Are online file converters safe?
Most conversion sites upload your file to a server you know nothing about. Sometimes that is fine. Sometimes it is a very bad idea.
Published
What normally happens when you upload
On a conventional conversion site, your file is transmitted to the operator's servers, written to disk, processed, and stored until a download link is fetched or a retention timer expires. That means a complete copy of your file exists on infrastructure you do not control, for a period you cannot verify, accessible to anyone with access to those systems.
For a holiday photo, that is a fair trade for a free service. For a signed contract, a passport scan, a medical letter, an unreleased recording or a spreadsheet of customer records, it is a decision worth making deliberately rather than by accident.
Read what the policy actually promises
"We delete your files after 24 hours" is a retention commitment, not a privacy one. It says nothing about who can read the file during those 24 hours, whether backups persist beyond it, which sub-processors receive it, or which jurisdiction the servers sit in.
The questions worth asking are narrower: is the file encrypted at rest, who on staff can access it, is it shared with any third party, and which country's law applies if someone demands it.
How to tell which kind of site you are on
The reliable test takes ten seconds. Open your browser's developer tools, switch to the Network tab, and convert a file.
- Server-side: you will see a large POST or PUT request roughly the size of your file, then polling, then a download.
- Client-side: you will see the page's own code load, possibly a one-off engine download, and then no request carrying your file at all.
A second clue: disconnect from the internet after the page has loaded. A converter that still works was never sending your file anywhere.
What in-browser conversion changes
Browsers can now run real conversion engines locally through WebAssembly — including ffmpeg, the same engine many desktop converters are built on. The file is read from disk into the page's memory, converted by your own processor, and handed back.
No upload endpoint exists, so there is no server copy to leak, subpoena, retain or sell. The operator cannot see the contents, the filename or even the file type, because none of it is transmitted. That is a structural guarantee rather than a policy promise, which is a meaningfully different thing.
It is also not magic. The page still comes from a server, and in-browser tools often pull their larger engines from public CDNs, so you are trusting the site operator and every provider it loads code from not to ship malicious code. That is the same trust any website requires. What you can check for yourself is narrower but still useful: open the Network tab while you convert, and you will see no request carrying your file. That check covers the visit you watched, not future ones.
A sensible rule
Ask one question: would it matter if a stranger read this file? If the answer is no, use whatever is convenient. If the answer is yes, use a converter that runs locally, or desktop software. The cost of being wrong is asymmetric — the convenience saved is small, the exposure is not.
Before sharing a photo anywhere, you can also remove its EXIF data, which strips the location and camera details many photos carry. Every tool on this site works the same way described above: in your browser, with nothing uploaded.