Issue
AI Memory Updated Without Asking
AI memory, saved context, preference, or durable state is updated without the user clearly asking for or approving that update.
What This Looks Like
The AI saves, updates, or changes memory without the user clearly asking for that durable update. A preference, fact, instruction, project detail, role assumption, or context item may become persistent even though the user intended it to apply only to the current task or conversation.
Why It Matters
Memory changes can affect future outputs. If memory updates without clear approval, local context can become durable behavior, and users may not know why later responses changed. This creates problems around consent, scope, correction, and trust in persistent AI behavior.
Structural Signal
A temporary or local context item crosses into persistent memory without a declared authority boundary. The issue is not simply that the AI remembered something; it is that memory state changed without clear user authorization.
Common Triggers
- The system treats repeated statements as memory candidates without explicit approval
- Local task context is saved as a durable preference
- The user corrects something for one task and the correction becomes global
- Memory update behavior is not visible at the moment it happens
- The system does not separate temporary context from persistent memory
- The user cannot tell which statements may update memory
When to Use This Issue
Use this Issue when AI memory or durable context is updated without the user clearly requesting, approving, or understanding that persistent change.
When Not to Use This Issue
Do not use this Issue when memory is merely used incorrectly or not used at all. Do not use it when the user explicitly asked the system to remember the information.
Category
Primary Pattern
Declared Patterns
Missing Authority
A structural condition where a region, action, state, or decision path exists without a declared governing authority.
Persistence Instability
A structural condition where persisted state cannot be stored and restored into an equivalent structure without alteration.
Boundary Leakage
A structural condition where effects, permissions, state, data, or authority cross a declared boundary without an authorized exception.
Derived Primary Lenses
Absence Lens
Detects structurally required elements that are missing from the observed structure.
Authority Overlay Lens
Maps declared authority hierarchies onto observed structure to detect absence, override, or conflict.
Isolation Boundary Lens
Evaluates whether structural constraints, effects, and regions remain contained within declared boundaries.
Propagation Lens
Traces how structural declarations, effects, or state changes propagate across boundaries or stages.
Reference Stability Lens
Evaluates whether structural references, identifiers, nodes, and edges remain consistent across execution cycles or comparable states.
Derived Secondary Lenses
Boundary Compliance Lens
Evaluates observed structure against declared boundary posture, including allow, block, and exception rules.
Conflict Lens
Detects mutually incompatible constraints, claims, states, or declarations that cannot be simultaneously satisfied.
Determinism Lens
Evaluates whether identical structural inputs produce equivalent structural outputs across repeated executions.
Reconciliation Lens
Evaluates whether structural changes align with declared authority updates, version changes, or reconciliation rules.
Search Intents
- AI memory updated without asking
- ChatGPT memory changed without permission
- AI saved memory without approval
- memory updated unexpectedly
- AI remembered something I did not ask it to
- persistent memory changed without consent