Using statistical data to build models for Artificial Intelligence (AI) sounds like a great idea, however, the real question that businesses need to think about is, “Is it enough to deal with human behaviours?”
In response to the constantly increasing demand for a reliable workforce with lower costs and lesser problems, the dreamers and the realists of the world have replied with AI as the way forward. It’s a logical step considering many of the problems today are the product of human behaviour. Certainly, human behaviour is not the easiest thing to deal with its erratic and varied responses. To a certain degree, these futurists have seen some success in pulling from statistical data collected to build mathematical algorithms to predict certain human behaviours. This is an incredible feat for sure and the future will certainly see more of this. So prepare now, as I say…
One point is that, change, even when it isn’t as drastic as AI replacing human jobs, is stressful on human behaviour enough, isn’t it? Despite one of the reasons for AI being how erratic/dynamic human behaviour is, which our history shows that we can already presuppose resistance to this change, human beings are the overwhelming majority. Acceptance was not an easily solved problem when machines and robots took over manufacturing jobs. In fact, it’s still not an entirely solved problem. Whether or not it is implemented ethically and professionally well, that’s a different discussion. It is however, something that you may want to consider in terms of the consequences of certain decisions.
The other important point about AI is that there are of course a lot things still needed to advance the collection of data as well as its reliability and accuracy to make this a reality. Because frankly, we are very much still in the infancy of such futuristic developments and there hasn’t been enough brave people willing to take the risk and adopt the change. Let’s face it, the whole point of entrusting AI with the job is that it does a better job, is more reliable, and is more accurate. Because if it isn’t, there’s not much point to organizations being OVER-reliant on it so early in its implementation at this point.
Not Quite. Not Yet.
And truthfully, what human behaviour can be accurately interpreted by numbers at this point? Could you take statistical data of the past 2 years of traffic jams and understand where traffic jams come from to predict or solve it? Or how build a prediction model to accurately pinpoint when people will from country A will travel to country B, and how much they will spend? Isn’t that information relayed to human beings to act on? The reality is that the day when computers and AI is able to take over human cognition is still quite some ways off. To do the what we want computer systems to do, we need to understand what are doing, wouldn’t we? And frankly, neither psychology nor the visionaries are where they can definitively say they are there yet.
What that means is that basically, there is still a lot of things we don’t understand about basic human behaviour to even program AI to be useful as we want or expect it to be. What we do know is that data or statistics isn’t enough to understand the intricacies of human behaviour. So in a very short word, the answer is “No”. Or rather, not quite yet. Until the work has been put in to properly understand the dynamic and constantly changing nature of human behaviour, human beings are still the best form of observation and reasoning capable to deal with of human behaviour.
So What Can We Do?
Behavioural science has advanced a great deal in modern times thanks to a lot of great work by Dr. Ekman and many of his peers in helping us understand our emotions and what triggers them. In Dr. Ekman’s case, he’s even tested and applied it into observe the way emotions behave when we lie. Which when you think about it, even in a world where AI provided human beings with information to act on, this would be a new task or skill required of human beings to check if it is genuine or falsified. And many of us aren’t simply don’t understand our own emotions let alone others, or are trained to do the work of evaluating truth and credibility.
The programs Dr. Ekman has designed are geared to provide you a good foundation of the basic understanding of what emotions are, how they are triggered, and how to recognise them in ourselves and in others. This allows you to then apply it to enrich the way you read and deal with others. The research behind it has been widely accepted which gives you accurate, reliable scientifically supported data as you make better decisions in the way we perceive communication and our interactions with others.
Click here to:
Let us know what articles you would like us to read about regarding emotions and lie detection.