Problem
An assistant may use details from earlier conversations without explaining where they came from. People need to see what it saved and control what it uses.
Solution
Show a short notice when the assistant saves a detail for later. State what was saved and let people remove it immediately.
When saved information shapes an answer, make that use visible. Provide a readable list of saved items with dates and controls to edit or delete each one.
Keep memories separate by project or workspace when information should not cross those boundaries. Let people choose where memory is enabled and explain how long saved information is kept.
Use when
People return to related work and would benefit from keeping useful context across conversations.
Don’t use when
Use is occasional, anonymous, or shared across people on one account. Memory built from a shared login models a person who doesn’t exist and produces confidently wrong personalisation.
Trade-offs
Memory reduces repeated explanations but can make the same prompt produce different answers. Frequent notices add clutter. Saved details can become outdated or appear in the wrong context, so review and deletion controls matter.
Checklist
- Is there a visible moment when something is written to memory?
- Can that write be undone at the moment it happens?
- Is every stored item listed in plain language, with a date, individually deletable?
- When an answer used a memory, can the viewer tell?
- Is memory scoped by workspace or project, or global to the account?
- What’s the default, and was it opted into?
- What does deleting a memory actually delete, and by when?
- Can the viewer see what would change if they turned memory off?
- Does a shared or team account build one merged profile?
- Is memory carried into exports, shared conversations, and screenshots?
Compare
ChatGPT keeps an editable list of stored details and marks the turn where a memory is written. The write is visible at the one moment a viewer can still judge it. Gemini ties personalisation to past chats as a whole rather than to extracted facts, and answers the recall question on request, which trades per-item precision for a simpler mental model. Claude leans on projects, so the persistent material is a body of documents the viewer assembled deliberately rather than inferences drawn from conversation. Notion keeps the equivalent state in the workspace itself, where what the assistant knows is whatever the viewer already wrote down and has permission to read.
Related
Conversation history stores the exchanges themselves. Custom instructions contain guidance people write deliberately. Scoped context controls which workspace information applies.
First-run state explains data use before people begin. Citation chip identifies external sources rather than saved personal details.