Meet Baxter, an industrial robot built by scientists at the Massachusetts Institute of Technology (MIT).
Baxter looks like a pretty typical robot, with large bulky arms and a screen for a face. But what makes this robot special is the way it is controlled – via the human brain.
In order to control Baxter, humans must first wear an electroencephalography (EEG) monitor, which is a sci-fi cap that measures brain activity.
We’ve already used EEG technology to communicate with other humans, most notably patients who suffer from “locked-in syndrome”. However, most of the time the use of this technology requires a period of training. With Baxter, the researchers at MIT came up with a system that responds to natural human responses.
In this case, Baxter actually responds to someone thinking that it screwed up.
And while it sounds like the researchers may be trying to shame poor old Baxter, they are actually looking to push-the-envelope regarding how we communicate with robots.
As of 2017, we have two primary ways of talking to our metal friends: voice and touch. The problem with both of these inputs as that they take time. By reading our minds, robots will be able to respond to requests in milliseconds.
Baxter works by responding to brain signals called “error-related potentials” (ErrPs), which are signals our brains release when noticing mistakes.
Baxter is given a simple task – in this case sorting paint and wire – while a human subject wears an EEG. The human and Baxter are connected so when Baxter detects an ErrP, it corrects its behaviour. If no ErrP is detected, then Baxter continues along the same path.
For now, the technology only works in binary – meaning that Baxter can only distinguish between “right” or “wrong”. Sorting wire and paint is an extremely simple task but could be the first step to more complex human-robot mind-based interaction.
Imagine being able to instantaneously tell a robot to do a certain action, without needing to type a command, push a button or even say a word. A streamlined approach like that would improve our abilities to supervise factory robots, driverless cars, and other technologies we haven’t even invented yet.
Mind-reading robots sound great, but it is easy to imagine doomsday scenarios.
For example, we’d have to be very careful in telling the robots which thoughts to follow. We all have negative thoughts occasionally. At times, perfectly normal people may suddenly think of doing something destructive or harmful. They are called “intrusive thoughts” and according to psychologists such thoughts are totally normal.
The French (of course they do) have a fancy word for these sudden dark thoughts: “l’appel du vide” which means the “call of the void”. Try saying it while fully dressed in black and smoking for full effect, like a proper existentialist.
Perhaps intrusive thoughts are unaccompanied by detectable brain signals, and we won’t have to worry about them. What we do know is that the brain is extremely complex, so reading more complex processes accurately is far from simple.