[Cryptography] "Perpetual Encryption"

Phillip Hallam-Baker phill at hallambaker.com
Wed Mar 29 17:03:15 EDT 2017


On Wed, Mar 29, 2017 at 10:41 AM, Bill Cox <waywardgeek at gmail.com> wrote:

> This scheme scores highly, except that it has no web site:
>
> I invented an amazing truly unhackable super-encryption algorithm.  It is
> also super-simple:
>
> 1) Generate n bits of true random data that have no bias or any detectable
> non-randomness
> 2) Manually deliver this OTP random bits to the recipient, then go home.
>

​There are two tests that are critical. Attackers being unable to break the
system and users being able to use it in practice.

​Any system that claims to have a one time pad is either utterly
impractical or not a one time pad. Thus the claim is a near infallible
indicator of bogosity.


The one class of system involving an OTP that could arguably be harder to
break (but not impossible) ​is to mix a random stream of bits into the
stream in a fashion that effectively increases the key length.

​So lets say your initial key is k.

We encrypt the first AES block as follows.

Generate a random value the same size as the block. R
Let RH, DH be the high bytes of the random and data blocks respectively and
RL, DL be the low.

Output = AES (DH+RH, k) + AES (DL+RL, k)​
​k' = AES (R, k)​


​What this achieves is it doubles the size of the cipherstream and stalls
the AES encryption every two blocks.

What it might achieve is some degree of additional work factor ​but I doubt
it.


A more difficult to break arrangement might be

Output = AES (R, k) + AES (D^R, k')
​k' = AES (R, k)​


​It is of course total bollocks though.​ It is not making the cipher
unbreakable and the work factor is actually unchanged.

​There are infinitely many similar schemes. And they all fall short because
the width of the key isn't increased.​
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.metzdowd.com/pipermail/cryptography/attachments/20170329/382005da/attachment.html>


More information about the cryptography mailing list