3

I have made a Veracrypt Containment that uses AES to encrypt everything inside. But my question is: can I securely put the containment file in public (everybody can access it) without anybody able to access the files inside it?

PS. Some information below may help to answer the question:

  1. I use VeraCrypt to encrypt files.
  2. It uses AES-256.
  3. My password-length is 32 characters.
  4. The password contains characters from ASCII
  • 3
    If you do not mind that everybody will know that you have some important secret information, and assuming your password isn't guessable, then yes. In some situations, merely the fact that someone might know that you have something that's worth protecting may be undesirable. A metal door and iron bars on your home may discourage burglars, or it may encourage them (after all, there must be something valuable in there). – Damon Mar 27 '15 at 15:04
  • 1
    I have something very important but can the AES protect it? – John-mc Mar 27 '15 at 15:05
  • 3
    AES can, the question is, can you? If what you have is really important, and the "bad guys" know that you have it, then you are making youself a target (see e.g. wrench attack). – Damon Mar 27 '15 at 15:07
4

AES-256 is perfectly adequate (in fact, AES-128 is perfectly sufficient too, and arguably better according to some experts' opinion) to protect any data that you may have, provided that

  1. Your password is not only long (as you stated) but also not guessable.
  2. The software you use has no backdoor (Veracrypt to my knowledge doesn't have one).

However, the mere fact that someone knows you have something important may already be undesirable. It may rise their attention.

If someone sufficiently evil (criminals or governmental agencies) is sufficiently convinced that you have something of value that they want, but they cannot get it without the key, they will force you to hand over the key.
In other words, anything you encrypt is, in the end, only as secure as you are (which is not very secure at all, if you care about your health).

  • Even according to those expert opinions, AES128's key schedule is only "safer" against related key attacks, which are not an issue the way AES is used practically everywhere. When related key attacks are out of scope, AES256 is a good bit safer (but because of the extra rounds, not because of the double key size). – forest Dec 31 '17 at 23:33
  • @forest: It's probably a bit a matter of opinion there. Methinks that rel key attacks that require yottabyte amounts of memory and are 2^99 complex are irrelevant, but to some people it seems to matter. To me, it doesn't matter whether the algorithm uses 10 or 14 rounds. Practically, 10 rounds are "unbreakable" (unless you are officially listed as Public Enemy #1), as are 14 rounds, or 75 rounds. Those same people who might break 10 rounds will however cut off fingers of yours one by one rather than dedicate billions worth of equipment/time on one person. So algorithm-wise it's all the same. – Damon Jan 1 '18 at 11:51
  • So, in one word, AES-256, why not. But AES-128... again, why not. Both are fine really (in my opinion, for my "normal, non-public-enemy" use). The encryption isn't the weak link in the chain. – Damon Jan 1 '18 at 11:52
  • Yeah I agree and was just nitpicking, both AES128 and AES256 are sufficient (in non-quantum situations where only 256 is) as long as the key is strong. Though I think the related key attacks are actually fairly practical, and AES256 when used in such a vulnerable way result in less than 2^128 effective keyspace. Hopefully either size is sufficient with a strong key, even if you are public enemy #1. :P – forest Jan 2 '18 at 3:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.