Update on robots and AIs: Will Robots Take Our Jobs?

The Pew Report

Recently we did an eight part series on the boundaries between the human and digital world. The first segment was on the new reality of robots. This is a field that keeps moving, however, and recently the debate about whether robots will end up taking our jobs was fueled anew by Pew Research releasing the results of canvassing a group of experts. Though the survey was really just a summary of expert predictions, it got a lot of press because some 48% of these experts said that robots will, indeed take away a lot of jobs.

A couple caveats, though. First, the obvious corollary that 52% of experts surveyed don’t think robots will take our jobs did not get any headline space in any of the articles reporting on the survey. Second, the word “expert” should be taken with a grain of salt. The list of “key respondents” for the survey is very impressive, but they are not weighted against the much longer list of people solicited from list-serves and through associations, few of whom are trained in all the fields that would go into making broad sociological/economic predictions.

Still, it did revitalize the debate—which is very important. After all, if robots do take our jobs, what will we do? How will we provide people incomes necessary to distribute goods and services? How will we manage the potential for even greater income inequality? Will people find meaningful ways to live their lives without traditional jobs?

Looking at the opinions of some of the key respondents may be the best way to tease out the respective arguments. Some of the more optimistic thinkers are making some form of the argument that technology always creates jobs at about the same rate at which it destroys them. Vint Cerf is among the thinkers who believe this, which gives the position a lot of credibility.

However, some of the “optimists” put faith in the less compelling argument that certain jobs can only be done by humans. While this may be true, that number is certainly smaller than the number of jobs overall. Also, Pew groups into this 52% those who believe that the timeline given (2025) is too short to have an effect on the job market. This is a lot different than saying more jobs will be created than destroyed. There is also a group who claim that our “social, legal, and regulatory structures will minimize the impact on employment.” To me, this position seems to except that robots will take our jobs, but says we will take measures to deal with it.

On the other side of the discussion, we have people like Jerry Michalski, founder of the Relationship Economy eXpedition (REX), who says “Automation is Voldemort: the terrifying force nobody is willing to name.” In his view the race between automation and human work will inevitably be won by robots. Tom Standage, digital editor for The Economist, points out that this wave of innovation is different from those in the past, and there is no reason to assume it won’t change society. “Previous technological revolutions happened much more slowly, so people had longer to retrain.” Mark Nall, program manager for NASA, amplifies on this: “robotics and AI are different. Due to their versatility and growing capabilities, not just a few economic sectors will be affected, but whole swaths will be.”

Our Take

There are some pretty compelling reasons to think that the so-called pessimists have it right this time.

The assumption that innovation always creates more jobs than it consumes is not based on any fundamental law. It is just an observation of the recent past 60 years or so. The rate of change now is different than anything we’ve seen over that time.
Digital automation has the quality of being scalable and repeatable over wide sectors. The old joke was that “a computer took my job. Well, a computer, two programmers, a system engineer…” But the current reality that two programmers can create solution that, if it replaces one job, can just as easily replace a thousand.
Automated solutions, whether they be programs or robots, are easy to upgrade and alter on a massive scale. Your computer probably downloads updates to various programs at least once a week. Our current education system and social structure is not equipped to retrain a workforce at anything approaching this rate of change.

On the other hand, we don’t think that robots taking our jobs is necessarily a pessimistic viewpoint.

Jobs can be rich, challenging, fulfilling. But many aren’t. The only reason people do them is to earn enough money to meet their needs. If we can meet the needs of these people some other way, than good riddance to those jobs.
Our current job model of working from 9 to 5 (or 6, or later) is very artificial. It is not particularly good for many workers, and doesn’t lead to real productivity. Instead of losing whole jobs, society could easily scale back and share some of the more interesting jobs among several workers.

All of this will require some way to redistribute goods and services to meet the basic needs for everyone, as well as provide opportunities for those who want to grow, learn, and do more. But if this were to happen, the robotic future might be an optimistic future instead of a pessimistic one.