There was much angst surrounding AI as it loomed as a potential part of daily life, even among the so-called AI experts. But is it warranted? Physicist Niels Bohr is famously reputed to have said, “prediction is very difficult, especially about the future.” And decision scientist Philip Tetlock confirms this sentiment, claiming that most experts are subject to the same decision-making biases as the average human—and don’t have special inside knowledge—and therefore can be prone to frame blindness leading to inaccurate predictions.
For example, nine years ago, Elon Musk predicted that driverless cars were only a year or two away—rendering truck and taxi drivers to the employment scrap heap. Similarly, when spreadsheets were first promulgated in the 1980s, accountancy was labeled a sunset profession. Likewise, the emergence of the internet was associated with predictions of the death of retail shops—notably supermarkets—as many believed we would be buying everything online and waiting for the delivery to arrive at our doors.
The reality is, however, much more complex. New technologies are rarely perfect substitutes for existing activities. They typically make aspects of existing tasks easier, which allows the opportunity for reshaping the activities of those undertaking them. Vacuum cleaning in my house no longer requires me to hover over the hoover, but it still requires someone to prepare for the robot to do its task—clearing up the dog toys and taking the dog for a walk so he doesn’t get freaked out, or worse: chasing it all over the house. So, just as much time is spent on the vacuuming task, but productivity is increased by doubling it up with the dog walking. Thus, none of the household labor-saving innovations of the past two hundred years have (despite their promises) made housekeeping redundant. Rather, they have allowed housekeepers to do many other things besides.
This is starting to become clear as AI becomes more prevalent in our daily lives. Just as was illustrated when computers and the internet became ubiquitous, obtaining the productivity gains required rethinking and rearranging existing uses of time (and equipment) to sustainably lock in productivity gains.
Thus, spreadsheets did not make accountants redundant, but spreadsheets did change their role. Rather than just prepare accounts and budgets, accountants could now explore the economic effects of a wide range of strategies. Reducing marginal costs—in this case, using spreadsheets—makes it possible for workers to do more or engage in different value-creating opportunities that weren’t possible when time (the scarce resource) was required for more mundane tasks.
While fears remain, evidence suggests that though headcounts are dropping in certain industries, this isn’t necessarily because AI is replacing entire jobs. Rather, in challenging economic circumstances, existing workers can use AI tools to do their existing jobs more efficiently. Meanwhile, new tasks and industries emerge as the freed-up labor can be deployed to devise new ways to develop and utilize AI tools further—like accountants with the introduction of the spreadsheet. Or, they can do a greater volume of tasks that still require human effort in the same amount of time—such as walking the dog while vacuuming is automated.
The prospect for new industries is the most exciting. But to properly consider this, one needs to take a long-term view. For the past two hundred years, new forms of energy have led to the replacement of nearly all physical labor by machines. This has enhanced productivity, but also has a cost—the physical health of the laborers replaced. A healthy physical life requires the human body to exert more energy than is generally required for fundamental economic and social sustainability. Thus, entire new industries have emerged. Gyms now proliferate to meet daily fitness needs, and an ideal vacation—which used to be relaxing on the beach, for example, to recover physically—is beginning to be replaced with activity-based excursions: cycling in Europe, hiking in the Himalayas, or taking the Camaro pilgrimage.
As AI starts to replace mental tasks, it begs the question of whether the transition will come at the cost of thinking skills essential to a healthy human existence. If it does—and evidence is tending towards this conclusion—then it is likely that new industries will emerge, as did gyms and active tourism, to meet those needs. This may already be occurring in the fast-growing puzzles and games sector.
Or maybe, as John Maynard Keynes once predicted, with increased productivity the average working week will be reduced to fifteen hours—enabling humans to focus on maintaining their physical and mental health in their newly-reclaimed leisure hours? My dog, I suspect, is betting on this prospect.
The post AI Is Changing—Not Stealing—Our Jobs and Lives appeared first on American Enterprise Institute – AEI.
