In the wake of the Westminster attack, Amber Rudd took to the BBC’s Andrew Marr Show to revive David Cameron’s attempt to associate WhatsApp and encryption with terrorism. As then, Rudd’s comments have been roundly rejected by technology companies, internet security experts and most of the media. Remarkably, Rudd didn’t even pretend to fully understand the technology – describing experts as “those who understand the necessary hashtags to stop this stuff even being put up”.
Deciphering her statement isn’t straightforward, but she appeared to have two main points. Firstly, a more general sense that technology companies have a “responsibility to engage with government, to engage with law enforcement agencies”. Google recently faced a backlash as the Cabinet Office joined several companies in pulling their adverts from YouTube after it emerged they were appearing alongside extremist material. When it comes to publicly hosted extremist material, of course tech companies must act responsibly and in accordance with local laws.
Rudd’s other point is far more specific – perhaps more specific than even she realises – and far more problematic: “we need to make sure that our intelligence services have the ability to get into situations like encrypted WhatsApp.” WhatsApp currently use end-to-end encryption, which makes this technically impossible.
Why would services like WhatsApp want to adopt end-to-end encryption? While many tech companies are public champions of privacy, I suspect the main reason is that they would rather avoid the messy fallout of a major hack. Imagine waking up to the headline: “WhatsApp hack sees billions of private conversations being leaked online”. As with the Ashley Madison hack, it wouldn’t be long before websites sprung up providing a convenient search functionality. Soon, anyone could look up conversations had by friends, colleagues and public figures.
You might think that tech companies like WhatsApp would be relatively safe from attack — but last year alone saw data breaches from Yahoo, LinkedIn and Tumblr.
You might think that tech companies like WhatsApp would be relatively safe from attack – but last year alone saw data breaches from Yahoo, LinkedIn and Tumblr. Currently if hackers did get access to WhatsApp’s servers, any private messages they could access would be encrypted and virtually impossible to decipher.
Perhaps it’s worth returning to Rudd’s own slightly muddled request: “There should be no place for terrorists to hide. We need to make sure that organisations like WhatsApp, and there are plenty of others like that, don’t provide a secret place for terrorists to communicate with each other. It used to be that people would steam open envelopes, or just listen in on phones when they wanted to find out what people were doing — legally, with a warrant.
Rudd would like to frame the debate as one that poses privacy against security. Many people might sympathise with the example of privacy being broken in very specific situations – in the interests of national security, for example, and with a legal basis. But for WhatsApp to make such a request technically possible, they would have to remove end-to-end encryption entirely – thus fundamentally weakening the information security of each and every user of the service.
Amber Rudd focused on WhatsApp because it is one of the most popular platforms, with billions of end-to-end encrypted messages sent every day. It’s worth saying that there are plenty of alternatives, including apps such as Signal, Silent Phone and Telegram. But really all these platforms just add a pleasant user experience to something that Amber Rudd has no hope of defeating: mathematics, and the internet. With a small amount of technical ability, it’s possible for anyone to communicate using end-to-end encryption without involving a large tech company at all. OpenPGP, for example, makes it possible for anyone to end-to-end encrypt their email – completely outside of the control of their email provider (e.g. Gmail).
Removing the privacy provided by end-to-end encryption in the name of security is not a fair trade. And even if it was, neither government nor technology companies can stop anyone dedicated to online privacy downloading their own encryption tools. So if the government does one day get its way, let’s just hope the terrorists know even less about internet security than Amber Rudd.
for the geeks: how encryption actually works
End-to-end encryption is complex in the detail, but remarkably simple in principle. I encountered very rudimentary cryptography when passing secret notes in the school classroom. Starting with the assumption that a note could be intercepted by a classmate or teacher, we shifted each letter of a message up by a specific number of characters – three, for example. Let’s say our message was “Big Brother is watching you”. The “b” would become an “e”, “i” would become “l”, etc. Our final message would read “Elj Eurwkhu lv zdwfklqj brx”.
Modern encryption works in a very similar way. We have a “key” – in the above example, the key is “three”. And then we have an “algorithm” – above, the algorithm is “shift every letter of the message up by the number indicated by the key”. If you want to communicate securely with another party, both need to know the algorithm and the key.
Of course, my example is incredibly easy to break – there are only 26 different possible combinations. Over hundreds of years cryptographers have designed algorithms that are far more complex, and we now use keys are hundreds of characters long. Correctly implemented, breaking modern encryption would require an impractical amount of computing power. My favourite crypto-geek t-shirt simply says “trust the math” – without the correct keys, it is as good as impossible to break anything protected with modern encryption standards.
Encrypted communication with most online services – social networks, search, retailers, banks – involves a key shared by the user, and the service. Imagine I want to update my Facebook status, perhaps locating myself on a map in the process. My update is encrypted on my device, before travelling over cellular networks, WiFi and physical cables – all of which are potentially vulnerable to interception by malicious hackers, local law enforcement and international intelligence agencies. Once it arrives, Facebook uses their key to decipher my message and store it in their own database – if anyone else is permitted to see that content (e.g. a friend), their device requests and receives that update using their own key.
For a while, the default on many messaging services like Whatsapp was to use a similar system. When they received a private message they deciphered it, then relayed it to my contact – using a separate key. Whatsapp could, if they so wished, read those messages, store them in a database, and supply them to third parties (such as law enforcement). However, private messenger apps have been realising that they don’t need to access messages sent privately. Instead, they facilitate an exchange that only I and the recipient have the key to decrypt. WhatsApp have no way of reading or accessing my messages, even if they wanted to.
Credits
Text John Ridpath
Image via Flickr