Deleting a ChatGPT chat feels like deleting the data. It is only one deletion, and one detail can appear in several places at once. The Memory page lists five places where one piece of information can appear: the memory summary, saved memories, regular and archived chats, files in Library, and connected apps. So a detail shared in a chat, saved as a memory, and saved as a Library file needs at least three separate deletions, not one. Counted for this guide, there are six removal actions, and they remove different things. This guide maps each one to what it actually removes, and what it leaves behind.
Does deleting a ChatGPT chat permanently delete it?
From your account view, yes, immediately. From OpenAI’s systems, on a schedule. When a chat is deleted, ChatGPT removes it from the account view immediately, according to OpenAI’s deleting and archiving page, and schedules it for permanent deletion from its systems within 30 days, unless it was already de-identified and disassociated from the account or OpenAI must retain it longer for security or legal obligations. A deleted chat cannot be restored in ChatGPT or through OpenAI Support. That 30-day schedule is the honest meaning of “permanent” here.
To delete a single chat, work through these steps in order.
- Open chat history, select the more options menu for the chat, select Delete, then confirm the deletion.
- Open the app, select the menu icon, press and hold the conversation title, then select Delete.
- Open Settings, select Data controls, select Delete all chats, review the confirmation message, then continue.
The first step is the web route, the second is iOS and Android, and the third deletes every chat at once. Delete-all affects chats in the current account or workspace, including chats in projects, and it cannot be undone. One more exception matters for work accounts: authorized compliance tools may still access eligible workspace data under the organization’s retention policy, even after a chat is deleted. Available controls depend on sign-in, plan, and workspace settings.
Files follow their own clocks, according to OpenAI’s chat and file retention page. Files uploaded as knowledge to a custom GPT are retained until the GPT is deleted, and files added to a project are retained until the project or file is deleted or a workspace retention policy removes them; after deletion, the associated data is scheduled for permanent deletion within 30 days, unless security or legal obligations require longer. Retention depends on the type of chat or file, the action taken, and any workspace retention policy that applies. For Enterprise workspaces, some transient files not saved to Library can expire after 48 hours unless another retention period applies, and after a Library file is deleted, expires, or otherwise becomes inactive, OpenAI backups may retain it for up to 30 additional days unless OpenAI must retain it longer for security or legal obligations.
Does archiving a ChatGPT chat delete it?
No. Archiving hides a chat; it deletes nothing. Archive a chat from history by selecting the more options menu and then Archive, with no confirmation required, and the chat leaves the regular sidebar but remains searchable under the same retention rules as other saved chats, according to the same deleting and archiving page. Archiving does not change a chat’s retention period, and regular and archived chats remain saved to the account until deleted or an applicable workspace retention policy removes them. From Archived chats, managed under Settings and then Data controls, with the exact label depending on device and Manage selected if it appears, you can select Unarchive to return a chat to regular history, or Delete to delete it. Archive is the right tool for clearing the sidebar. It is the wrong tool for removing data, and one practical trap comes with it: archiving a chat used for scheduled-task results can cause the task to pause on a later run.
How do I delete what ChatGPT remembers about me?
Start from the five sources, because deleting the chat alone leaves the rest standing. Memory controls live under Settings, then Personalization, then Memory, according to OpenAI’s memory page, though memory features and controls can vary by plan, region, platform, and workspace settings. When Memory is enabled, ChatGPT can remember relevant preferences and details from chats and other available sources, including past chats, saved memories, custom instructions, files in Library, and content from connected apps such as Gmail. Saved memories are stored separately from chat history, so deleting the original chat does not automatically delete a separate saved memory. To remove one piece of information properly, delete the saved memory and the chat where it was first shared, and also remove it from the other sources the Memory page lists: the memory summary, saved memories, regular and archived chats, files in Library, and connected apps.
Three controls sound like deletion and are not. Turning off Memory does not delete past chats, and if Memory is turned on later, ChatGPT may create new remembered information from chats that remain in history. “Delete and turn off memory” deletes the remembered information shown in the summary and turns Memory off, but it does not delete past chats. “Don’t mention this again” reduces future references to the information; it does not delete the original source. Telling ChatGPT not to use information in future responses changes personalization behavior but does not delete the underlying chat, file, email, or other source. Two more limits complete the picture: deleting a saved memory prevents it from being used in future personalization, but it does not remove mentions of that information from past conversations, and OpenAI may retain logs of deleted saved memories for up to 30 days for safety and debugging purposes. If Reference chat history is turned off, information remembered from past chats is scheduled for deletion from OpenAI systems within 30 days, while the original chats remain in history unless deleted, and turning it off does not delete saved memories. The memory summary itself is a high-level view and does not necessarily include every detail or source that Memory can reference.
How do I stop ChatGPT from using my chats to train its models?
Use the training control, not the memory control, because the two are separate settings. Memory controls how ChatGPT personalizes responses; model-training controls determine whether eligible content can be used to improve OpenAI models, according to OpenAI’s data controls page. On personal workspaces with ChatGPT Free, Plus, or Pro, data sharing is enabled by default. On the web while signed in, open the account menu, select Settings, select Data controls, turn off Improve the model for everyone, then select Done. On iOS and Android, open the sidebar, select the profile icon to open Settings, select Data controls, and turn the same setting off. When signed in, the setting applies to the account across web and mobile devices; when using ChatGPT without signing in, the choice is saved in that browser.
Turning the setting off has a precise scope. New conversations will not be used to train OpenAI models, apart from the feedback exception below and the Trusted Access exception below, but they can still appear in chat history, and turning it off does not delete or hide saved chats. Archiving a chat does not change whether it can be used to train models; the Improve the model for everyone setting continues to apply. Two exceptions catch people out. If feedback is submitted after opting out, such as selecting thumbs up or thumbs down on a response, the entire conversation associated with that feedback may be used to train OpenAI models. And during Trusted Access sessions on a personal account, such as Daybreak Blue sessions, OpenAI may use content from interactions to improve its models even when Improve the model for everyone is off, to strengthen safeguards against cyber misuse, while the setting still applies to other activity outside a Trusted Access session. The defaults run the other way for work and developer use: by default, OpenAI does not use content from ChatGPT Business, Enterprise, Edu, or ChatGPT for Healthcare workspaces to train its models, and does not use inputs and outputs provided through the API Platform either. If Codex is used on a personal ChatGPT plan, the same Improve the model for everyone setting applies to Codex tasks. Codex environment data is separate: Codex has its own Include environments setting that controls whether additional context from Codex environments can be used to train models, and changing the ChatGPT setting does not change that setting. The pages state only that new conversations will not be used, so treat the opt-out as forward-looking only.
What is a ChatGPT temporary chat, and does it save anything?
A temporary chat is the preventive tool: a conversation that stays out of history unless you choose to save it. According to OpenAI’s temporary chat page, a temporary chat is best for a conversation you do not want saved to history unless you choose to keep it, while a regular chat is best for one you want to keep and return to later. While they remain temporary, temporary chats do not appear in chat history, do not create or update memories, and are not used to improve OpenAI models, though OpenAI may retain a copy for up to 30 days for safety purposes. Before the conversation begins, you choose whether the temporary chat is personalized: a personalized temporary chat can use existing memories, custom instructions, and plugins, subject to account and workspace settings, but even with personalization on, a temporary chat does not create or update memories. If a temporary chat is saved, it becomes a regular chat in history and follows the account’s personalization and model-improvement settings, so saving one after the fact gives up the protections that made it temporary.
How do I delete my ChatGPT account, and can I get it back?
Deleting the account is permanent, and it is the only one of the six removal actions counted for this guide that ends your access to ChatGPT and the API and schedules account data for deletion within 30 days, except for a limited set OpenAI may retain longer where required or permitted by law. According to OpenAI’s account deletion page, signed-in ChatGPT Free, Go, Plus, and Pro users can delete their account from settings, while self-service account deletion is not available in managed workspaces and should be directed to the workspace owner. On the web, sign in, open the account menu, select Settings, select Account, select Delete next to Delete account, enter your account email or phone number, type DELETE when prompted, and select Permanently delete my account. If you cannot sign in, use the Privacy Portal to submit a verified deletion request, and if the app returns to chats without a deletion confirmation, use the Privacy Portal steps, since completing a sign-in prompt by itself does not confirm that the account was deleted.
Three consequences deserve a pause before confirming. Deleted accounts cannot be reactivated, but a new account can be created with the same email address after 30 days, provided the prior account was fully deleted. Deleting an account does not reset the API verification limit: a phone number can verify the first API key for up to three accounts, and deleting an account does not reset that limit. Before deleting, check any subscriptions billed through the Apple App Store or Google Play. Deleting the account cancels subscriptions billed directly by ChatGPT, but it does not cancel an Apple subscription, so cancel any mobile subscription you no longer want before deleting, and if the email address for the OpenAI account is also used to access an Ads account, deleting it may prevent accessing or managing that Ads account while active campaigns could continue serving and accruing charges, so pause or end active campaigns first. The account deletion page itself lists gentler alternatives: to stop paying, cancel the subscription; to remove a conversation, delete or archive it; to review memory controls, use Memory settings.
Export before you delete anything you might want back. Signed-in ChatGPT Free, Go, Plus, and Pro users can request a copy of their data from settings, according to OpenAI’s export page, and eligible ChatGPT Edu workspaces may allow exports when an administrator enables them, while self-service export is not available for Business, Enterprise, or Healthcare workspaces. On the web, sign in, open the profile menu, select Settings, select Data controls, select Export under Export data, and confirm the export. ChatGPT sends an email or SMS message when the export is ready, exports can take up to 7 days to arrive, and the download link expires 24 hours after receipt, so download it while signed in to the same account that requested it. The downloaded ZIP file includes chat history and other relevant account data, but an export cannot restore chats that have already been deleted.
The 30-day figure recurs across the pages: deleted chats, account data, GPT and project file data, temporary-chat safety copies, reference-chat-history information, and deleted saved-memory logs, against one confirmed shorter exception, Enterprise transient files at 48 hours.
The rule that ties the six actions counted here together is simple: match the deletion to the target. Deleting a chat removes the chat. Deleting a memory removes the memory. Deleting a Library file removes the file. Turning off training stops future training use, apart from the feedback and Trusted Access exceptions. Only deleting the account ends access and schedules account data for deletion within 30 days, except for a limited set retained longer where required or permitted by law. And when the target is everything, export first, because no deletion in ChatGPT can be undone.
