Van Robotics (copy)

Abii, a robot designed by Columbia-based Vän Robotics to tutor elementary school students in math, will be commercially available in the fall with 700 robots in production.

The prospect of learning math from a robot that gets better at teaching the more lessons it gives is either awe-inspiring or terrifying, depending on how many episodes of “Black Mirror” one might have watched in the past few months.

Of course, the cute, smiling math tutor looks more like C-3PO than a Terminator, and its limited “machine learning” capabilities are basically just a way to figure out which lessons work best for which students — not a way to plot the destruction of humanity.


But the real insidious thing about Abii, the $999 robot designed by a Columbia-based company, is the apparently unskeptical enthusiasm with which it has been received.

Nonrobot math tutors, of course, already very much exist. Without having tested Abii myself, I feel confident that a decent human math tutor would still generally outperform even the most advanced nonhuman teacher.

And $999 will buy quite a few tutoring sessions with the average college student or starting teacher. They could use the extra cash.

That gets to the broader problem with automation and the reckless, often illogical push to embrace machine learning and rudimentary forms of artificial intelligence without considering the consequences or even — ironically in this case — doing the math.

So far, there are surprisingly few jobs that robots can consistently perform better than humans. There are even fewer jobs they can do without fairly intensive human supervision. And in so many cases, adding robots to the mix means cutting human wages without saving any money overall.

Consider the case of self-checkouts. They’re not a very new technology, but they have been widely embraced. Some customers apparently prefer them to human cashiers, but their overall benefit to business owners is dubious at best.

Self-checkout machines are incredibly costly, requiring an up-front investment several times higher than the average full-time retail employee salary. They also tend to lead to an increase in shoplifting. And they still require human supervision, albeit with fewer employees than might otherwise be necessary.

The end result is that business owners save little if any money while providing an arguably less personal shopping experience and pumping less cash into the immediate local economy by employing fewer workers. How is that progress?

Ditto the concern that self-driving cars will replace taxi drivers — or more recently, Uber or Lyft drivers — food delivery drivers and truck drivers.

To a certain extent, self-driving cars already exist. But they don’t work very well, so they need humans in the driver’s seat just in case something goes wrong. It’s going to be a very long time — and many billions of dollars in research and development — before that’s no longer the case.

So the question again is why are we spending so much time and money on a questionably useful technology that would eliminate millions of jobs instead of simply working on tools, such as the safety features that self-driving cars already depend on, that would help humans do those jobs more easily and more safely?

Get a weekly recap of South Carolina opinion and analysis from The Post and Courier in your inbox on Monday evenings.

Or better yet, why don’t we spend more time and money on a society in which cars — self-driving or otherwise — are less necessary for routine tasks?

The cost-benefit equation of trying to replace human drivers but not cars just doesn’t make much sense.

None of this is to suggest that we shouldn’t keep working on smarter machines and better ways to teach math and a simpler way to buy groceries. But a relentless focus on technology too often overlooks the obvious solution — the human beings already doing that work.

Spending a little more money on improving workers’ lives rather than trying to replace them with machines would almost certainly be better for corporate bottom lines and customer satisfaction in the long-term than, say, a kiosk that rings up a cucumber.

And while overworked math teachers and perplexed parents would probably welcome any help available — even if it’s from a robot — they have at least a few advantages that Abii won’t ever be able to replicate. Like real brains and real hearts.

Ed Buckley is an editorial writer with The Post and Courier.

We're improving out commenting experience.

We’ve temporarily removed comments from articles while we work on a new and better commenting experience. In the meantime, subscribers are encouraged to join the conversation at our Post and Courier Subscribers group on Facebook.