What will the future of work look like and why is it important?
I’ve asked myself that question a lot of times since I started investigating how artificial intelligence (AI) is beginning to automate tasks which seemed impossible a few decades ago.
Bit by bit, AI is eating its way through work that we’ve considered exclusively human as long as we can remember.
Driving a car. Managing employees. Composing music. Writing emails and booking meetings.
These are all tasks that most people would still consider too complex for a machine to do properly.
Yet, we’re already seeing how self driving cars are threatening millions of driving jobs in the US alone in the coming decades. Companies are building software capable of hiring and firing employees, delegating tasks and ultimately replacing middle management. As a secretary, you’ll no longer need to book meetings for people as this can already be done by a bot called Amy.
Even a thing so uniquely human as composing music can also be done automatically with the help of AI. You simply pick a style of music, a mood and the length of the track you want and voila! The song magically appears in a few seconds. It may not be a beatiful work of art, but it’s definitely on par with many of the other mediocre human made beats out there.
And we’re only seeing this as the beginning of a new technological revolution.
So where does that leave us as human beings?
I asked a couple of AI researchers for their insight, and their message was actually encouraging for all those who deal with art, creativity, empathy and human connection in their work.
As it seems, this frontier will be harder for machines to pass than we’re sometimes led to believe.
A Paradigm Shift That Impacts Everybody
From the outside, it’s easy to see technological development as a series of gradual improvements.
The self driving car is not a new concept, you may rightly point out. Already in the 1960’s, there were experiments with self driving cars — the reason they didn’t get famous was that they didn’t really work that well.
And that’s just one larger example. Automation can also be about the smallest things such as an app which turns on the lights when you enter your home so you don’t have to push a button.
From this perspective, self driving cars as we know them today or the bot that books meetings for you can easily be mistaken as refinements of previously commercialised technologies.
Grasping the difference is crucial to understand how your job and industry could be affected in the future and what to do about it.
The main difference lies in how these different systems operate.
Some of the first self driving cars weren’t capable of making complex decisions. They were actually just appearing to be autonomous vehicles while in reality they were simply following magnetic cables embedded in the road.
In other words, they weren’t intelligent.
The same thing goes for the app which turns on the lights when you enter the home. It simply makes a pre-programmed response to a pre-defined condition (when door opens, turn on light).
On a very basic level, this is how computer systems have operated up until recently (and still do in most cases).
The main point is that all the decisions the computer make, has already been laid out in advance by the programmer. The human stand behind the wheel.
Today’s artificially intelligent systems work in a completely different way.
Instead of saying: When this happens, do that, programmers train AI systems to learn in a way which for simplicity’s sake can be compared to how we learn as human beings.
By showing a neural network thousands of images with cats, it may learn to distinguish a cat in new images.
The programmers do not explain each feature of a cat, such as the shape of the head, ear and whiskers. No, they simply show many different examples of cats and let the system learn by itself how a cat looks.
As you can imagine, this way the computer can see patterns that the programmer hasn’t even thought about. This is actually what happened during an epoch-making Go match between Lee Sedol, one of the world’s best Go players, and AlphaGo, an artifically intelligent system created by the Google-owned AI company DeepMind. During a pivotal point of the game, AlphaGo made a move which puzzled everybody watching the match, including its human opponent and the experienced commentators, who thought the move to be a mistake.
Only it wasn’t, because AlphaGo ended up winning the game and ultimately the entire match.
Apparently, by watching millions of moves from expert players, the AI system had ‘seen’ an opportunity which even its makers couldn’t have foreseen.
In a nutshell, this move shows how systems being built in this very moment are able to learn even complicated tasks and do them better than humans.
But luckily for us, that’s not the whole story. Although it’s easy to think of artifical intelligence as a primitive version of human intelligence and then extrapolate from that, it’s not the right way to think of it.
As it turns out, as humans we have unique capacities which AI is nowhere near replacing. And deciding to tap into this potential is what will define us in the years to come and decide for us whether a computer can replace our work or not.
To learn more about how we as human beings can leverage our strengths and beat the robots, tune in tomorrow. It’ll take a surprising turn when it shows that beating the robots is a lot less complicated than we think and it actually is more about not beating them but simply learning to live with them in a new way.
Be sure you don’t miss a new episode in my blog post series by subscribing below.
If you care, please tell me what you loved and what you hated about this post and why. Feedback is sincerely appreciated! In fact, it’s the only way I can improve.