If you have ever wondered what ChatGPT actually knows about you, OpenAI just gave you a way to look. There is now a memory summary, a single screen that lists the details ChatGPT is carrying about you, when it last changed, and a box to fix or remove anything that is wrong.
For a business owner who put a half-baked detail into ChatGPT six months ago and forgot about it, this is a two-minute audit worth doing this week.
What changed
OpenAI rebuilt how ChatGPT remembers things. The update is called "Memory that stays more up to date," dated June 4, 2026. In OpenAI's words: "We've upgraded memory so ChatGPT can better keep your context up to date, helping responses stay more relevant."
Two things are different now.
First, memory updates on its own. The old version waited for you to manage it by hand. The new one keeps track automatically: "Memories are now updated automatically, with ChatGPT keeping track of the details it determines are most important so it can continue building on the context you've already shared."
Second, you can see it. OpenAI says, "You can see the information that ChatGPT remembers in the memory summary." At the top of that summary, it shows you how recent it is. OpenAI's example reads like "2 hours ago."
A second update on June 12, 2026 added more control, including a way to wipe the whole thing and shut it off.
Paid Plus and Pro accounts get twice as much memory capacity as free plans under the rebuild.
Source: OpenAI Help Center, 2026Why an owner should care
You wear every hat. You do not have an IT person to sort this out for you. So the value here is plain: ChatGPT has been quietly building a picture of you and your work, and now you can read that picture and correct it.
Three reasons this is worth your attention.
Wrong details bend every answer. Say ChatGPT still thinks you run a three-person shop when you now have eight people. Or it holds an old service you stopped offering. Every answer it gives you leans on that stale detail. Reading the summary and fixing one wrong thing is the cheapest way to make the tool more reliable.
Privacy is now a setting, not a guess. Before you paste a client list or a draft contract into ChatGPT, you can decide what it keeps. There is a "Delete and turn off memory" option, a Temporary Chat mode that remembers nothing, and an opt-out on training. More on each below.
Knowing the limits is its own skill. OpenAI is honest that the summary does not show everything. "While the memory summary should capture the most important details, it will not include everything that ChatGPT remembers based on your chats." Understanding the difference between what you can see and the broader picture it holds is the kind of judgment that separates someone who uses AI nervously from someone who uses it well.
Understanding the difference between what you can see and the broader picture it holds is the kind of judgment that separates someone who uses AI nervously from someone who uses it well.
That last point is the heart of The 7 Levels of AI Proficiency, the measurable standard we built at LaunchReady for AI-capable companies. The early levels are about getting comfortable with a tool. The higher levels are about keeping a human hand on it: checking what it holds, correcting it, and deciding what it is allowed to do. Reading your own memory summary is a small rep in exactly that habit.
How to find it and fix it
You do not need any technical setup. Here is the path.
Read what it knows
Open ChatGPT and find your memory summary. It lists the details ChatGPT is carrying and stamps the top with when it last updated.
Read it like you would read a note someone else wrote about you. Look for anything outdated, anything that was only true for one project, anything that was never quite right.
Correct it two ways
OpenAI gives you two edit paths.
You can type the fix. "Type what you want changed into the text box at the bottom of the memory summary and it will update accordingly."
Or you can highlight a specific line. "You can highlight any text in the memory summary to make a specific correction, or select 'Don't mention this again.'"
One thing to understand about that second option. "Don't mention this again" tells ChatGPT to stop bringing something up. It does not erase it. OpenAI is direct about this: it "helps reduce unwanted references, but it does not delete the information."
Delete it for real
To fully remove something, OpenAI says you have to delete it everywhere it lives: "every source where it appears, including past chats, archived chats, files, the memory summary, and disconnect any connected apps that may contain that information."
If you want a clean slate, there is now a single switch. From the three-dot menu, choose "Delete and turn off memory." OpenAI notes one thing to keep in mind: "This does not delete your past chats." And if you flip memory back on later, it can rebuild from those old chats. "If you turn memory back on later, ChatGPT may create new memories from chats that remain in your chat history, including older chats."
The delete control is on the web for everyone and was rolling out to mobile.
The privacy levers, plain and simple
If you handle anything sensitive, client records, payroll notes, deal terms, these are the controls to know.
Temporary Chat. For a one-off conversation you do not want kept, use this. "Temporary Chats don't use existing memories or create new memories." Nothing goes in, nothing comes out.
Training opt-out. ChatGPT may use your content to improve its models, but only with one setting on. OpenAI says: "If you have the 'Improve the model for everyone' setting turned on, we may use content you've shared with ChatGPT, including past chats, saved memories, and memories from those chats, to help improve our models. You can turn this setting off anytime in your Data Controls." Worth a look in your settings.
Business and team accounts are different. OpenAI states, "we do not train on content from ChatGPT Business, Enterprise, and Edu customers by default." And if you run a team account, an admin can set memory for everyone: "ChatGPT Enterprise workspace owners can turn Memory on or off for all users in their Admin Settings."
A few things to know before you dig in
A couple of details that save confusion.
The summary is a high-level view, not a full list. OpenAI describes memory as "a continually updated synthesis of context from your past chats, which may be broader than what can be shown as individual items in a summary." So if you want to confirm whether ChatGPT knows something specific, OpenAI's tip is simple: "just ask in chat."
This is still rolling out. The rebuild started with Plus and Pro users in the US and was expanding "to Free and Go plans and additional countries over the next few weeks." Paid Plus and Pro accounts also get "twice as much memory capacity." If you do not see the full memory summary yet on your plan or in your country, that is why.
You can go back. If the new automatic system is not for you, OpenAI lets you revert: "go to Settings > Memory > Saved memories" for the legacy version.
Your next step
Spend two minutes on this. Open ChatGPT, find your memory summary, and read what it thinks it knows about you and your business. Fix one thing that is wrong or out of date.
That one rep, checking what the tool holds and correcting it, is the same habit that carries through every level of working well with AI. Start small. Start with the screen that finally shows you what is in there.
Related reading: Level 3: The Lieutenant (Critical Thinker).
Sources
Frequently Asked Questions
Does ChatGPT remember everything I type?
It keeps what it decides is most important and synthesizes context from your past chats. The memory summary shows you a high-level view of that, though OpenAI says it will not include everything ChatGPT remembers. To check a specific detail, ask it directly in chat.
If I select Don't mention this again, is it gone?
No. That reduces how often ChatGPT brings something up, but the information stays. To remove it fully, you have to delete it from every place it appears, including past chats, archived chats, files, the summary, and any connected apps.
Can I turn memory off completely?
Yes. From the three-dot menu, choose Delete and turn off memory. Your past chats stay, and if you turn memory back on later, ChatGPT may build new memories from those older chats.
Is my business data used to train the model?
For personal accounts, only if Improve the model for everyone is on, and you can switch it off in Data Controls. For ChatGPT Business, Enterprise, and Edu accounts, OpenAI does not train on that content by default.
Find your AI Proficiency level
The free 7 Levels assessment places you across seven stages of AI capability. Under ten minutes. Research-backed scoring.