Page 1 of 1

#1 NSA-Induced Compromised Encryption Exploited In The Wild

Posted: Wed Dec 23, 2015 9:59 am
by White Haven
Tom's Hardware wrote:Recently, Juniper Networks announced that it found “unauthorized code” (backdoor) in its NetScreen firewall operating system, ScreenOS. According to securityexperts, that “unauthorized code” may, in fact, have been “authorized” - by the NSA. Other hackers merely stumbled upon it and started using it as their own, with a few small changes to it.

The NSA backdoor in question is the controversial Dual EC algorithm for random number generation. It was standardized by the National Institute of Standards and Technology(NIST) in 2006 and then uncovered as a backdoor (PDF) by a Microsoft cryptographer (Neils Ferguson) in 2007. Many security experts have pointed out since then that the algorithm was orders of magnitude slower than other alternatives, so it never made sense for the NIST to standardize it in the first place - unless it was meant to be used as a backdoor.

Image

Then, in 2013, thanks to some Snowden documents, it was revealed that the NSA paid RSA, an influential security company, $10 million to adopt Dual EC in its BSAFE security suite, and quickly thereafter NIST approved it as a standard as well.


Despite all of this, Juniper said that because it changed some parameters in it, Dual EC should be safe to use, and it will continue to use it in its systems:

“ScreenOS does make use of the Dual_EC_DRBG standard, but is designed to not use Dual_EC_DRBG as its primary random number generator. ScreenOS uses it in a way that should not be vulnerable to the possible issue that has been brought to light. Instead of using the NIST recommended curve points it uses self-generated basis points and then takes the output as an input to FIPS/ANSI X.9.31 PRNG, which is the random number generator used in ScreenOS cryptographic operations.”

According to cryptography and network security professor Matthew Green, and others, this argument makes no sense, as the backdoor will continue to remain a backdoor, just with different parameters, as he explained on Twitter in more layman terms:

Image

Even worse is that Juniper issued a "fix" that merely reverts the attackers' changed parameters to the original values Juniper was using for its firewalls. If those values get discovered by attackers, they could once again gain access to the system.

Matthew Green added in a later article on his Cryptography Engineering blog:

“To sum up, some hacker or group of hackers attacker noticed an existing backdoor in the Juniper software, which may have been intentional or unintentional -- you be the judge! They then piggybacked on top of it to build a backdoor of their own, something they were able to do because all of the hard work had already been done for them. The end result was a period in which someone -- maybe a foreign government -- was able to decrypt Juniper traffic in the U.S. and around the world.

And all because Juniper had already paved the road.”

As Green also noted, this discussion is important to have because over the past few months, the U.S. government has tried to promote the argument that technology companies need to agree to a backdoor in their products, despite the almost unanimous cries from the security community that this would be devastating for digital security everywhere.

Some companies such as Apple, but also Google, have become the main targets of the U.S. government for their support of strong security and encryption. At the same time it’s been calling for security-weakening backdoors, the government has been arguing for increased “cybersecurity,” as a way to pass CISA, the surveillance bill in disguise, which has already been signed into law as the “Cybersecurity Act of 2015.”

“The problem with cryptographic backdoors is not that they're the only way that an attacker can break into our cryptographic systems. It's merely that they're one of the best. They take care of the hard work, the laying of plumbing and electrical wiring, so attackers can simply walk in and change the drapes,” Matthew Green added.

The whole Juniper backdoor situation should put to rest the “good guys” backdoor discussion, because as we can now see not just in theory, but also in practice, there’s no such thing as a backdoor that only the good guys (whoever they say they are) can use. A backdoor can and will be used by any attacker interested enough to look at someone’s backdoored code. As Green said, it also makes it much easier for them to exploit systems that otherwise would be more secure.

We have already asked Juniper Networks twice if there has been any cooperation between the company and the NSA regarding the use of the Dual EC algorithm, but so far it has refused to answer this question. We have also asked the company whether after all of these findings it is ready to completely eliminate the Dual EC code from its systems, because as many security experts note, reverting the parameters back to the company's original parameters is not a solution - eliminating Dual EC completely is. We'll update this article if a reply arrives.

Image

Lucian Armasu is a Contributing Writer for Tom's Hardware. You can follow him at @lucian_armasu.
For a bit of context for anyone who's not been following the IT side of recent governmental shenanigans, various entities and persons in US politics and government have been pushing for law enforcement and intelligence agencies to have the capability to bypass encryption on internet communications. IT companies and general IT-cognizant individuals have been insisting that this was a bad idea due to the fact that such an ability would not remain in the sole hands of government agencies for long and that even providing for that ability to exist requires weakening existing security measures to begin with.

About that...

EDIT: One additional bit of context that I missed, regarding the Dual EC random-number generator. Proper encryption relies on having a strong random number generator, strong being defined as highly random. If you use a weak random number generator (one thing intelligence agencies that want weaker civilian encryption have been trying to push), then someone with intimate knowledge of the tendencies of that RNG can crack encryption that uses it far more quickly. In layman's terms, if I know the RNG used to generate a number between 1 and 10 almost always chooses a number between 6 and 9, I don't have to waste my time guessing 1-5. In actual use it's of course far more complicated than that, but that should get the idea across for contextual purposes.