Product design question

How would you improve ChatGPT's memory feature for power users?

Practice this question out loud. An AI interviewer asks it, follows up like a real interviewer would, and scores your answer. Type or speak.

What this question tests

Product design for a feature that requires balancing personalization value against control and privacy concerns for advanced users.

Clarify the power user's pain point: memory currently feels like a black box, they cannot see, edit, or organize what ChatGPT has learned about them.
Propose a memory dashboard: a browsable, searchable view of stored memories, grouped by topic, so users can audit and prune what is retained.
Add selective memory scopes, letting a user create separate memory contexts for work versus personal use, so memory does not bleed across unrelated projects.
Improve transparency in-conversation, showing a subtle indicator when a response used a stored memory, so behavior does not feel unexplained.
Define success: reduction in memory-related complaints or deletions, and increase in memory-dependent feature usage among power users specifically.

Leads with control and visibility, a searchable memory dashboard, since power users specifically want to audit and prune, not just benefit passively.
Proposes scoped memory contexts, like separating work and personal, which addresses a real complaint about memory bleeding across unrelated conversations.
Adds an in-conversation transparency signal so a surprising response is traceable to a specific stored memory, building trust in the feature.
Names a concrete metric, like reduction in manual memory deletions after launch, as a proxy for the feature working as intended rather than annoying users.

Adding more automatic memory without addressing the core complaint of lacking visibility and control.
Ignoring the risk of memory bleeding between unrelated contexts, like work and personal use.

Chapters of the AI PM course that teach what this question tests.