I’m just back from a family trip to Chicago, New York and points in between, aided by talking Google maps and other modern marvels, which oddly has me thinking a lot about the role of machines in our lives.
The discussion is often framed as Man vs. Machine, which is alliterative, but also inaccurate. First, to be inclusive, we need to go with Human vs. Machine. Second, I’ve concluded it’s not actually a battle. It’s more a collaboration; an exercise in figuring out how humans and machines can complement each other to make life easier and more fulfilling.
The Human and Machine epiphany hit me while I was traveling with my family on an epic summer road trip consisting of planes, trains and automobiles. Like almost everybody in the 21st century, we lined up flights, hotels, Uber rides, an Airbnb stay and a rental car on the Internet and through mobile apps. We endeavored to find our way around Manhattan and Chicago with Google Maps and public transit apps and sites like HopStop and CTA Train Tracker.
And in the process I was reminded repeatedly of a Harvard Business Review blog post by Andrew McAfee that I’ve mentioned before. McAfee argues — fairly convincingly — that algorithms are simply better than human beings at coming up with the right answer. McAfee says human expertise is important on the front-end, designing the algorithm, for instance, but that human reasoning is almost always harmful when it’s used to second-guess a data-driven conclusion.
It is a hard notion to swallow, as McAfee points out in his piece:
“Of course, this is not going to be an easy switch to make in most organizations. Most of the people making decisions today believe they’re pretty good at it, certainly better than a soulless and stripped-down algorithm, and they also believe that taking away much of their decision-making authority will reduce their power and their value. The first of these two perceptions is clearly wrong; the second one a lot less so.”
McAfee is definitely onto something: There is a balance; a place where the mix of human and machine is optimal — though I’m not sure I’m ready to accept his extreme human hands-off conclusion. (Maybe I’m just one of the stubborn folks he references above.)
It seems to me that machines should be our help-mates. They provide the data we need to make the right decisions for our enterprises. But here’s the thing: Nobody is perfect. And when people aren’t perfect, neither are the machines (think algorithms for instance) that they build. Machines can provide the wrong suggestions because of biases and shortcomings baked into the human-made models that they rely on.
So, the human-and-machine model requires that the humans are aware of the potential pitfalls and their impact on the computer-generated results that the machines yield — all of which makes for better results.
But we also need to be vigilant and willing to assess the results of our human/machine collaborations and tweak the balance between human and machine input when necessary.
There are examples both trivial and tragic of times when the balance has tipped too far in one direction or the other. Consider, for instance, the National Transportation Board finding that pilots’ over-reliance on automated cockpit technology was one contributor to the fatal 2013 Asiana airlines crash in San Francisco.
I plan to pay a lot more attention to the human and machine dynamic in the coming weeks and months. It’s a fascinating tension that has been described in terms of “thick data” and viewed in this story from British publication Marketing as a battle between big data and “magic moments” that are built with the help of true human understanding.
“One of the struggles in wrestling big data into little, magic moments for people is in reserving the time and resources to discover and, importantly, respond to the human stories inside the numbers,” the Marketing story says. “We spend more energy collecting information than listening and responding to it in unusual and surprisingly human ways.”
I’m going to be looking for stories that have at their heart this struggle to figure out when human insight is needed to fully leverage data and when the decisions before us are better left to the machine.
I’m also going to pay a lot more attention to how this tension plays out in my own life. I started on my family’s recent trip, where the tension between the two was evident and where human distrust of the machine surfaced in odd ways (such as my wife, Alice’s, habit, when I was driving, of reading the directions from Google Maps before the automated voice had a chance to say its piece).
In the end, our machine-aided vacation experiences represented something of a mixed bag:
There was our Uber ride to our hotel near Midway Airport. The driver announced that his GPS had given out. I turned to my iPhone to provide directions, including a right turn that our driver insisted was a left turn, despite the clear instructions of my mapping software. He turned left; we got lost. Score one for machine data over human judgement.
And there was our bus trip from Manhattan to Hoboken. My map described our stop as 14th and Bloomfield, but the on-board bus announcement referred to a different cross street. My wife and daughter insisted we get off (based on seeing the intersection we wanted out the bus window). I insisted we stay on the bus and we missed our stop. Score one for human judgement over machine data.
Then American Airlines’ reservation software mysteriously upgraded my wife and daughter Riley’s coach seats, purchased with AAdvantage miles, to first class. My seat, purchased with cash, remained in coach. Score one for the machine, particularly if you ended up in first class. (Which, did I mention, I did not?)
One thing we can be sure of in the swirl of uncertainty is that machines and the big data they crunch are here to stay. Every day you read about a new way to leverage data, machine learning and robotics. (The latest offering, from the Wall Street Journal, is about a new form of advertising in which consumers “converse” with a smart bot that builds a bond between brand and customer.)
Our challenge is to figure out not only how to co-exist, but how to leverage and combine our unique strengths.
Machine photo by Frédéric Bisson and Uber driver photo by Jason Tester published under Creative Commons license. Subway photo by Mike Cassidy.
Mike Cassidy is BloomReach’s storyteller. Reach him at email@example.com and follow him on Twitter at @mikecassidy.