Preventing Your Body From Getting Hacked With Visual Cryptography
Face, eye and thumb scans are forever. A technology called “visual cryptography” may offer the best hope for keeping them protected.
2017 seems likely to go down as the year when we all finally realized that passwords, social security numbers and other familiar login credentials were obsolete. The Equifax breach, which exposed the private data of almost 45% of the U.S. population to hackers, was a tipping point. Added to a numbing list of multi-million account break-ins at Yahoo, SONY Entertainment, Zappos, Experian, The SEC and Target (to name just a few), Equifax simply made it undeniable: By now, your sensitive personal data has almost certainly fallen into the hands of some bad actor.
Banks, hospitals and other entities that store your personal data are responding with systems that use a different kind of login credential: your body. Through the use of biometrics — a technique where your face, palm, ear, iris or some other body part is scanned and stored, allowing you to login only when you step in front of a scanner again — they hope to gain control over a data security crisis, which is causing over $16 billion in losses per year due to identity theft, according to Credit.com. But even biometric scans must be stored in a database somewhere, and that’s given rise to fears that they can be stolen just like passwords and social security numbers. A unique approach called visual cryptography, however, makes it possible to encrypt scans in a way that renders them useless to a hacker even if they are stolen.
Actually, biometrics are already used worldwide by banks, and here in the U.S., Apple has introduced fingerprint and facial recognition login on its phones and mobile payment apps. According to an FBI adviser quoted by NBC News, more than 500 million biometric scanners will be in use worldwide by 2019.
However, after being burned by the old login methods, some iPhone users are apprehensive about using this new login approach. That’s because biometric data is forever. If someone hacks your face, you can’t go and create a new one as you can with a password. There have, in fact, already been some large data hacks like the 2014 one at the U.S. Office of Personnel Management, during which millions of records were stolen — including many fingerprint scans.
Fortunately, some key signs indicate that biometric scans can be made a good deal more secure than passwords — if they’re encrypted and stored in the right way.
There are two points of potential security failure for biometric scans: at the capture device (which could be anything from a camera at a reception desk to a mobile phone) and in the database where the scan is stored. Efforts to secure the “front end” where the scan is captured have had uneven results. Recognition of a face, for example, can be “spoofed” by bad lighting, the face of a twin or by something as simple as holding a photograph in front of a scanner.
Trying to fool scanners to login to individual accounts would be incredibly laborious for a hacker, though. “Back end” database hacks — like the Equifax one — are potentially far more damaging. They can allow for theft of huge volumes of data in a heartbeat, sometimes without detection.
Biometric scans can offer some unique security options simply because they are not lines of text, like usernames and passwords. They’re most secure when stored as digital records — only lines of ones and zeros. That allows for a kind of encryption called “visual cryptography” that poses real problems for hackers. In this approach, the scan that’s taken of your eye, face, fingerprint or any other feature is digitized into a series of ones and zeros, and then surrounded with a mass of other code. If a hacker accesses your scan in the database, he literally finds that there is “no there there.” According to London-based Veridium, a provider of visual cryptography, your biometric information appears as a useless mass of digital white noise.
Front End Protection
When you go to login to your account by stepping in front of an eye scanner, that scan is encrypted (surrounded by a mass of confusing code) right there in the scanning device. It’s then sent to the database, where it’s laid over your stored scan file, revealing only the elements that are common to both. The junk code is cancelled out and your scan is authenticated. But an unencrypted version of your scan is never stored. Even if a hacker gains back end access to, say, a hospital database containing your health information, he can’t find a usable login credential for your account.
Imagine the word “hello” written in red and surrounded by thousands of other colored dots to the point where you cannot read it. Now take another file with the same word in red, also surrounded with different dots that make it unreadable. Only when you overlay the two files can you see the element that’s common to them both: the word “hello” in red. This is essentially how visual cryptography protects biometric scan security. It’s an added bonus that during the matching process, the scans move across the Internet in such a heavily encrypted format that even a “man in the middle” attack will, again, only provide the hacker with a bunch of digital gibberish.
No technology can protect against the kind of weak management oversight that appears to be at the root of the Equifax hack. The company’s executives have made the almost incredible admission that a good deal of customer data was stored as plain text with no encryption whatever. But because biometric security is a new tool being integrated in more up-to-date databases, it should provide a level of protection that all the numbers, characters and letters in a password cannot.