AI Memory
Persistent Context Users Can See and Control
Instead of re-explaining context every conversation, AI maintains a visible, editable memory. Users see what the AI "knows" about them and can correct, delete, or add memories. Memory becomes a first-class UI element, not a hidden system prompt.
Framing
The problem
Persistent context is useful but risky when it is hidden from users.
The pattern
Expose memory as a visible, editable user-facing artifact.
Why chat breaks here
Hidden memory silently shapes outputs without inspection or correction.
Risks
Stored memories can become creepy, stale, or over-personalized.
Avoid when
The product does not need cross-session continuity or personalization.
Use when
Continuity across sessions creates real user value — and silent memory would feel intrusive.
DOPE evaluation
- Directability
- Context is curated per card — switch, inline-edit, delete, or add a memory — not wiped or kept wholesale
- Observability
- Every memory is a card with its provenance ("Source: 12 conversations"), a staleness dot, and an on/off state — plus a running "8 of 9 active" count, not a hidden system prompt
- Predictability
- Each toggle maps to a specific output change — flip "avoids buzzwords" off and the headline goes hype, flip the format preference on and prose becomes bullets
- Explainability
- The response preview names which memories shaped it as chips — active ones lit, disabled ones struck through — so influence is attributed, not assumed
In the wild
- ChatGPT Memory (OpenAI) — Settings → Personalization → Manage memories. Apr 2025 added memory-source visibility ("which memory was used"); Oct 2025 added auto-management with prioritize/deprioritize. Flagship example.
- Claude Memory (Anthropic) — Settings → Capabilities → "View and edit memory" (Sept 2025). Pencil icon in lower-left of memory summary lets users add custom instructions. Stored as plain Markdown for transparency. Pro/Max only.
- Claude Code auto-memory (Anthropic) — v2.1.59+ writes memory to CLAUDE.md files Claude shows you and lets you edit. Tier hierarchy (~/.claude, project, subdirectory) is unusually transparent — the file IS the memory.
FAQ
When should I use the AI Memory pattern?
Continuity across sessions creates real user value — and silent memory would feel intrusive.
When should I avoid the AI Memory pattern?
The product does not need cross-session continuity or personalization.
What problem does AI Memory solve?
Persistent context is useful but risky when it is hidden from users.
Why is chat the wrong fit for this?
Hidden memory silently shapes outputs without inspection or correction.
Related patterns
- Often paired with: Constraint Rules Engine — Memory holds preferences; rules hold hard limits — both visible and editable.
- Often paired with: Confidence Signals — Show what is remembered alongside how confident the model is in using it.
- Alternative to: Autonomy Dial — Persisted preferences vs an explicit dial for delegation level.