“Unbreakable” Data Encryption Challenge

[Terry Olaes]

Permanent Privacy announced that they have the world’s first practical data encryption system that is absolutely unbreakable. They’re so sure of this that they’re offering $1 million USD to anyone that can crack it. Break out your decoder ring and Enigma machine and get “crackin’!”

Permanent Privacy has been verified by Peter Schweitzer, one of Harvard's top cryptanalysts, and for the inevitable cynics Permanent Privacy is offering $1,000,000 to anyone who can decipher a sample of ciphertext.

 

[Shlomo]

I'm sure by the time somebody cracks it, they'll put down the million dollar challenge.

I guess this is kind of like saying that HD DVD is uncrackable, and backing it with a one week million dollar challenge. Then a month later somebody breaks it....

 

[wasted]

but you first have to purchase a 39 dollar license ... nice marketing scheme

 

[matrix563]

heh. probably won't take someone long to do it

 

[InorganicMatter]

It's kind of hard to judge this without knowing what they are using (the article makes no mention). Is it 48-bit DES? The hackers are gonna have a field day. 256-bit AES + 128-bit Blowfish? Computationally infeasible; it's gonna take quantum computing before we can break the really high end stuff.

 

[glutto]

Funny how the ol' DMCA hasn't been mentioned yet.

Has it been forgotten already, or should we all pretend to be blindsided when The DMCA Attacks?

 

[DanNeely]

It's kind of hard to judge this without knowing what they are using (the article makes no mention). Is it 48-bit DES? The hackers are gonna have a field day. 256-bit AES + 128-bit Blowfish? Computationally infeasible; it's gonna take quantum computing before we can break the really high end stuff.

... one time pad. Unless the pad is compromised theoretically impossible.

 

[defuseme2k]

if the zodiac killer can release ciphers no one has broken, then I'm sure some company can do it eventually

 

[Rational Egoist]

I call BS on this. There's no way that they can make that claim. Basically what they're saying is that they encrypt gibberish with AES and for that reason you'll never know if you've got it right.
Here's their mistake: They have to convert from plain text to the gibberish in some reversible fashion. This is a part of the encryption. I remember someone asking about this in a CS course my freshman year at CMU. Rudich said that doing something like that doesn't make the encryption any stronger. I'd like to see what others who know more about cryptography will say.

 

[krupted]

if anything can be decoded it can be cracked, or at least that's what i've always thought. where it gets interesting is when we start using fiber optics with single photons as keys... any sort of detection at all and the key is killed because photons are destroyed when observed. that means one time use only, no matter what by the law of physics. THAT is truly uncrackable stuff.

 

[preinheimer]

It's PR, nothing more.

I had a long post refuting the point of all this, but someone with a lot more credibility than I did it years ago: http://www.schneier.com/crypto-gram-9812.html

 

[w1retap]

lol... I love these 'news' articles.

 

[k1DBLITZ]

I once read an article where they stated people are now using GPU's to crack passwords and how exponentially faster they were than a CPU performing the same operation.

Wonder if a smart programmer and a couple gtx280's could handle this.

 

[Vorazan]

OMG! Its 50 gigs of midget porn!

 

[jimmyb]

... one time pad. Unless the pad is compromised theoretically impossible.

Very true. And it's not as if one-time pads are a new method of encrypting data...

 

[EXreaction]

This looks like a scam.

It is practically impossible to crack something without knowing at least a little bit about the encryption scheme. They could have encrypted a bunch of crap and you'd never be able to tell if you cracked it or not.

 

[knyghtryda]

Its interesting how they want the hackers to attack the cipher text instead of the app. The easiest way to crack anything is the directly attack the app and try to find memory leaks/buffer overruns etc... to exploit. No sane hacker is gonna try and break something that 1) is just cipher text and 2) is encrypted with something like AES 256-bit. The only way to make this challenge "fair" (and therefore very very costly to the company) is for them to release the algorithm and the app to test, and for them to release part of the cleartext and have the challenge be revealing the remainder. Not open source, mind you, but just a executable and a some white pages to hammer at. An "unbreakable" cipher should be just as solid if part of the message is known as if none of the message is known. If they did that however they would probably be cracked in a week due to some coding error in the app...