SpiderLabs Blog

Who's in the Driver's Seat?

Written by SpiderLabs Anterior | Apr 26, 2011 6:23:00 AM

Events over the last seven days have dramatically underlined the pitfalls and difficulties of online security to consumers. To kick off, we had the news that both Apple's iPhone and Google's Droid were keeping rather too much data on their user's physical movements. Then, in the middle of the week, Sony's PlayStation network suffered an outage that was caused, to quote Sony's carefully chosen words, by an "external intrusion." Next up, there was another outage -- cause unclear -- this time with Amazon's much-touted Elastic Cloud 2 service.

The signals we are getting from all this are hardly encouraging. Now that we are well beyond the hype about Web 2.0, its painfully obvious that digital asset protection is in crisis. In my view, it's a crisis so grave that it could potentially freeze both innovation and purchasing patterns, should consumers decide that the risks, whether emanating from a mobile device or a remote network, just ain't worth it.

In fact, it was a fourth news story that drove home to me -- literally, you might say -- that sense of crisis. Increasingly, the cars we drive are dependent on computer technology to manage such critical tasks as braking and steering. And if the last week of online misery tells us anything, it's this: wherever there is a computer and a network interface, there is a gateway for malicious code to come riding through and mess things up.

Says an AP report on the potential for hackers to strike at those behind the wheel: "While there is no evidence that anyone has hacked into auto computer systems to compromise safety or steal vehicles, industry groups are studying the issue in hopes of getting ahead of future cyber-attacks."

We live in a society where we can be shown all of the potential problems with a technology and even a Proof of Concept of the issues, but unless there is crime committed or someone is injured, we shrug it off.

Earlier this year, a group of researchers at the University of Washington in Seattle demonstrated that a cyber attack could remove control of a car from a driver by targeting the electronic control units (ECUs) that run the vehicles' key functions. According to this interesting piece in Scientific American, they did this by inserting malicious software onto a car's computer system, using its Bluetooth and cell phone connections.

"The unending chess match between hackers and security programmers," Scientific American concluded, "has found a new playing field." It's an especially dangerous field from a consumer standpoint, because most people are simply not conscious of the fact that when they are driving a car, they are operating a cluster of computers.

At a hacking conference, THOTCON, in Chicago a few weeks ago, a security researcher (Chris Roberts) gave a talk on continued research into this area. He found similar problem in large farm-grade tractors. He was able to make modifications to the parameters to change the depth of seed placement. Slightly tweaking these settings could results in the difference between 1000 acres of corn or no corn at all.

However distressing it is to have your smartphone or PC hacked, deep down, most people also know that it's a risk associated with your device. Someone else controlling your brakes or cruise control while you are driving? Your produce prices skyrocketing because a group performed mass compromise of tractor config files? Consumers don't expect that. These are headaches that an ailing auto industry and farming community do not need.

Automobile security is, then, a challenge in which manufacturers should move with an urgency that breaks the speed limit. In our Global Security Report for 2011, SpiderLabs warned that over the next few years, mobile attacks may well surpass those against desktops. Are the cars that drive on the roads or industrial equipment like tractors involved in production of a food source the next frontier?

In the business software world, security researchers often find their reports of vulnerabilities, when disclosed responsibly, are welcomed with open arms. They often see fixes developed and implemented in a few short months. In other industries, if anyone actually responds to the email, security researchers are typically placated with a canned response.

Proactive techniques can neutralize these systemic weaknesses. But if manufacturers don't start rapidly adopting the necessary critical measures -- such as creating an organizational standard for all applications, assisting enterprises to take control of the configuration of mobile devices, empowering incident response teams to mobilize at the first hint of a hostile intrusion, and responding to legitimate concerns before "evidence of a real-world problem" exists -- we can expect more weeks like the one just past. And worse.