Nicholas Weaver Explains how QUANTUM Works

An excellent essay. For the non-technical, his conclusion is the most important:

Everything we’ve seen about QUANTUM and other internet activity can be replicated with a surprisingly moderate budget, using existing tools with just a little modification.

The biggest limitation on QUANTUM is location: The attacker must be able to see a request which identifies the target. Since the same techniques can work on a Wi-Fi network, a $50 Raspberry Pi, located in a Foggy Bottom Starbucks, can provide any country, big and small, with a little window of QUANTUM exploitation. A foreign government can perform the QUANTUM attack NSA-style wherever your traffic passes through their country.

And that’s the bottom line with the NSA’s QUANTUM program. The NSA does not have a monopoly on the technology, and their widespread use acts as implicit permission to others, both nation-state and criminal.

Moreover, until we fix the underlying Internet architecture that makes QUANTUM attacks possible, we are vulnerable to all of those attackers.

Posted on March 14, 2014 at 2:01 PM10 Comments


Clive Robinson March 15, 2014 6:09 AM

It puzzels me that the NSA should use, let alone use by prefrence, QUANTUM (mind you the Greatfirewall of China and others supposadly use similar).

It is easily detected and would not be difficult to defend against in many ways. Which begs the question of “reliance” and what the NSA do when the “sea state” changes, which it finaly appears to be.

The first thing to note is QUANTUM is “old tech” in computing terms it’s three to five generations behind, thus we need to think about how we would do similar today and more importantly how we would stop it.

As I and others have noted befor comments on this blog about similar attacks actually pre-date the TAO noticably enough that some have joked about copyright violation by the NSA.

And it’s an important observation for various reasons, because it indicates the NSA has limitations on what it can do that actualy ‘puts it behind the curve’ in the field, not ten or thirty years ahead as some would have you beleive under the Myths and Legands. The question is “Are those limitations real or imposed?[1]” and “Does it matter as long as they are in place?” as we are fighting todays enemy not tomorrows. But the age old question arises of “why fight if you are clever enough not to?”

You can only be attacked if you can be found, so sometimes the best defence is to put yourself out of harms way as Mr Greenwald observes in his article the problem is,

    … the NSA’s malware gives the agency unfettered access to a target’s computer before the user protects their communications with encryption.

To do this the NSA have to first get the malware onto the target’s machine and contrary to “myth and legand” they are not omnipitent or omnipresent beings as this quote shows,

    In many cases, firewalls and other security measures do not appear to pose much of an obstacle to the NSA. Indeed, the agency’s hackers appear confident in their ability to circumvent any security mechanism that stands between them and compromising a computer or network. “If we can get the target to visit us in some sort of web browser, we can probably own them,” an agency hacker boasts in one secret document. “The only limitation is the‘how.’”

And there is the crux of the NSA problem like vampire lore they can only cross your threshold if you invite them in, in some way. That is the NSA need to trick ‘the target computer into communicating with an NSA controled computer’ if it does not then the malware cannot be put onto the computer remotely.

I suspect this problem is not lost on the NSA, in the same way it’s not lost on the marketing industry. After all you can only send an RSVP invite if you know where to post it, so the only solution is to “broadcast it” in some method which might include leafleting every residence there is via “junk mail”. And it’s this later strategy the NSA appear to be addopting. Even so “junk mail” is known to have a poor return rate hence the prefrence for “targeted marketing”. At the end of the day all of these NSA aproaches are taken from the field of marketing which is all the cyber-crooks have done as well, so again the NSA is “behind the curve” on this as well.

Whilst “junk mail” is as a minimum a “refuse issue” riseing to anoying most of us know how to deal with it when it drops into our letter boxes, we rarely read it much less act on it. Thus the marketing approach can be defeted when it comes to our physical letter boxes. However due to the very poor state of our computer software the NSA can “post-a-roach” that climbes out of your letter box and scurries to the nearest cover to then attack your home. Again in the real world we have solutions –use a post office boxes and open your mail in a cafe etc– we need to translate this solution into an online equivalent.

As discused on this blog in the past there are many solutions to these various problems you can do individualy, however this still leaves the “upstream issue”. Currently the biggest upstream problem with the Internet is the “all roads lead to Rome” issue of it’s current physical construction and routing mechanisms with “choke points” located in the Five Eyes nations. Avoiding these choke points is by no means impossable but most –if not all– need “out of band” solutions that just don’t exist or are not practicle or flawed in some way. Until we solve the upstream issues the NSA does not need to be ahead of the curve, nor even on it, marginaly behind is just fine.

[1] The fact the NSA appears “behind the curve” is one I’ve thought about on a number of occasions and as has been pointed out by Nicholas Weaver often they hold themselves back by tripping over their own feet for various beauracratic or compartmentalisation issues.

Gweihir March 15, 2014 10:26 AM

One source of the problem is, of course, unsafe habits in the first place. If I do an SSH login to a host I have logged into before, no amount of NSA QUANTUMing will help them one bit. At best, the can DoS me and become obvious. Now, the very existence of things like (Open)SSH tell me that this is not a new risk at all and was well understood at least 2 decades ago.

The other issue is of course that the IT world has failed to remedy the utterly pathetic state of software security for a long time now.

Some level of vigilance is necessary, or totalitarianism creeps in in one form or another. The safeguard democracy is supposed to offer only works if people vote intelligently, and at need for new political parties. I see basically nothing significant happening in that direction.

Nick P March 15, 2014 11:04 AM

@ Gweihir

“Now, the very existence of things like (Open)SSH tell me that this is not a new risk at all and was well understood at least 2 decades ago. ”

Goes back even further. The risks of networked computers were mentioned in the Anderson report (1972) that created INFOSEC field. He outlines specific problems, solutions of the time, and options they were considering. The problems he mentioned still apply today. The decentralized Internet has added plenty more compared to their terminals and timesharing machines.

Other historical papers for curious readers:

Nick P March 15, 2014 11:07 AM

Edit: Forgive me, the Ware report created the INFOSEC field in terms of asking the questions. Anderson report was the first to try to answer them.

Chris Abbott March 15, 2014 9:19 PM

@Gweihir, Nick P

IT security is pathetically weak, and I’m surprised it took this much to wake people up to that. It’s been horrible for years. I just did a job for a law firm that got pwned by a former contractor and potentially dozens or more case files and privileged atty/client info were taken. What needs to happen is that the IT world desperately needs to make security easier to use or the default for the average joe. It’s easy for us, but helping my friends, family, girlfriend, and customers setup and be able to use crypto and security software has been like banging my head against a wall. There’s a real lack of tools out there for the non-tech people.

Amy Higgins March 21, 2014 11:33 AM

QUANTUM must be stopped. It’s just awful. Every time I run Windows update I fear that my computer is being spied on. What can we do about this?

Jarth April 24, 2015 10:47 AM

Encryption ? What about ipsec-like encapsulation based on some dynamic scheme ? Provides identity, integrity but will not freak out as many people and not create a false sense of security.

Leave a comment


Allowed HTML <a href="URL"> • <em> <cite> <i> • <strong> <b> • <sub> <sup> • <ul> <ol> <li> • <blockquote> <pre> Markdown Extra syntax via

Sidebar photo of Bruce Schneier by Joe MacInnis.