to respond—to evaluate alternatives, migrate if necessary, adjust processes—depends on having enough time. Short notice periods pressure holders toward accepting changes they have not fully evaluated.

Notice may not reach the holder at all. Email filters may catch the message. The holder may have provided an old email address. The notification may arrive during a period when the holder is not checking that channel. The service fulfilled its notice obligation by sending the communication. Whether the holder received and understood it is the holder's problem.

Even clear notice leaves the holder in a reactive position. They learn what the service decided. They do not learn what alternatives were considered, what tradeoffs were weighed, or what might be negotiable. The notice announces a conclusion. It does not invite dialogue. The holder's response options are limited to accepting the conclusion or leaving.

---

The Exit Problem

When term changes are unacceptable, the holder's remedy is exit. Stop using the service. Find an alternative. But exit may not be straightforward, especially for services embedded in custody infrastructure.

Some services hold custody or partial control over bitcoin. Exiting these services requires moving the bitcoin elsewhere. The service must cooperate in this movement. If the term changes include restrictions on withdrawals, exit itself becomes harder. The holder discovers that the ability to leave was also subject to the service's terms.

Some services provide functionality that is hard to replicate. A particular backup solution, a specific security feature, a unique interface—if no equivalent exists elsewhere, exit means losing capability. The holder may accept unfavorable term changes because the alternative is worse: doing without the functionality entirely.

Some services are interconnected with other parts of the holder's setup. The hardware wallet works with specific software. The software expects specific services. Pulling out one piece may require reconfiguring others. The holder faces cascading changes, each with its own learning curve and risk of error. Exit from one service becomes a project affecting the entire custody arrangement.

Exit also has timing constraints. The holder may need the service during the period when they are trying to leave it. A backup service is needed most when something goes wrong. A wallet interface is needed to move funds. The holder cannot fully exit until they have successfully transitioned, and transitioning may require using the service under the new terms they are trying to avoid.

---

Trust Degradation

Unilateral term changes erode the trust relationship between holder and service. Each change reminds the holder that the relationship is not between equals. Each change demonstrates that past terms provide no guarantee of future terms. The holder begins to question whether any current terms are stable.

Repeated changes compound the effect. A service that changed terms once might do so again. A service that changed terms in ways the holder disliked might make future changes the holder dislikes even more. The trajectory is unclear but trending in a direction the holder cannot control. Uncertainty about future terms becomes a persistent feature of the relationship.

Trust degradation affects how holders use the service. They may reduce reliance on features that seem most likely to change. They may maintain backup options they would not otherwise need. They may invest less in learning the service's capabilities, knowing those capabilities may not persist. The holder hedges against a service they once trusted fully.

The service may not notice this trust degradation in any single user. Aggregate metrics may remain stable even as individual users become more guarded. The erosion is invisible until it reaches a threshold where users begin actively leaving. By then, the trust deficit has accumulated over many changes and cannot be quickly reversed.

---

Documentation Instability

Custody documentation that references services becomes unstable when terms change. Instructions written under old terms may no longer apply. References to features that no longer exist confuse rather than clarify. The documentation describes a service that has since shifted into something different.

Holders rarely update documentation after service term changes. The change was not their choice. Updating documentation to reflect it requires effort they did not ask to spend. The documentation drifts out of sync with reality. Future users of that documentation—heirs, executors, backup parties—encounter instructions that do not match what they see.

Services rarely provide migration guidance for documentation. Their term change announcements address the service itself, not the holder's personal documentation ecosystem. The holder is left to figure out which of their notes, instructions, and references need updating. Some will be missed. The documentation becomes partially obsolete without any clear indication of which parts are affected.

Over time, documentation may reference services that no longer exist at all. A service that changed terms repeatedly may eventually shut down or transform beyond recognition. The holder's historical documentation becomes an artifact of a different era. Anyone trying to follow it encounters a world that no longer exists.

---

The Structural Position

The ability of services to change terms unilaterally is not a malfunction. It is how service relationships are structured. Consumer services operate at scale. Individual negotiation is not economically viable. Terms must be standardized. Standardized terms must be changeable as circumstances change. The holder's position in this structure is always subordinate.

This structure exists across all services, not just bitcoin-related ones. Email providers, cloud storage, social networks—all operate the same way. Bitcoin holders are not uniquely disadvantaged. But bitcoin custody may involve higher stakes than typical consumer services. The asymmetry that is merely annoying for email becomes significant when the service affects access to substantial value.

Self-custody bitcoin avoids some service dependencies but not all. A holder who controls their own keys still may depend on software, hardware manufacturers, blockchain infrastructure providers, and information sources. Pure independence is difficult to achieve. Most holders accept some service relationships as part of their custody setup. Those relationships carry the asymmetry this memo describes.

Understanding the structure does not change it. The holder cannot negotiate their way to equal footing. They cannot compel services to maintain terms indefinitely. They can only recognize the asymmetry and account for it in their planning. The dependency exists. The question is whether the holder acknowledges it.

---

Assessment

When a bitcoin service changes terms unilaterally, the holder faces the structural reality of asymmetric dependency. Services set terms. Holders accept them or leave. Negotiation is not available. The holder's leverage is limited to exit, and exit carries its own costs that the service does not share.

Term changes can affect fees, features, data handling, and jurisdictional availability. Notice provides information but not influence. The holder learns what changed, not whether it can be changed back. Exit is the remedy, but exit requires time, effort, and alternatives that may not exist or may themselves be subject to future term changes.

The asymmetry is structural, not incidental. Services operate at scale with standardized terms that must be modifiable. Individual holders are interchangeable from the service's perspective. This is how consumer services work. Bitcoin custody that depends on such services inherits this dynamic. The dependency may be necessary, but it is not equal.

---

System Context

Examining Bitcoin Custody Under Stress

DCA Bitcoin Storage Strategy: Modeled Custody Drift, Timing Boundaries, and Inheritance Effects

Bitcoin Custody Reference for Spouse

← Return to CustodyStress