One of the big things governments around the world want is a backdoor into end-to-end encrypted messages and many are considering trying to force the big tech companies to provide such backdoors.
Governments present what, on the surface, sound like reasonable arguments in favour of such things. They talk about catching terrorists and child molesters and most of us would like to see such people caught, but if you scratch away the surface, beyond the emotive arguments, things are altogether less clear.
Let’s have a look at what they mean by end-to-end encryption and why it’s necessary. This is an extremely simple example that ignores things like public-private key systems and many of the intricacies related to the way these systems actually work.
If I send a plain, unencrypted message to you, it’ll first go to my ISP where it’ll be bounced around a few servers, then it’ll find its way to an internet backbone where it’ll be bounced around a few more servers, then it’ll find its way to your ISP and eventually your computer.
Just getting from my computer to google.com requires seven hops to interim servers and each one of these interim servers is a place where someone could read our unencrypted message via a so-called man-in-the-middle attack.
If we get fed up of people snooping on our messages we might decide to use end-to-end encryption. I could send a key to you by one method — carrier pigeon for example — and then I’ll encrypt all my future messages with that key. Now all the man-in-the-middle would see is gobbledegook but you could decrypt the message at your end because I’ve previously given you the key to do so.
This isn’t a bulletproof method of security. Someone could still intercept the carrier pigeon involved in the initial key exchange, make a note of the key and then send the carrier pigeon on its way so that we’re none the wiser (although pigeons have no road sense and it would probably have been run over anyway). They could also attempt to decrypt our messages by using brute force to find the key, although if we’ve chosen suitably complex keys we could make it unreasonable for them to do that in any realistic time.
An attacker could of course just hack into either of our computers and read the messages before or after transmission but that would be too time-consuming for mass surveillance. Although if we used the method I described that might be the only way they’d read our messages because intercepting carrier pigeons might be tricky. There are also a number of other ways we could mitigate some of the danger but I won’t go into them here.
What a government’s mass surveillance programme wants is a way to be this man-in-the-middle and, if a service controls both the key generation and message transmission processes, a suitable backdoor could theoretically be provided. The service provider could essentially be the man-in-the-middle for its own service and then pass the necessary information on to government spooks.
The thing is, if you purposely build in a back door, it’s there for everyone. The WannaCry-based ransomware that struck computers in 2017 exploited a backdoor Windows had provided for the NSA, for example. It’s easy to say such backdoors will only be available to governments but hackers specialise in finding backdoors.
A government may say this is a risk worth taking to catch terrorists, child molesters and other criminals, but is it really? The thing is, you may be able to strong-arm big tech into providing these backdoors but what will criminals do then? They’ll go to smaller independent services. So then you might make it illegal for any company to provide encryption without a backdoor, but criminals would then use apps from Russia, China or somewhere else where a nation’s long arm of the law just isn’t long enough. Or the criminals will develop their own messaging systems, just as ISIS did.
There are wider issues too. Do we really want to allow a government to snoop like this? In our western democracies we may have been lulled into a false sense of security by the relatively stable governments we have, but if history teaches us anything it’s that these things don’t last. And aren’t programmes of mass surveillance the sort of thing despotic governments aim for? Haven’t the terrorist already won if they’re forcing us into this?
There is a price to pay for freedom and our involvement in the many wars that have resulted in the loss of many lives shows we’re willing to pay that price. It seems odd that we’re planning to sabotage that from within, particularly as I doubt this sort of mass surveillance would have stopped many attacks anyway, and it certainly won’t when criminals get wise to it. What you’d be left with is a system that merely monitors all the innocent people, the criminals having gone elsewhere. In fact, the criminals will be the only ones with effective security systems.
So even if you think mass surveillance is a good idea in principal you’re unlikely to realise the perceived benefits of it, although you’ll still get all the downsides. It may provide some short-term results of course, but they won't last.
Even a former MI5 boss thinks that weakening encryption is a bad idea. Of course your current intelligence leaders want mass surveillance — it makes their jobs a little bit easier — but the decision ultimately lies with governments.
I don’t believe our government is actually planning anything nefarious here. There is just a woeful lack of technical understanding within governments around the world. That was probably most apparent when, talking about his own mass surveillance plans, the Australian Prime Minister declared the laws of mathematics don’t apply there:
The laws of Australia prevail in Australia, I can assure you of that. The laws of mathematics are very commendable but the only law that applies in Australia is the law of Australia.