What skills will set you apart in the age of automation? | David Epstein | Big Think

In a rapidly changing work world, it's important to be a constant learner to be able to change and evolve your skills. Especially when you're facing automation of certain types of work. So, I want you to think about a spectrum of work to get automated.

On one part of the spectrum is chess. Chess is based on rules; there are very clear patterns that repeat. That is a great situation for computers. Computers are really good at patterns, which is why they have made exponential progress in chess. Now, the chess app on your iPhone can beat the best human chess player in the world.

In the middle of the spectrum, maybe you think about self-driving cars. We have made great progress there; there are rules of the road, and there are regular repeating patterns. But there are significant challenges that remain.

At our far end of the spectrum, we have something like, say, cancer research. IBM Watson had a lot of hype but actually underperformed that hype in such a way that I talked to AI researchers; some of them worried that they would damage the reputation of AI in health research going forward. As one psychologist I talked to put it, the reason Watson triumphed at Jeopardy but failed in cancer research is that we know the answers to Jeopardy.

So, if you want to have skills that continue to be valuable, you have to keep learning things. You have to be in some of these more amorphous fields. Almost so, I want to share one example of how this has played out in the past. When ATMs were created, the thought was that this would do away with bank tellers for good, right? Bank tellers did repetitive transactions and so you would not need them anymore.

But in fact, as more ATMs came online, there were more jobs for bank tellers. What happened was that each branch needed fewer tellers. So, each of the banks became cheaper, and banks opened more branches. Thus, there were more tellers. However, the job of tellers changed completely; it was no longer someone who could do repetitive transactions. Rather, they had to learn marketing skills and customer service and have this much wider array of broad skills.

Because those broader skills and integrating different types of information are what differentiate humans from computers. The psychologist Robbin Hogart categorized domains of learning as going from the kind to the wicked kind. Learning environments where areas of patterns repeat, there is a wealth of previous data, there are clear rules, and feedback is apparent.

In those kinds of areas, like chess, computers really thrive. On the other end of the spectrum are wicked environments where not only is information unclear, but rules don’t necessarily repeat. People aren’t waiting for each other to take turns; feedback may be delayed, and if you get it all, it may be inaccurate. Human behavior is involved.

Those are areas where computers don’t do as well. There are quite a lot of so-called soft skills: how to deal with human behavior and how to adjust to things that are changing in real time and interpret signals that are very difficult to quantify. That’s an area that’s very, very difficult for computers, but humans have a huge advantage.

So, those kinds of soft skills are really important and will be for a long time to come.