In too many accidents the ‘human error’ is at fault. Autopilots and Artificial Intelligence are developed with the aim to diminish the frequency of accidents. At the same time, only humans can deal with certain unexpected situations and find ways to manage them. These are two sides of the coin of the ‘human factor’ and we are struggling to reconcile them.
One of the issues of the Fourth Revolution is that the border between the intelligent automation and the area which still requires active human input is moving fast. Commercial aircraft flying is already largely automated. In a few years, automobile driving will be automated to drastically lower accidents. Still there will always be some situations where humans need to take over because they go beyond what the automation can deal with. And this means that increasingly, humans will largely monitor automated systems and at the same time will be required to be able to deal with extraordinary situations.
This is because humans are expected to be able to deal with a wider, more systemic view of the situation and find a way to move forward. This type of intervention will be prone to a high rate of failure in particular if there is not too much time to analyse the situation. Still we will continue to rely on this human intervention in extreme cases, sometimes with unsatisfactory results.
Finding exactly how humans can contribute best and setting up the right ergonomics so that this intervention is effective is a key area of research.
How strange is it that at the same time we complain loudly about the fallibility of humans and still expect them to deal with those extra-ordinary situations automated systems can’t deal with!