NSA has backdoors to all software

above page proves AES is not hackable.

it's really not...

As I mentioned, AES is secure for now. The related key attack and the biclique attack are the best known attacks on AES currently. That being said there is a common saying in the crypto world that attacks always get better, and never worse.

It's doubtful that AES is a perfect cipher, the only provably perfect cipher in existence is a vernam cipher (one-time pad). Right now AES-128 is projected to be strong in to 2030 by then there will be an AES2, but there could always be breakthroughs.

These attacks are all simply related to the math behind AES. In the real world you need an implementation (details of which can be attacked), and you have to actually assemble a cryptosystem (again details which can be attacked). AES by itself does not ensure confidentiality, it has to be used correctly and there are lots of applications that don't. Browse the StackOverflow [encryption] tag someday and see what some people actually put in their Android/iOS/PC applications and call it secure.
 
it's really not...

WEP also was really not. ;)

ANY encryption can also be brute forced. Not a matter of if it gets cracked but a matter of when. Of course it can take a long time. Though an encryped file is simply limited by the computer power you throw at it. I imagine the NSA has ASICs and other such devices for this.
 
WEP also was really not. ;)

ANY encryption can also be brute forced. Not a matter of if it gets cracked but a matter of when. Of course it can take a long time. Though an encryped file is simply limited by the computer power you throw at it. I imagine the NSA has ASICs and other such devices for this.

Though WEP-64 was indeed brute forced. WEP as a crypto protocol also has serious flaws that could be exploited for cryptanalysis, i.e 24-bit IV. To add on top of that, If WEP wasn't already completely broken, the cipher WEP used, RC4, is also broken now, though this is a more recent development.

On top of the protocol flaws in the spec there were also numerous weaknesses introduced by implementations created by vendors of WEP capable wireless gear such as using ASCII as key material. When it rains, it pours I guess.
 

The annoying thing about articles like these, to me, is that they claim attacks take a specific amount of time, instead of mentioning that it's an average or maximum amount of time. Even with AES-1048576, there's still a chance that the very first attempt just happens to get it right. Of course, it's extremely unlikely, but it's still incorrect to say that an attack "will" take a specific amount of time. I tend to expect that journalists that report on technology should understand things like these (and also a more common mistake that also annoys me - confusing "x% greater than ___" and "x% as much as ___" - I'm glad idiots like that are journalists and not engineers!)
 
Last edited:
The annoying thing about articles like these, to me, is that they claim attacks take a specific amount of time, instead of mentioning that it's an average or maximum amount of time. Even with AES-1048576, there's still a chance that the very first attempt just happens to get it right. Of course, it's extremely unlikely, but it's still incorrect to say that an attack "will" take a specific amount of time. I tend to expect that journalists that report on technology should understand things like these (and also a more common mistake that also annoys me - confusing "x% greater than ___" and "x% as much as ___" - I'm glad idiots like that are journalists and not engineers!)

To be fair, most instances I saw were the journalist quoting someone. It's just a watered down way of explaining an attacks time complexity. Guessing the key is an attack (if you can call it an attack) all ciphers are vulnerable too, but it isn't really relevant because it isn't something you can't defend against. This is simply a corollary to "ciphers are vulnerable to brute-force attacks" and doesn't really need to be stated. The only cipher immune to brute-force attacks is a one-time pad which is unsuitable for most use-cases.
 
Last edited:
The annoying thing about articles like these, to me, is that they claim attacks take a specific amount of time, instead of mentioning that it's an average or maximum amount of time. Even with AES-1048576, there's still a chance that the very first attempt just happens to get it right. Of course, it's extremely unlikely, but it's still incorrect to say that an attack "will" take a specific amount of time. I tend to expect that journalists that report on technology should understand things like these (and also a more common mistake that also annoys me - confusing "x% greater than ___" and "x% as much as ___" - I'm glad idiots like that are journalists and not engineers!)

+1 for mentioning this.
 
This is why I try to avoid proprietary software now and my stuff is hosted in Canada. I'd be ignorant in saying I'm 100% safe, but I'd imagine I'm much safer than if I used proprietary software and cloud based services that are in the US.

You realize that by putting the stuff overseas, you are allowing the NSA to do whatever they want?

At least if it was in the US, they are hoops they ahve to jump through to get at it.
 
You realize that by putting the stuff overseas, you are allowing the NSA to do whatever they want?

At least if it was in the US, they are hoops they ahve to jump through to get at it.

To pile onto this comment, if you do business outside of the US with a company that has US infrastructure, your information potentially could be accessed under the Patriot Act.

For some fun reading, Google U.S.A. Patriot Act MS Office 365
 
Last edited:
You realize that by putting the stuff overseas, you are allowing the NSA to do whatever they want?

At least if it was in the US, they are hoops they ahve to jump through to get at it.

Those are only political/legal loops, I doubt the NSA cares about following any of that. They do what they want no matter what. Their only limits is technology and I'm sure they have all that sorted.

By having my stuff in Canada there is less of my traffic going through the states. Though I just checked and for some reason the traffic goes to NY then comes back to Canada... that is unnerving and I did not figure this till I did a traceroute. Most of that stuff is encrypted though so it adds an extra step for them to get to the actual data.
 
Last edited:
that is unnerving and I did not figure this till I did a traceroute. Most of that stuff is encrypted though so it adds an extra step for them to get to the actual data.

There's a little more to it than that: (Room 641A)

Even when your traffic crosses oceans.

USS Jimmy Carter

Encryption does add a step as you mention, but the NSA is everywhere, directly on the wire ... capturing and storing information. It's likely it's all being decrypted out of band.

Granted, your not likely a target, but you could be caught in the net. Ha, no pun intended!
 
Last edited:
I'm not so sure about that (Room 641A)

Not even if it crosses oceans.

USS Jimmy Carter

Encryption does add a step as you mention, but the NSA is everywhere, directly on the wire ... capturing and storing information. It's likely it's all being decrypted out of band.

Granted, your not likely a target, but you could be caught in the net. Ha, no pun intended!

Yeah knew about that, but I originally figured my traffic would not even pass through the states at all but turns out it does. Geographically it makes no sense. For some reason it is going through a Global Crossing node before returning to Canada. I would imagine NYC has a wire tap centre and it's probably why the traffic takes this odd detour. Ether way, Canada has CSEC which is basically the NSA. Really seems there is no way around this unethical spying. (illegal would be a bad word to use, government just makes it legal if it pleases them)

Hyperboria/meshnet really needs to take off. This stuff is getting so bad.
 
Last edited:
Back
Top