Is It Safe to Upload Your Resume to ChatGPT? Settings + Checklist
A resume is a dense bundle of personal data: your name, phone, email, often your address, your full employment history, and sometimes details that don't belong on it at all. Uploading it to ChatGPT is not reckless, but the defaults are not the most private option either. On consumer plans, OpenAI may use your conversations to train its models unless you opt out, and chats stay in your history until you delete them. Here is what OpenAI's own documentation says, and what to change before you upload.
What happens to a resume you upload by default
Four documented defaults matter for a personal ChatGPT account (Free, Go, Plus or Pro):
| Default | What OpenAI documents | Source |
|---|---|---|
| Model training | Content may be used to train models unless you opt out | Data use article |
| Chat history | Regular chats stay saved until you delete them | Retention article |
| Library files | Files can be saved separately from the chat; deleting the chat doesn't delete them | Upload article |
| Memory | When on, ChatGPT can remember details from chats and files in Library | Memory article |
The training default only applies to individual plans. OpenAI says that by default it doesn't use content from ChatGPT Business, Enterprise or Edu workspaces, or the API, to train models. That doesn't make a work account the place for a job search: OpenAI notes that your organization's own privacy, retention and access policies still apply to managed workspaces.
The training opt-out, and its one exception
OpenAI's Data controls article gives the path: account menu → Settings → Data controls → turn off Improve the model for everyone. On mobile it's sidebar → profile icon → Data controls. When signed in, the setting follows your account across devices.
After that, OpenAI says your new conversations won't be used to train its models. Two caveats from the same pages:
- It isn't retroactive in the way you might hope. It covers new conversations from that point on. Turn it off before you upload.
- Feedback overrides it. If you rate a response with a thumbs up or down, OpenAI says "the entire conversation associated with that feedback may be used to train OpenAI models", even after you've opted out. Don't rate responses in a chat containing your resume.
The opt-out doesn't delete or hide anything. Your chats stay in history until you remove them.
Temporary chat is the simplest private option
For a one-off tailoring session, Temporary chat does most of the work for you. Open a new chat, select Temporary, and choose Unpersonalized before your first message. OpenAI documents that a temporary chat:
- doesn't appear in your chat history
- doesn't create or update memories
- isn't used to improve models while it stays temporary
- may be kept for up to 30 days for safety purposes
Two things to know. You can't reopen a temporary chat after you leave it unless you saved it, so copy the finished resume out first. And if you do save it, it becomes a regular chat and follows your normal training and memory settings.
Memory: why your address can follow you around
With Memory on, ChatGPT can draw on past chats, saved memories and files in Library, according to the Memory article. That's convenient for preferences but not for your phone number. Deleting a chat doesn't necessarily delete a saved memory created from it; OpenAI says to remove both, and that it may keep logs of deleted saved memories for up to 30 days for safety and debugging.
To check, open Settings → Personalization → Memory, or ask ChatGPT what it remembers about you, and delete anything job-search related you don't want kept.
Redaction checklist before any upload
Removing data is more reliable than any setting. Work on a copy of your resume, not the original:
None of the redacted items help with tailoring. The assistant needs your experience and the posting, not how to reach you. Keep your job titles, employers, dates and numbers: those are what the rewrite works from, and removing them produces generic output. If an employer name is itself sensitive, replace it with a description such as 'a regional bank' and swap the real name back in afterwards. Put your contact details back in the final file. If you're unsure what belongs on a resume at all, see should you put your address on a resume and adding a LinkedIn URL.
A safe upload, step by step
- Make a redacted copy of your resume using the checklist above.
- In ChatGPT, go to Settings → Data controls and turn off Improve the model for everyone.
- Start a Temporary, Unpersonalized chat.
- Paste the resume text, or attach the redacted file, then paste the posting.
- Tailor using prompts like those in our ChatGPT tailoring prompts. Don't rate responses.
- Copy the result into your real resume file and add your contact details back.
- If you used a regular chat instead: delete the chat, then open Library and delete the uploaded file too.
On deletion, the retention article says a deleted chat is removed from your view immediately and scheduled for permanent deletion within 30 days, unless it was already de-identified or OpenAI must keep it for security or legal reasons.
How the other assistants compare
If privacy is your deciding factor, the defaults differ:
| Assistant | Training on your chats by default | Private mode | Kept for |
|---|---|---|---|
| ChatGPT | Yes, opt-out available | Temporary chat | Up to 30 days |
| Gemini | Yes while Keep Activity is on | Temporary chat | 72 hours |
| Claude | Your choice in Privacy settings | Incognito chat | 30 days |
| Copilot app | Not used to train foundation models | Delete chats manually | Files up to 18 months |
Google's privacy hub adds that chats reviewed by humans are kept for up to 3 years, disconnected from your account. Whichever tool you use, redaction protects you more than any toggle.
Sources
- OpenAI Help Center, Data controls in ChatGPT
- OpenAI Help Center, How your data is used to improve model performance
- OpenAI Help Center, Temporary chat in ChatGPT
- OpenAI Help Center, Memory in ChatGPT
- OpenAI Help Center, Chat and file retention in ChatGPT
- OpenAI Help Center, Uploading files and audio to ChatGPT
- Google, Gemini Apps Privacy Hub
- Anthropic Privacy Center, Is my data used for model training?
- Anthropic, Use incognito chats
- Microsoft Support, Microsoft Copilot for individuals: your activity history
- Microsoft Support, File upload in Microsoft Copilot
Frequently asked questions
- Can I get a copy of, or delete, everything ChatGPT holds about me?
- Yes. OpenAI's Data controls article says signed-in Free, Go, Plus and Pro users can request a copy of their data from settings and can delete their account there. Account deletion is permanent and affects its data and subscriptions.
- Does turning off chat training delete my old chats?
- No. OpenAI says turning off 'Improve the model for everyone' does not delete or hide saved chats. Delete the chats yourself, and delete any uploaded file saved in Library separately.
- Is a Temporary chat completely private?
- No. OpenAI says it may keep a copy for up to 30 days for safety purposes. It is not added to your history, not used for training while it stays temporary, and does not update memory.
- Is pasting my resume safer than uploading the file?
- The same settings apply either way, but pasting lets you see exactly what you are sharing and avoids creating a separate file copy in Library. It also does not count toward the Free plan's 3 uploads a day.