New York Times blogger Nick Bilton writes, in “Disruptions: As New Targets for Hackers, Your Car and Your House“, that researchers at the DefCon conference reported hacking into hybrid cars and aggressively influencing their behavior. The blog says
The researchers completely disabled a driver’s ability to control a vehicle. No brakes. Distorted steering. All with a click of a button.
Eek. Even more vivid descriptions can be found on the Forbes.com site by a journalist who was driving a hacked car and experienced the effects first hand. It’s definitely frightening, but strikes me as a little overblown. The current round of hacks require that the hacker have a device physically connected to the diagnostic port on the car, so it’s not like any old baddy with a smartphone can crash your car. On the other hand, wireless attacks on cars have been reported in the past. Colleagues in my former department showed how to mess up a car by communicating with its wireless tire sensors from bluetooth transmitters by the side of the road. So, while people are not reporting that any of these life-threatening interventions are happening in the real world, it should give one pause.
In addition to sounding the alarm about breaking into cars over the Internet, the article cites some risks of using Lockitron, an online door lock that I have on order, online lighting like the Hue I have, and Internet-controlled refrigerators. There’s a real concern that Internet-enabled user-programmable devices of the kind I advocate could open us up to some serious malicious mischief if designers don’t keep security at the forefront of system design.
The line that convinced me it was worth making my first blog post in many weeks, however, was in the context of the INAX wireless toilet. This state-of-the-art commode sports, not one, but four automatic features. Here’s the description from their site:
- Automatic Lid & Heated Seat: When you approach the toilet, the lid opens and the heated seat is activated.
- Sound Module: When the lid automatically opens, music from the sound card will begin to play and the deodorizer will be activated.
- Automatic Flushing & Deodorizing: When you step away from the toilet, it will flush automatically.
- Self-Closing Lid: When you are finished, the lid closes automatically, the deodorizer deactivates and the air purifier will activate emitting ions to cleanse the air in the room surrounding the bowl.
Actually, that sounds kind of awesome. Maybe not worth the $5600 price tag, though. And it’s even more expensive if you want to control these features, as there is a $500 remote control center that they offer. Happily, these features can also be controlled from your smartphone! Unhappily, the bluetooth security code built into the toilet is, by default, set to ‘0000’. So, anyone with the free app can control any toilets in the vicinity. Oh, ick. Now, although I wasn’t able to verify it, the Times blog post claims that the company has issued a patch, and summarized the situation with this gem:
Yes, in the future, you will need to download security updates for your toilet.
That cracked me up.
Unfortunately, in the process of researching the topic for (I thought) a quick blog post, I ran across an article that envisions what happens when homes are only as secure as the passwords people use to secure their website access. In short, it summarizes the various physical items that are being put online and pairs them with what we know about the history of attempts to secure websites, which hasn’t been stellar. One of this posts concluding thoughts about the security situation for controllable devices:
The rush for vendors to compete in this extremely fast moving market will inevitably result in rushing aspects of the product design and we know very well from past incidents that security is one of the areas most frequently overlooked in favour of delivering features.
Ok, that makes me a bit paranoid. As more stuff goes online, developers and end-users will get sloppy and opportunities for troublemakers to hurt people and property expand tremendously. It seems kind of inevitable.
Although I never watched Battlestar Gallactica, I saw most of the pilot. A critical plot point is that all the ships in the defense fleet have all their weapons disabled by a backdoor in the control system software installed by the enemy. The only battlestar that survives the vicious attack is the one with the paranoid captain who refused to trick out his ship with the latest in software control.
Boy, I sure hope there’s another choice between refusing to build programmable systems and getting exterminated. I would really like to have my programmable devices and my life. Perhaps there’s a role for universities to create a lightweight free open security package that could be offered to developers everywhere? Maybe, if it’s easy enough, consumers will demand a certain minimal level of security in the products they buy? Perchance such developments might help stave off the doomsday scenario? Help?