Here's The Most Obvious, Terrifying Flaw In Google's Self-Driving Car Prototype: The 'Panic Button'

Most people are slightly scared of Google's new self-driving car prototype because the driverless vehicle has no steering wheel or brakes. You just push a button and it takes you where you want to go. It's like sitting in the car equivalent of one of those airport monorails.

But it's not the lack of brakes or steering wheel that makes the cars potentially terrifying. In theory, a world of software-driven cars programmed not to bump into anything would be a lot safer than the roads are today - full of drunks, texting teens, and humans who make fatal mistakes at the wheel.

Rather, it's the apparent lack of a manual backup system to those automated brakes and steering wheels. Instead, Google has installed a "panic button" - literally a red "e-stop" button for emergencies.

In a regular car, if the brakes fail then a human driver has several other systems she can manipulate to bring the car safely to a halt. She can shift into a low gear; she can put the car in neutral and let it coast; she can yank the handbrake; she can steer around oncoming cars while the car runs out of momentum; she can steer into a skid to avoid a spinout (easier said than done, admittedly). In a worst-case scenario, she can point the car at a human-free object - like a grassy median - and crash the thing (relatively) safely.

In Google's new vehicle, none of those manual backups exists. You just hit the button and hope that it works - even though the only reason you're hitting the button is because it has stopped working properly.

Here's how Google describes it:

They won't have a steering wheel, accelerator pedal, or brake pedal… because they don't need them. Our software and sensors do all the work. The vehicles will be very basic-we want to learn from them and adapt them as quickly as possible-but they will take you where you want to go at the push of a button. And that's an important step toward improving road safety and transforming mobility for millions of people.

One parallel here is airplanes. They can fly vast distances on autopilot. Systems will even kick in to override the pilot and "drive" the plane if the pilot makes a mistake, like pointing it at the ground. But pilots still have the ability to take control of a plane. The two systems - human and non - work in tandem.

But not at Google's car division.

This is likely to slow the regulatory changes needed to get widespread adoption of driverless cars. The obvious bottom line is that while we may one day accept - and even enjoy - being driven around by robots, almost all of us would prefer the idea of being able to grab the wheel or hit a brake pedal in the event that the machine goes on the fritz. After all, we've yet to develop a mistake-free machine of any kind.