Digital interactions need more human skills, not less.
Turning away from R2-D2 and looking for C-3PO, robotics can teach us a lesson on our own human skills.
For those of you who have succeeded in avoiding the efficient Star Wars marketing over the last 40 years, these are the names of the two robot characters featuring in the 7 movies of the franchise. R2-D2 is the one looking like a dustbin on wheels and speaking a synthetic language that humans in the future seem to have learned. C-3PO is the humanoid one, who speaks fluent English, albeit like a 19th century butler.
Humanoid computers
Robots are all the craze right now and they seem to be the next embodiment of the extraordinary transformations computers are causing in our daily lives. We had “one computer on each desk”, then “one smartphone in each pocket”, will we have “robots everywhere”? I was recently in a conference about robots and I finally got the answer to a question that was bothering me since 1977: will future robots be more like R2-D2 or C-3PO?
It turns out that the first type is already here: R2-D2-like robots are already used in many places to perform a limited number of tasks they have been programmed – and physically designed – for. This last part is important. These non-humanoid robots, once produced, can only perform the limited set of tasks they have been designed for in the factory. One can improve their software to slightly alter the way things are done, they can even learn themselves how to do it better but not much more. As an example, do not expect your self-driving car to tuck you into bed when you’re ill anytime soon.
C-3POs, however, are the ones receiving most of the investment of tech companies right now and the reason is simple: they will be able to interact with the world we have created around us to perform – any – task. Our kitchens, our offices, our factories, our walkways have been created so that we can interact with them and perform any task. Just think of the difficulties disabled persons face when trying to interact with the same world. Humanoid robots, when available, will be able to be programmed or trained to do anything we can, with the same hardware.
So, for very practical (and economic) reasons, future robots will gradually be more human-like, physically looking more like C-3PO than R2-D2.
Human-like communication
Will this physical constraint extend to the way robots will communicate in the future? Without a doubt. It is established that communication is only for a small fraction happening in the words we say but more significantly in the tone of our voice, our body language and our behaviour. Universities and corporations are investing huge amount of capital to improve the way computers, and soon robots, will be able to accurately read people, cognitively and emotionally, and then interact appropriately with them.
At the same conference I mentioned before, I already had the opportunity to interact with a small humanoid robot able to “read” emotions in my face and adapting to it with the colour of its eyes, its posture, a few words. And the main selling argument for using it in old people nursing homes was: “it is so cute it gets adopted right away”! So we can be sure, we’re getting there.
In the meantime, we mere humans are still around, gradually losing our repetitive tasks and activities to R2-D2s but still conducting most of the interactions with our environment and our peers while waiting for C-3POs to come. Our societies need to prepare themselves to this future with many questions requiring an answer right now.
One question I submit: isn’t it ironic that while investment is poured into making robots more skilled in human interactions, companies are doing little to make us humans more skilled in human interactions? Face to face or through digital media, there is no doubt that efficiency requires a high level of quality in these interactions to sell then deliver value.
When so much effort is put into making computers more human-like in their communication, can managers still pretend that these skills are not important? When most predictions announce that it will work with robots, can managers still pretend that these skills cannot be taught to humans?