By clicking “Accept All Cookies,” you agree to the storing of cookies on your device to enhance site navigation and analyze site usage.

Skip to main content

Learning to Work with Our Increasingly Smart Machines

June 14, 2016

A few weeks ago I attended the annual conference of MIT’s Initiative on the Digital Economy.  The day-long conference featured a number of interesting talks on the impact of digital technologies on business, the economy and society.  Tom Davenport – Babson College professor and IDE Fellow – gave one such talk on technology and jobs, based on his recently published book Only Humans Need Apply: Winners and Losers in the Age of Smart Machines, co-authored with HBR editor Julia Kirby.

Davenport started his talk by noting that over the past two centuries we’ve seen three distinct stages of automation, based on the kinds of jobs that were replaced by machines.  The machines of the first automation era “relieved humans of work that was manually exhausting,” making up for our physical limitations – steam engines and electric motors enhanced our physical power while railroads, cars and airplanes helped us go faster and farther.

Next came the automation of jobs involving routine tasks that could be well described by a set of rules and were thus prime candidates for IT substitution.  “Era Two automation doesn’t only affect office workers.  It washes across the entire services-based economy that arose after massive productivity gains wiped out jobs in agriculture, then manufacturing.”  It threatened many transactional service jobs that “are so routinized that they are simple to translate into code,” from bank tellers to airline reservations clerks.

We’ve now entered the third era of automation.  Our increasingly smart machines are “now breathing down our necks…  This time the potential victims are not tellers and tollbooth collectors, much less farmers and factory workers, but rather all those knowledge workers who assumed they were immune from job displacement by machines…” including – as Davenport and Kirby poignantly remind us – “People like the writers and readers of this book.”

Earlier this year Google’s AlphaGo won a match against one of the world’s top Go players.  AI-based machines can now play championship-level Go, assist in the diagnosis and treatment of rare forms of cancer, and navigate our roads as self-driving carsencroaching into activities that not long ago were viewed as the exclusive domain of humans.  “Brilliant technologies can now decide, learn, predict, and even comprehend much faster and more accurately than the human brain, and their progress is accelerating.  Where will this leave lawyers, nurses, teachers, and editors?”

Part of the answer is that most jobs involve an amalgam of tasks or processes.  Some of these tasks are more routine in nature, while others require judgement, social skills and other human capabilities.  The more routine and rules-based the task, the more amenable it is to automation.  But just because some of the tasks have been automated, does not imply that the whole job has disappeared.  To the contrary, automating the more routine parts of a job will often increase the productivity and quality of workers by complementing their skills with machines and computers, as well as enabling them to focus on those aspect of the job that most need their attention.

“Most workers eagerly embrace the machines that save them from the day-in and day-out chores of their jobs that take up time and add nothing to their net knowledge… write Danvenpot and Kirby.  “People want the extra productivity they get from state-of-the-art tools because it frees up capacity for them to take on more interesting challenges… automation of one task after another tends not to be seen as the infiltrating enemy by employees.”

“And neither is it seen as a problem by most customers.  When a task can be performed well by a machine, they prefer it, too.  Obviously, paying customers appreciate when higher productivity means that prices go down; while some people might cherish paying higher prices to enjoy artisanal products and services, most go for the product that does the job at the lowest price possible.  But beyond price, automation often improves quality, reliability, and convenience. When ATMs arrived, customers didn’t complain about the automated option.  By now, few could imagine life without them.”

Many knowledge work tasks are at risk of being automated, warned Davenport.  But while some knowledge workers will lose their jobs, it will likely be on the margins rather than the whole job going away.  Perhaps we’ll need eight lawyers instead of 10.  But, there is no room for complacency, he added, listing several examples of tasks within knowledge jobs that are quite susceptible to automation, including online class content within teaching, e-discovery in law firms, and automated cancer detection in radiology.

So, “what are humans good for anyway?”  What kinds of activities require specifically human skills?  The book discusses a number of potential answers to these questions – including those of eminent mathematician and philosopher Norbert Wiener in his 1950 book, The Human Use of Human Beings, and the more recent views of IDE founders Erik Brynjolfsson and Andy McAfee in their 2014 best-seller, The Second Machine Age.

All in all, four major skills areas emerge where – at least for now – humans are still considerably superior to machines:

  • Expert thinking:  People are very good at pattern recognition, allowing us to imagine new ways of solving problems based on our knowledge of what’s worked well in other areas.
  • Creativity and ideation:  People keep coming up with new scientific breakthroughs, gripping novels, and great new business ideas.
  • Complex communications:  Millions of years of evolution enable humans to broadly interpret a situation and read people’s emotions and body language – skills that are crucial for interpersonal activities like nurturing, coaching, motivating and leading.
  • Dexterity and mobility: Humans are very good at many tacitly learned, common sense tasks, such as being a waiter, which might involve walking across a crowded restaurant, serving a table, and taking dishes back into the kitchen.

In the end, it all comes down to learning to race with rather than against the machines.  Over the past two centuries we’ve successfully adapted to Industrial Age machines. It would have made no sense to look at the Industrial Revolution as a race between humans and steam power to see who is stronger, or between humans and cars to see who is faster.  Similarly, we must now learn to adapt to and work with our increasingly smart machines.

The key, wrote Davenport and Kirby in a related HBR article, is to “reframe the threat of automation as an opportunity for augmentation…  What if, rather than asking the traditional question:  What tasks currently performed by humans will soon be done more cheaply and rapidly by machines?, we ask a new one: What new feats might people achieve if they had better thinking machines to assist them?  Instead of seeing work as a zero-sum game with machines taking an ever greater share, we might see growing possibilities for employment.”

This is in fact what we’ve been doing for the past couple of centuries – from the textile machines at the dawn of the Industrial Revolution to the productivity apps of our digital revolution.  In such an augmentationenvironment, humans and machines support each other, with the machines making the human more productive, and the human ensuring that the computer is doing a good job, is on the lookout for the common-sense mistakes that computers often make, and making sure that the machine keeps learning and improving.

Davenport finished his IDE talk by discussing five key ways for humans to partner with smart machines, so together they can do things much better than either could on its own:

  • Stepping In – Master the details of the system and learn its strengths and weaknesses, including when it needs to be carefully monitored and improved.
  • Stepping Up – Move up above the automated systems, developing big-picture insights, decisions and views that are too unstructured and sweeping for computers to make on their own.
  • Stepping Aside – Focus on areas that humans do better than computers, at least for now, such as selling, motivating people or explaining the decisions that computers have made.
  • Stepping Narrowly – Find an area that is so specialized and narrow that it’s not worth automating.
  • Stepping Forward – Build the next generation of computers, robots and AI tools, using the human talent for thinking outside the box and envision new tools, applications and business opportunities that don’t yet exist.

“Today, many knowledge workers are fearful of the rise of the machines,” write Tom Davenport and Julia Kirby in the book’s concluding paragraph.  “We should be concerned, given the potential for these unprecedented tools to make us redundant.  But we should not feel helpless in the midst of the large-scale change unfolding around us.  The steps are there for us to take.  It’s up to us, individually and collectively, to strike new, positive relationships with the machines we have made so capable.  With our powers combined, we can make our workplaces, and our world, better than they have ever been.”



This blog first appeared on June 6 here.

Instead of seeing work as a zero-sum game with machines taking an ever-greater share, we might see growing possibilities for employment.