# How big is the issue of no file size obfuscation with some encryption tools? **URL:** https://discuss.privacyguides.net/t/how-big-is-the-issue-of-no-file-size-obfuscation-with-some-encryption-tools/37391 **Category:** General **Created:** 2026-04-23T22:31:53Z **Posts:** 4 ## Post 1 by @Kabo — 2026-04-23T22:31:53Z I have researched that some encryption tools like Cryptomator, have no file size obfuscation. Source: > > > ## Basic Info > > This is a > - bug report. (privacy) > - feature request. (randomi…ze file size) > > ## Description > > From reading the documentation on the [security architecture](https://cryptomator.org/architecture/) it looks like the file size is not obfuscated. > > The file size is not exactly the original file size, but all the headers and metadata have fixed size. So it should be possible to calculate the exact original file size from only the encrypted file. > Especially for bigger files this could be enough information to identify the file(s). E.g. a collection of a series/movies. > > A small randomly sized block could fix that. E.g add a block of 0% to 1% of filesize. My question is: How big of a deal is this for privacy? Cryptomator protects the file name and content, but if someone stores media files, determining the exact file size, could enable someone else to identify the content. Because of for example a song.mp3 the unencrypted version is publicly available. --- ## Post 2 by @anon96036341 — 2026-04-24T05:46:01Z The answer to your question comes down to your threat model and if such a things is really a concern for you. But I honestly don’t think this is an issue because you can still maintain plausible deniability and no one can prove anything about your data stored unless it’s decrypted. --- ## Post 3 by @Kabo — 2026-05-06T22:53:02Z > [@anon96036341](#): > > But I honestly don’t think this is an issue because you can still maintain plausible deniability and no one can prove anything about your data stored unless it’s decrypted. I read in some other forum that someone got convicted based on file type matching. And apart from that, I am not only worried for legal reason, but also for general privacy reason --- ## Post 4 by @Colter — 2026-06-04T16:06:31Z > [@anon96036341](#): > > But I honestly don’t think this is an issue because you can still maintain plausible deniability and no one can prove anything about your data stored unless it’s decrypted. I think plausible deniability is the lowest bar for privacy, that is only a bit above no privacy. So I would say “there is plausible deniability, so we don’t need to improve it”