Don't stop here
Hand-picked guides our readers explore right after this one.
Master ChatGPT with advanced prompting techniques, mega-prompts, and proven frameworks
Read the guideStunning image generation with Midjourney prompt mastery
Read the guideAI prompts for product strategy, user research, roadmapping, and stakeholder communication
Read the guide'Unknown error occurred' is ChatGPT's catch-all message when an upload fails, and the wording tells you nothing about the cause, which is why the same file can succeed on a second account and fail on yours. In practice the message covers five very different failures: the file is too large or too long, the PDF is encrypted or scanned rather than text, a browser extension is breaking the upload request, your daily upload allowance is spent, or OpenAI's file-processing backend is briefly degraded. Reports through 2026 include the odd pattern of Plus and Pro accounts failing while free accounts uploaded the same file successfully, which is a strong sign of a backend issue rather than anything wrong with your document. Work through the checks below in order, cheapest first. If the file itself is the problem, converting or re-saving it fixes the upload permanently rather than just for one attempt.
The attachment shows a progress bar, then 'Unknown error occurred'
PDFs fail while plain text or DOCX versions of the same content upload fine
The same file uploads on a different account, browser, or device
Uploads fail only in your main browser and work in a private window
ChatGPT accepts the file but then says it cannot see or read its contents
Failures cluster together, then everything works again hours later
Uploads have practical ceilings (commonly cited around 10 MB per file for chat attachments, with tighter effective limits on very long PDFs). Large scanned documents blow past these quickly because the images inside inflate the file far beyond what the page count suggests.
A scanned PDF is a stack of images with no extractable text layer, and a password protected or permissions-restricted PDF cannot be opened for parsing at all. Both surface as the same generic error rather than a useful message.
Ad blockers, privacy extensions, and ChatGPT-specific add-ons intercept the upload request. This is the most common cause when uploads suddenly stop working in one browser but succeed in a private window or on your phone.
File uploads are metered separately from messages and reset on a rolling window. When the allowance is spent, uploads fail with a generic error rather than an explicit limit notice.
Users have repeatedly reported waves of upload failures affecting paid accounts while free accounts worked, which points at the processing backend rather than your file. These waves typically clear within a few hours without any action on your part.
When to try: First, it takes 30 seconds and rules out half the causes
Open a private or incognito window, sign in, and upload the same file. If it works there, the cause is an extension or stale cached session in your normal browser, not the file. Disable extensions one at a time in your main browser to find the culprit.
When to try: If the file is large or image heavy
Right click the file and check its size. Anything approaching or above 10 MB is a likely failure. Compress the PDF (most PDF tools have a 'reduce file size' or 'optimize' option), or split a long document into parts and upload them one at a time.
When to try: If the file is a PDF, especially one from a bank, government portal, or scanner
Open the PDF and use Print, then 'Save as PDF' to produce a clean copy. This removes password protection you can already open, drops broken embedded objects, and flattens unusual structure that the parser chokes on. Upload the fresh copy.
When to try: When re-saving the PDF did not help
Export or copy the content into DOCX, TXT, or Markdown and upload that instead. Text formats almost never fail, and for text-heavy documents you lose nothing. For spreadsheets, save as CSV rather than XLSX if the upload keeps failing.
When to try: If the document was scanned or photographed
If the PDF is a photo or scan, run it through an OCR step first so it contains real text. Most PDF readers, Google Drive (open the PDF with Google Docs), and phone scanner apps can do this. Without a text layer, the upload may succeed but ChatGPT will still see nothing useful.
When to try: When individual uploads keep failing but you need the content in now
Put the file, or several files, into a single ZIP archive and upload that. Users have repeatedly found ZIPs go through when the individual files do not, because the archive takes a different processing path. Ask ChatGPT to extract and read the contents after uploading.
When to try: If desktop uploads fail consistently
Send the file to yourself and upload it from the ChatGPT mobile app. This bypasses browser extensions, corporate network filtering, and any local proxy entirely, and often succeeds when the desktop browser will not.
When to try: When even a tiny .txt file fails to upload
If several different files fail in a row, including small plain text ones, the problem is almost certainly on OpenAI's side. Check status.openai.com, then try again in two to five hours. Community reports on these upload waves consistently describe them resolving without user action.
Keep working documents under about 10 MB and split long PDFs into parts before uploading
Store reference material as text or DOCX rather than scanned PDFs so it always parses
Keep a private window bookmarked for ChatGPT so you can test extension issues instantly
OCR scans as part of your filing routine, not at the moment you need an answer
Contact OpenAI support through help.openai.com if uploads fail on every file type including small plain text files, across two different devices and networks, for more than a day, while status.openai.com shows no incident. Include your plan, the exact filename and size, timestamps, and the fact that a private window and mobile app were both tested. If only one specific file fails and everything else works, that is a file problem rather than a support ticket.
It is a generic upload failure message, not a specific diagnosis. It covers oversized files, encrypted or scanned PDFs, browser extensions blocking the request, a spent daily upload allowance, and temporary problems with OpenAI's file processing. The message is identical for all of them, which is why you have to test rather than read the cause off the screen.
PDFs have to be parsed before ChatGPT can read them, and that step fails on encrypted files, permission-restricted files, scanned images with no text layer, and files with unusual internal structure. Text and DOCX skip most of that parsing, which is why they almost always succeed.
Often, yes. Users report ZIP archives uploading successfully when the same files fail individually, because the archive follows a different processing path. Upload the ZIP, then ask ChatGPT to extract and read the contents. It is a workaround rather than a fix, but it unblocks you in the moment.
It can be. File uploads are metered separately from messages, and when the allowance is exhausted the failure shows as a generic error rather than a clear limit notice. If uploads worked earlier in the day and stopped after heavy use, wait for the rolling window to reset and try again.
That pattern has been reported repeatedly during upload outages in 2026 and points at OpenAI's backend rather than your account or file. Different account tiers can be routed differently during degradation. There is no fix from your side beyond waiting a few hours and retrying.
Product behavior and limits can change. These primary sources were used to verify this guide.