AES-256 Encryption Possibly Now Broken After New Method To Speedily Factor Integ

[conspirosphere.tk]

Not sure if BTC-relevant:

http://investmentwatchblog.com/encryption-security-may-not-be-secure-anymore/

the students found that intermediate keys created specifically within the AES-256 encryption algorithm had cryptographically weak output that followed a Quadratic curve when initial keys contained identifiable Fibonacci sequences, non-evenly divisible values including PI, Catalan numbers and Mersenne primes which allowed the students to estimate possible integer factors allowing them to recover the initial encryption key within as little as 100 hours compute time.

[Lauda]

The title on the link is false though:"Encryption Security May Not Be Secure Anymore". They can't possibly talk about all encryption algorithms. I'm not fully certain, but I think that (all of) the algorithms that are used in Bitcoin should not be affected by this. This development of this story will certainly be interesting.

the students found that intermediate keys created specifically -snip-

I'd like to know the meaning of the words 'intermediate keys' in this context. I would also not be surprised by this part:

According to industry experts, the discovery also raises the question of whether the ‘flaw’ was actually an inbuilt component of encryption technology, introduced on purpose by its creators or encryption standards modifiers.

[NyeFe]

AES isn't related to bitcoins protocol.

[Yogafan00000]

AES isn't related to bitcoins protocol.

Also, the article is a load of shit.  There's no press release, or names or any evidence of any mathematical breakthrough.  They took a bunch of fancy maths terms and mushed them together to make a hoax article.

[CIYAM]

AES isn't related to bitcoins protocol.

Correct - but AES256 is used for encrypting wallets (with Bitcoin Core).

I do agree that the article looks a bit suspicious though.

[iamnotback]

More info:

http://www.abovetopsecret.com/forum/thread1120355/pg1

From that description it appears that it may only apply to private keys containing mostly (or only?) alphanumeric characters?

And there is still a possibility it is a hoax.

I read else where that the NSA expects to break AES-256 by 2018 any way. That appears to unsubstantiated.

Wild conjecture is the other potential damning outcome is if it is shown that this was likely a planted security hole. That could have ramifications on why does Bitcoin use what may be potentially an insecure form of double-hashing in that if so then maybe that was a planted weakness.

[Cuntabula]

I read else where that the NSA expects to break AES-256 by 2018 any way.

That's why destroying Satoshi's coins is so imperative!

[CIYAM]

That's why destroying Satoshi's coins is so imperative!

Huh?

(as Satoshi's wallets were created before encryption was even used in Bitcoin wallets I hardly see how that would be relevant)

[numismatist]

AES isn't related to bitcoins protocol.

Also, the article is a load of shit.  There's no press release, or names or any evidence of any mathematical breakthrough.  They took a bunch of fancy maths terms and mushed them together to make a hoax article.

Wallet files encryption algos could be changed in between days.
This article doesn't got me alarmed neither.

[Bitcoinpro]

Exchanges r getting desperate

obtaining 1 Bitcoin should be much harder

than 570 USD thats a damn easy investment

figure to put you in the bracket of the 15 million richest

people in the world.

[Foxpup]

Obviously fake article. Besides the obvious bullshit (what has factoring integers got to do with AES? or Fibonacci sequences to do with anything related to cryptography?) and the fact that it's only being published on tinfoil-hat conspiracy sites, the article specifically cites a press release from the University of Toronto as its source, which is trivially easy to verify. The University of Toronto has issued no such press release. And if the press release is a fabrication, so is the whole story, since there's literally nothing else backing it.

It's embarrassing that people actually fall for these hoaxes.

[thejaytiesto]

AES isn't related to bitcoins protocol.

Bitcoin Core uses AES256 to encrypt the wallet, so it's very relevant. Also, a lot of people use 7zip to encrypt their wallet.dat files inside .7z files, and 7zip uses AES256 algorithm for encryption.. if this is true, then I want to know what would be the ideal way to encrypt the wallet.dat file then.

[belmonty]

Obviously fake article. Besides the obvious bullshit (what has factoring integers got to do with AES? or Fibonacci sequences to do with anything related to cryptography?) and the fact that it's only being published on tinfoil-hat conspiracy sites, the article specifically cites a press release from the University of Toronto as its source, which is trivially easy to verify. The University of Toronto has issued no such press release. And if the press release is a fabrication, so is the whole story, since there's literally nothing else backing it.

It's embarrassing that people actually fall for these hoaxes.

I Google searched for a phrase from that quote and only found these few obscure blogs publishing the story. If it was for real it would be plastered all over mainstream news sites. The domain names of some of those blogs alone make it clear they are unreliable sites.

[BillyBobZorton]

Obviously fake article. Besides the obvious bullshit (what has factoring integers got to do with AES? or Fibonacci sequences to do with anything related to cryptography?) and the fact that it's only being published on tinfoil-hat conspiracy sites, the article specifically cites a press release from the University of Toronto as its source, which is trivially easy to verify. The University of Toronto has issued no such press release. And if the press release is a fabrication, so is the whole story, since there's literally nothing else backing it.

It's embarrassing that people actually fall for these hoaxes.

I Google searched for a phrase from that quote and only found these few obscure blogs publishing the story. If it was for real it would be plastered all over mainstream news sites. The domain names of some of those blogs alone make it clear they are unreliable sites.

This is indeed very strange... AES256 being broken would be HUGE news. Julian Assagne trusted aes256 to encrypt is famous "insurance.aes256" file, so I doubt we are anywhere near AES256 being broken.

[Cuntabula]

That's why destroying Satoshi's coins is so imperative!

Huh?

(as Satoshi's wallets were created before encryption was even used in Bitcoin wallets I hardly see how that would be relevant)

AES-256, ECDSA... Quantum computores, Amigo! Nothing is safe. Poison the wells, salt the fields!

This issue has been discussed for several years. I think that the very-rough consensus is that old coins should be destroyed before they are stolen to prevent disastrous monetary inflation. People joined Bitcoin with the understanding that coins would be permanently lost at some low rate, leading to long-term monetary deflation. Allowing lost coins to be recovered violates this assumption, and is a systemic security issue.

https://www.reddit.com/r/Bitcoin/comments/4isxjr/petition_to_protect_satoshis_coins/d30we6f

[greyhawk]

This isn't being reported by any reputable sources, nor has any real detail been provided and the initial press release has been removed. I call shenanigans.

[pereira4]

I would like to hear what a core dev like gmaxwell that posts here has to say about this, since wallets are encrypted with AES256.
If this was real, im sure they would have made a thread already telling everyone to stay cautious and they would be about to deploy a quick update-fix with a changed algo for the wallet encryption.

Nothing of this is happening so I call bullshit.

[greyhawk]

I would like to hear what a core dev like gmaxwell that posts here has to say about this, since wallets are encrypted with AES256.
If this was real, im sure they would have made a thread already telling everyone to stay cautious and they would be about to deploy a quick update-fix with a changed algo for the wallet encryption.

Nothing of this is happening so I call bullshit.

AES-256 along with most modern crypto will probably break slowly. A weakness will be found in some part of it and software using it will transition to another algo long before it is fully cracked.

[Cuntabula]

I would like to hear what a core dev like gmaxwell that posts here has to say about this, since wallets are encrypted with AES256.
If this was real, im sure they would have made a thread already telling everyone to stay cautious and they would be about to deploy a quick update-fix with a changed algo for the wallet encryption.

Nothing of this is happening so I call bullshit.

They'll get in touch with you as soon as they're cashed out & Blockstream buttons down its exit strategy

[Slark]

Exchanges r getting desperate

obtaining 1 Bitcoin should be much harder

than 570 USD thats a damn easy investment

figure to put you in the bracket of the 15 million richest

people in the world.

I don't think this info is prepared by bitcoin exchange lobbyists (but hey, you can never be sure).
But seeing how allergic bitcoin crowd is to any kind of 'not positive' news I wouldn't be surprised that bitcoin price would drop a little as a result.