Slashdot Mirror
English High Court Bans Publication of 0-Day Threat To Auto Immobilizers
An anonymous reader writes "The High Court — England's highest civil court — has temporarily banned the publication of a scientific paper that would reveal the details of a zero day vulnerability in vehicle immobilisers and, crucially, give details of how to crack the system. Motor manufacturers argued that revealing the details of the crack would allow criminals to steal cars. Could this presage the courts getting involved in what gets posted on your local Bugzilla? It certainly means that software giants who dislike security researchers publishing the full facts on vulnerabilities might want to consider a full legal route."
It sure is a good thing that England controls the entire Internet and that no one anywhere will be able to publish this information now.
I'm an American. I love this country and the freedoms that we used to have.
It's standard practice, when publishing about security flaws, to alert the producer of the products affected before doing so openly, only publishing when a) the hole is patched, or b) if they are ignoring the issue and refusing (or at least taking too long) to fix it.
If they have not given the manufacturer a reasonable amount of time to fix the problem, I can understand why they're being censored - it's unnecessarily dangerous. However, if this is simply the manufacturer trying even harder to pretend the problem doesn't exist, I would of course object strenuously, and support publishing the hole because that will not only force them to get a fix out ASAP, but will punish them for taking so long.
And, while TFA doesn't say either way on the issue, I would expect the latter, not the former.
Take a look at this year's Black Hat presentations. These are just the ones on vulnerabilities in embedded systems.
Security through obscurity does work, not very effectively, but it does. Or at least, the obscure system is more secure than the same system that is open.
For example - let's say I keep a backup key to my house buried somewhere in the yard or in a flowerpot ( there are many flowerpots and I chose one at random). While this is not as secure as not having the backup key, it is more secure than placing a sign indicating where the key is.
Same thing here - while the system is not as as it would have been if the vulnerability did not exist, if the exploit was published, then everyone would know how to hack it, even those who would not be able to come up with the hack on their own.
My car is too old to have a computer in it, but I use an aftermarket security "system" - I have to push a button (the button is visible and usually has another function) before I try to start the engine or it would crank, but not start. Now this would not be a problem for a competent thief - he would figure out how to circumvent this, it's not that difficult. However, some drug addict or a drunk teenager may just conclude that the car is broken and steal some other car instead.
Generally temporary injunctions like this are just until there is a full hearing. Volkswagen will probably have a fix in place by then, but the main purpose is to avoid doing irreversible damage until there can be a full hearing on the facts.
A temporary injunction is common in many types of cases and in no way indicates the court's opinion on the substantive issues. It's simply a recognition that they can't unpublish the information, so they need to wait until a decision is made before they publish. The same is often done with property disputes such as divorces. A temporary injunction orders both parties not to sell or otherwise dispose of the property until a decision is made as to ownership.
Ps - I don't care for the injunction. I would have preferred that the court hint at whether they think the case has merit, then let the researcher decide whether to release the information immediately, risking a successful suit for damages. The injunction, as a prior restraint on speech, is censorship. Still, it's best not to exaggerate the effect of the or intent of the injunction.
On the other hand, as these researchers learned, if you notify the company, they can get a court order against you. If you let the cat out of the bag without notifying them them, they can't really stop you. And if you figured it out, there is a good chance that the company knows about it already anyway. They simply don't have any incentive to correct it unless they know that the general public knows about it too.
I'm an American. I love this country and the freedoms that we used to have.
Yepp, the court fell for the oldest and most blatantely false argument of the full disclosure opponent.
The court assumes that bad guys don't already have this knowledge. From decades of experience in IT security we can conclude with near certainty that they do. What this provides is limited, short-term protection against those would-be thieves who don't, yet. Also, a false sense of security.
What would've happened if this had been published: The public would know, car manufacturers would (have to) scramble for a fix.
What will happen now: Nothing. The next model will be fixed, your current one will maybe get an update at the next maintainance cycle, but don't count on it.
The next years will be a great time to be a car thief.
Assorted stuff I do sometimes: Lemuria.org
Have a recent BMW? There is a known vulnerability where you can copy an actual key inside the car, using the data in the car's computer and the car's own transponder. BMW has not fixed this and won't fix it. The vulnerability is that BMW relied on being the only source of blank, programmable keys and having all the programming equipment in house. Once someone reversed the key system (the car itself contains unprotected, unencrypted key strings), they found out what electronics to put in the key and made blank keys and software to program them using the keys found in the car's computer. This is a massive problem that was out for probably at least a year before there was enough public attention to the enormous theft of BMWs with that system. I think that the number of BMWs stolen had quadrupled in that period. Right now, since BMW won't fix it, getting a BMW that suffers from this vulnerability is prohibitively expensive to insure, making their second hand value very low. It may be that insurers now require 3rd party alarm systems to be installed or something, I don't know, but the vendor didn't fix it and basically left their customers without a solution.
Right now, there's no indication that VW can and will fix this problem once it gets out. I highly doubt they will recall all vehicles and replace the parts that are vulnerable with a system that has the flaw removed. For all we know, that could cost thousands per vehicle and apply to all VAG cars from the last 10 years. That could be over 100M cars, worst case. Then again, if it'd only apply to a certain model and year and it is an affordable fix, they may actually do it, but I wouldn't count on them fixing anything.
I was promised a flying car. Where is my flying car?