Memory and Recall is a UX design pattern that enables the AI to store and reuse information from past interactions such as user preferences, feedback, goals or task history to provide continuity, reduce repetition, and enhance personalization.

Memory used to access information can be Ephemeral (short-term within a session) or Persistent (long-term across sessions) and may include conversational context, behavioural signals, or explicit inputs.

Why is it important

1. Enhances personalization by remembering past choices or preferences makes the user experience more relevant and efficient.

2. Reduces user burden by avoiding repeated input requests especially in multi-step or long-form tasks

3. Builds trust by reinforcing the feeling of continuity and partnership aligning with user’s mental models of how an assistant should behave.

4. Supports complex tasks like longitudinal workflows (e.g., project planning, learning journeys) by referencing or building on past progress.

   
   

   
   

How to use this pattern

1. Define the User Context and choose Memory type

   Choose memory type like Ephemeral or Persistent or both based upon use case. A shopping assistant might track interactions in real time without needing to persist data for future sessions whereas personal assistants need long-term memory for personalization.
   

2. Communicate Memories transparently

   Clearly communicate what’s being remembered and offer visibility (e.g., “Your preferences are saved here”).
   

3. Use Memory intelligently in User Interactions

   Recall information contextually (e.g., “Last time you preferred a lighter tone. Should I continue with that?”). Avoid over-reliance, fallback gracefully when memory fails or is reset.
   

4. Give Users Control

   Let users view, edit or delete stored memory. Make “delete memories” an accessible action. E.g. In ChatGPT you can view, update, or delete these memories anytime
   

Further Reading

Memory and States in LLM Applications Memory and new controls for ChatGPT