What does ChatGPT remember about me?

If ChatGPT has ever brought up something you told it weeks ago, you've seen Memory at work. Here's how to see what it's using about you, fix what's wrong, and remove what you don't want it to keep, based on OpenAI's own help page.

OpenAI says memory features and controls vary by plan, region, platform and workspace settings. If a setting below isn't in your account, it may not be available to you yet. Treat menu names as signposts.

Where ChatGPT's memory comes from

When Memory is on, ChatGPT can draw on more than one source. Depending on your plan and region, OpenAI lists:

Past chats (when Reference chat history is on)
Saved memories, details you asked it to remember or that it saved as useful context
Custom instructions you wrote yourself
Saved files in Library or Space
Connected apps, such as Gmail

OpenAI also notes that memory doesn't keep every detail from every chat. ChatGPT picks what seems relevant, and what it remembers can change over time.

Three ways to see what it remembers

Check the memory summary. Go to Settings > Personalization > Memory. Depending on your account you'll see a memory summary, a list of saved memories, or both. OpenAI describes the summary as a high-level view: it doesn't necessarily include every detail ChatGPT can use.
Look at Sources under an answer. When ChatGPT uses your personal context in a reply, Sources may appear below the response, pointing to the past chat, saved memory, custom instruction, file or email that shaped it. OpenAI says Sources may not show every factor, and they aren't included when you share a chat by link.
Just ask. OpenAI says you can ask ChatGPT what it remembers about you, for example: "What do you remember about me?"

Tired of re-checking these settings? The free one-page Mini Context Card gets your context into any AI in about 10 minutes. Get the free card

Fix something that's wrong

You can tell ChatGPT that something it remembers is wrong or out of date. Where the controls are available, you can also edit the memory summary directly: type a correction in the text field, or highlight the text and give a correction. Saved memories can be corrected or deleted too.

"Don't mention this again" vs. deleting

These sound alike but do different things:

Don't mention this again (when the option shows) cuts down future references to that information. It doesn't delete the chat, file, email or other place it came from.
Delete removes the remembered item itself. The menu (•••) on the summary also has Delete and turn off memory, which deletes what the summary shows and turns Memory off. It doesn't delete your past chats.

How to fully remove something

This is where most people get caught out. OpenAI says deleting a chat doesn't necessarily delete a separate saved memory made from it, and deleting a saved memory doesn't remove mentions of it from past chats. To really remove something, clear it from every place it lives:

the memory summary and your saved memories
the chat where you first shared it, including archived chats
any saved files in Library or Space that contain it
any connected app that holds it (disconnect the app if you no longer want ChatGPT to reach it)

OpenAI says deletions can take time to take effect and that it may keep logs of deleted saved memories for up to 30 days for safety and debugging. If you turn off Reference chat history, what ChatGPT remembered from past chats is scheduled for deletion within 30 days, but the chats themselves stay in your history unless you delete them.

Chat without memory

Before you start a temporary chat, you can choose Unpersonalized so it doesn't use your memories, custom instructions or plugins. You can't change that choice once the chat starts. Temporary chats don't create or update memories. If you save one, it becomes a regular chat and follows your normal memory settings.

Keep the important part where you can see it

Memory is something ChatGPT writes and updates on its own, and even its summary is only a partial view. For the details you always want it to have (your role, your projects, how you like answers), OpenAI's own distinction helps: Custom Instructions are direct guidance you write, while memory is what builds up across chats. Put the essentials in your instructions, and keep the longer version in a personal context file you own. It works in ChatGPT, Claude, Gemini and Grok, and you can paste it into any new chat with this line:

Here is my personal context file. Read it and use it for everything in this chat. You don't need to summarize it back to me. Just reply "Got it" and wait for my first request.

[paste your context file here]

Source: OpenAI Help Center, Memory in ChatGPT.

Related guides

More guides on how AI assistants remember you:

Start free in 10 minutes. Grab the free Mini Context Card: a one-page fill-in sheet to write your first context file.
Get the free card Full Context Kit ($19 suggested) The full kit includes a fill-in template, a worked example, a step-by-step setup guide for ChatGPT, Claude, Gemini and Grok, plus interview and monthly-refresh prompts.