Are online PDF tools safe? What happens to your file

Updated October 6, 2026

It depends on one thing: where the file is processed. Some tools send your PDF to a server, work on it there and send the result back. Others load a program into your browser and work on the file on your own device. Both can be run responsibly. They carry different risks, and you can tell which kind you are using in about a minute.

How a server-based tool handles your file

You choose a file, the browser uploads it over an encrypted connection, the service processes it on its servers, and you download the result. The uploaded file and the result stay on those servers for a while so the download link keeps working, and are then deleted.

How long varies. Among the services compared on this site, iLovePDF says files are deleted within two hours, Smallpdf and PDF24 say one hour, and Adobe says a file is deleted unless you sign in to save it. Each of those statements is taken from the service's own pages and linked from the comparisons.

What can go wrong with an upload

None of this means those services are careless. Some publish security certifications, and all of those compared here publish deletion rules. The point is narrower: for a period, a copy of your document exists somewhere you do not control, and you are relying on someone else to delete it.

  • The copy outlives the job. You needed the file for ten seconds. The service holds it for an hour or more.
  • You cannot check the deletion. A stated policy is a promise. It is usually kept, and you have no way to confirm it.
  • Signing in changes the rules. Files saved to an account are kept until you remove them.
  • Your own obligations may forbid it. An employer, a client contract or a professional duty of confidentiality can rule out sending a document to any third party, however well that party behaves.

For a restaurant menu none of this matters. For a payslip, a medical letter, a signed contract or someone else's personal data, it can.

How a browser-based tool works

A browser-based tool reverses the direction of travel. The website sends the program to you, and your file stays where it is.

  1. The page loads, along with a PDF library written in JavaScript. On PDFTools.studio that is pdf-lib for writing PDFs and pdf.js for reading and drawing them.
  2. You choose a file. The browser hands the page the contents of that one file, in memory. A web page cannot browse your folders; it sees only what you pick or drop.
  3. The library changes the file in your device's memory.
  4. The page offers the result as a download, built on your device.

The network is needed for step 1 and for nothing after it. When you close the tab the file is gone from memory.

What PDFTools.studio sends, and what it does not

Your documents are never sent. A few other things are, and you should know what they are.

  • Requests for the libraries and fonts. These come from public content delivery networks, which see your IP address and browser type as any website does.
  • OCR language data. OCR PDF downloads its engine and up to 3 MB of language data the first time it runs. The recognition happens on your device.
  • A web address, in one mode of one tool. The URL tab of HTML to PDF sends the address you type to a public proxy so the page can be fetched. Pasting HTML or uploading an HTML file does not.
  • Analytics and advertising, which you can refuse. Advertising stays off until you accept it. Where your device's time zone is in the EEA, the UK or Switzerland, Google Analytics waits for your answer in the cookie banner; in other time zones it runs until you switch it off under Cookie settings in the footer. The cookie policy lists everything stored in your browser.

How to check any PDF site yourself

You do not have to take a website's word for it, including this one. Open the browser's developer tools, switch to the Network tab, run the tool, and look for a request about the size of your file. A server-based tool shows one. A browser-based tool shows only small requests, for scripts, fonts and any analytics.

The full method, with the limits of what it proves, is in how to check if a site uploads your file.

When a server-based tool is the better choice

A browser is a constrained place to run software, and some jobs are done better on a server or in a desktop program:

  • Office conversions that must keep the layout. The converters here carry text only.
  • Files larger than your device can hold in memory. Most tools here accept up to 150 MB.
  • Changing existing text and keeping the font. The PDF Editor covers the original and draws your text in a standard font.
  • Signatures that need legal standing. Sign PDF places a signature image, not a certificate.

If the document is not sensitive and the job is one of these, use a server-based service and read its deletion policy first.

A checklist before you put a document into any online tool

  1. Would it matter if a stranger read this document? If not, use whatever is convenient.
  2. Does the site say where files are processed? Look for a security or privacy page.
  3. If files are uploaded, how long are they kept, and does signing in change that?
  4. Does a contract, an employer or a regulator restrict where this document may go?
  5. Can the job be done locally? Page operations such as merging, splitting and rotating usually can.
  6. Have you checked the Network tab once, with a harmless file?

Frequently asked questions

Is it safe to upload a PDF to an online converter?

For documents that are not sensitive, a reputable service with a published deletion policy is a reasonable choice. For confidential documents, a copy held on someone else's server for an hour or more is a risk you can avoid by using a tool that processes the file in your browser.

Can a website read the files on my computer?

No. A web page can read only the files you choose in the file picker or drop onto the page. It cannot browse your folders or open anything you did not select.

Do browser-based PDF tools work offline?

Partly. Once a page and its libraries have loaded, tools that use only pdf-lib, such as Delete Pages and Extract Pages, keep working with the network switched off. Tools that draw pages fetch one more file the first time they run, and OCR needs its language data, so run those once while online.

Sources

  • File API (MDN Web Docs): a web page can read a file only when you choose it in a file picker or drop it on the page.
  • Inspect network activity (Chrome for Developers): how to watch every request a page makes in the Network panel.

Tools used in this guide

More guides

Or see all guides and how this site compares with other PDF services.