Custom Instructions and Memory
People who use ChatGPT most days and are tired of re-explaining themselves in every chat
Prerequisites: Module 301 (ChatGPT 101) or equivalent experience. Module 02 (How to Prompt) helps.
Last verified 2026-09-17
Learning objectives
By the end, participants can:
- Explain the difference between custom instructions (you write them) and memory (ChatGPT writes it)
- Write a compact about-me and style block that fits the limit (1,500 characters on Free and Go, 5,000 on Plus, Pro, Business, Enterprise and Edu)
- Read, correct and, where needed, delete the memory summary at Settings > Personalization > Memory
- Choose a temporary chat, personalised or not, for the right reasons
- Explain how workspace memory settings on Business and Enterprise change what applies
What's in this module
7 lessons · 5 exercises · ~7 min reading
Starts withThree layers, and who writes each~1 minExecutive summary
You will learn how ChatGPT personalises itself: custom instructions you write, memory it builds from your chats, temporary chats that use neither, and how the three interact. End state is a written about-me and style setup in your custom instructions, a memory summary you have read and corrected, and a rule for when to use a temporary chat. After this module you can stop repeating yourself, keep memory honest, and know exactly which of your words ChatGPT is carrying into a new conversation.
Sources
- OpenAI Help Center, "ChatGPT Custom Instructions", the setting, its path on each platform, and the character limits: https://help.openai.com/en/articles/8096356-chatgpt-custom-instructions
- OpenAI Help Center, "Memory FAQ", how the memory summary works, how to edit and delete it, and the legacy saved-memories system: https://help.openai.com/en/articles/8590148-memory-faq
- OpenAI Help Center, "Temporary Chat FAQ", personalised versus non-personalised temporary chats and the 30-day retention: https://help.openai.com/en/articles/8914046-temporary-chat-faq
- OpenAI Help Center, "Data Controls FAQ", the consumer training switch that decides whether this material improves the model: https://help.openai.com/en/articles/7730893-data-controls-faq
- OpenAI Help Center, "Managing workspace settings in ChatGPT Enterprise", where admins control memory and improved memory by role: https://help.openai.com/en/articles/8411955-managing-workspace-settings-in-chatgpt-enterprise
Prefer the live room?
This module also runs inside our in-person bootcamps and workshops in Copenhagen.
More in Track B
Set Up Claude Cowork
How Cowork works and how to set up the folder and context files that make it effective.
Sound Like You
How to capture your writing voice, taste, and hard rules in one compact file any AI can use.
Stop Writing Like AI
The patterns that make text read as machine-written, from negative parallelism to banned vocabulary.