Manage AI Context Transparency with Memory Sources
Audit and refine AI memory by viewing specific context sources.
Use the new 'Sources' view under personalized responses to audit which past chats or instructions ChatGPT used to generate an answer, allowing you to delete outdated context.
The Scenario
ChatGPT provides a recommendation based on a project you finished months ago, and the context is now irrelevant. You need to see which specific memory it used so you can clear it and get accurate results for your current work.
Before & after
Users had to guess why ChatGPT gave a specific personalized answer or browse through 'Saved Memories' to find and delete old context. This took 5-10 minutes of trial and error.
Identify the 'Sources' below a response to see which past conversations or manual memories were used. This allows you to correct or delete specific context in 30 seconds.
The Prompt
Review my memory summary and tell me which past conversations you are using to personalize your answers for [PROJECT_OR_TOPIC_NAME]. Delete the memory if it's outdated.
The improved memory feature shows exactly which past chats or instructions influenced an answer. It provides a 'Memory summary' and 'Sources' links below responses for full transparency and control.
Source
ChatGPT Business - Release Notes | OpenAI Help Center"View Sources below personalized responses to see relevant context from memories, past chats, and custom instructions."