October 08, 2020 CYBER SECURITY, INTERNET OF THINGS

If You Can Connect It, Protect It

Let’s talk about the future of connected devices.  About how if you can connect it, then you need to protect it.  Allow me to explain.  The future of the Internet of Things (IoT) has to be that manufacturers view the security of their devices more seriously.  There’s no other way around it.

First of all, these devices come with default passwords which are normally published on the Internet since users need to download manuals to install or configure them.  Once installed, they never really update themselves, and chances are slim that users might update the software on their own.  For me, I counted at least 20 connected devices in my home and I’m not even a fan.  In fact, it took me years to purchase a Nest thermostat or a Ring doorbell.  Just recently, I was shopping for refrigerators and came across the Samsung model with a large tablet on the door.  You know , the one with which you can create a shopping list and then send it to your phone? Tell me, who’s ever going to say, “I need to update my fridge??“!!

Forget it.

That aside, IoT isn’t confined only to home gadgets as numerous other industries use portable devices too.  Think hospitals with their instruments and medical carts.  Or refineries with an entire Operational Technology (OT) department dealing specifically with IoT issues.  Sure, these are more likely to be updated with manufacturer passwords changed since someone’s job would otherwise be on the line.

All these devices have one thing in common – well, perhaps more than one – but this one thing is certain.  They’re all based on embedded Linux.  Microsoft tried, a decade ago, to fight an impossible battle by introducing Windows ME.  And lost.  Brutally.  Why?  Well, let’s be real here, nothing beats free.  Plus Linux is far superior to any other personal OS, let alone the poorly conceived child of Windows that was ME.  The issue with using embedded Linux is that there’s no inherent security.  Even though Linux is great to build security (case in point: all Firewalls today are based on Linux), that doesn’t mean it’s secure straight from the box.  Or that embedded Linux is secure.

Not at all.

To be clear, embedded means it’s been stripped of everything that isn’t strictly necessary.  A move that’s meant to reduce digital footprints so it fits into very small flash drives.

Yes, I’m oversimplifying.

And intentionally so since my focus is on the security aspects.  I can’t stress enough that if you can connect it, then you absolutely must protect it.  There’s no other way around it.

Same OS for all (translating into likely the same vulnerabilities as well), default passwords, no updates.  The worse cybersecurity scenario, all in one same place.  Try telling a CISO, “I’m going to give you a network of 1000 devices.  They’ll all run on the same operating system, unprotected.  They’ll have the same vulnerabilities.  Same default passwords and no, they can’t be updated.”  I guarantee you he’d quit.  Right there and then. And yet, this is today’s reality for IoT.

This is the primary reason why wired devices are such a cybersecurity nightmare.

Statistics abound on how hackers are exploiting these gadgets to launch attacks. One such act of aggression which catapulted IoTs to the forefront was a DDoS that involved a million such devices.  All of which had been taken over by the one same botnet.  Because yes, once you take over one of them, they’re just computers.

No more, no less.

Computers that can be reprogrammed for just about any purpose.  The computing capacity these devices hold is unimaginably vast, and certainly much larger than needed for their original purpose because there aren’t any smaller CPUs around.  Just as I can’t really purchase a 500GB HDD these days, I’d also have a hard time acquiring a ‘slow’ CPU.  They simply aren’t in the market anymore.  Manufacturers utilize what’s available.  What often means is  quad cores with computing power of Herculean proportions.  Think of it as though we’re putting more computing power into a light bulb than we’d need to go to the moon. Imagine that!  And when the security isn’t sufficient (and, by the way, it never is), things go really, REALLY wrong.

So what’s our solution to this?

In a company environment, that CISO, if he hasn’t resigned yet, may attempt to bolster the devices with security updates.  He might change passwords, and possibly even protect access to them with proper firewalls, IPS, etc.  But that’s not going to be the situation in a home or a small office where you’ll typically barely even find an unattended firewall.  And definitely zero security expertise.  In such an environment, those devices remain unsecured and as vulnerable as when they came out of the box.  The only solution is for them to be secure from the time they leave the manufacturing plant.  Straight out of the box, as the expression goes.

One other idea is to get rid of default passwords.

Generate some type of QR code allowing the user to “find” the access code online.  Or write that unique password on the box itself.  Hold on, don’t scream just yet.  This isn’t as bad an idea as it may sound.  If the password to my Ring doorbell was written on the cardboard box it came with, no hacker would know that.  It’d be a random, long string password that was automatically generated when the device was manufactured.  And then printed on the box.  Who’d see it once it was installed?  And even if you saw it at the store, what good would that do?  The analogy would be like finding a door key in the park.  You wouldn’t know which house it belongs to and as such, it doesn’t pose a threat.  On the flipside, that Ring doorbell now has a password that’s long, random, and doesn’t need to be changed.

Updates too can be automated.

My ISP already does that for their router.

Once in a while, especially in the middle of the night, that thing sounds like it’s being possessed by demons.  The first time I heard it, I thought it’d been hacked!!  Then I learned that it was the ISP running an update.  My security dashboard does the same, and one time, they decided to do it in the middle of the day just as I was trying to turn it on because I needed to leave the house.  Very annoying.  I had to wait 5 minutes, sitting on my couch.  Patiently watching that thing go through the motions, until it was done and ready again.  That being said, I did feel assured that at least those devices were being updated.  I’m fairly certain some of the updates were security related and that’s good.

Manufacturers must take ownership of updating these devices.

They cannot leave it to the general public because that simply isn’t going to happen, leaving them (and their owners) completely vulnerable.  This is what Microsoft had to do with their Windows computers.  This is also what IoT manufacturers will need to do if they wish to ensure their devices aren’t taken over by hackers in the time it takes to say ‘hacked’.

In the interim, if you can, change all the passwords to all the devices in your house and update their firmware.

Yes, I know.

I can only dream.