Balancing Automation With People

We’re well past the tipping point when it comes to automation, and it’s forcing society as a whole to come to terms with what it means today, what it portends for the not-too-distant future and how to overcome complex macroeconomic challenges in the interim.

Depending on your age, class, education and geographic location, the immediate impact of automation across all industry sectors will have varying degrees of impact. While jobs will be displaced as the pace of automation increases, there’s some reason to believe this acutely disruptive and inevitable technology will eventually create more new jobs over a period of time.

Will an entire generation of people be able to hold on and support themselves during this sea change? What must companies embracing automation, AI and machine learning do to speed up this revolutionary transition and make it as painless as possible for the hundreds of millions of workers who will be impacted as these technologies ascend? And if the purported upside of automation for labor never comes to fruition, will the ramifications nullify all or most of the benefits these technologies promised in the first place?

Without the benefit of a crystal ball, it’s informative to look back at the early days of automation for some insight into how this will all play out over the next couple of decades. Whether it’s music, fashion or commerce, each new iteration uses bits and pieces of the themes and processes of years past as the foundation for the next breakthrough advancement – for better or worse.

Back in the 1950s, the post-war boom that propelled the U.S. to the forefront of the world economy was largely built on the convenience and affordability of consumer goods such as televisions, appliances and, especially, automobiles. The American work ethic established during World War II, combined with the perfection of the assembly line approach to mass production, made the U.S. manufacturing juggernaut the envy of the world.

With this economic expansion came friction, particularly as it pertained to the sustained and highly acrimonious relationship between companies and the labor unions created to ensure an equitable distribution of these newfound profits as well as the physical safety and welfare of the workers.

Walter Reuther, a prominent civil rights leader who helped organize and finance the famous 1963 March on Washington, bailed out protesters jailed in Birmingham, and survived a couple of assassination attempts, also built the United Automobile Workers (UAW) into the most progressive and powerful labor union of its time, perhaps of any time.

While the exact details of the story are a bit cloudy, it’s been widely and repeatedly reported that sometime during the early 1950s Reuther was taking a tour of Ford Motor Co.’s latest and most automated automobile factory of the time when either Henry Ford II – the grandson of the founding mogul – or perhaps another high-ranking executive turned to him and said, “Walter, how are you going to get all those robots to pay your union dues?” He replied, “How are you going to get them to buy Fords?”

The UAW still exists and has expanded beyond the automobile industry to healthcare, aerospace, higher education and casino workers, among many others. And Ford, in spite of some bumps in the road, is still selling cars.

Another Ford, Martin Ford (unrelated), a futurist and best-selling author described automation and AI as the next “killer app” in his landmark 2009 book “The Lights in the Tunnel,” laying out the case for how and why these technologies would inevitably render a sizable portion of the world’s workforce obsolete.

While most of these displaced workers would be of the so-called low-skill variety – assembly line workers, servers, drivers, etc. – his 2015 follow-up, “Rise of the Robots: Technology and the Threat of a Jobless Future,” predicts that automation will soon encroach on high-skill occupations including software developers, attorneys and medical professionals.

According to prominent researchers and independent international business organizations, Ford’s ominous predictions appear to be prescient. By 2022, the World Economic Forum predicts that 75 million jobs will be lost as a result of automation.  However, it also claims that as many as 133 million new jobs could be created over this same period as industries master the delicate and unprecedented balancing act of blending machines and humans on the job while simultaneously helping workers acquire new skills for jobs that – for now anyway – cannot be done by robots.

There are, of course, many examples of the current trend toward robotic workers and job automation. Amazon is spending billions on robotics and automation to keep – in concert with human employees – picking and shipping millions of packages each day. MGM Resorts is rapidly rolling out automation in all its casinos and resorts as part of its ambitious MGM 2020 plan. And incrementally, casual dining and fast-food restaurants for the past couple years have rolled out on-table tablets or kiosks for customers to order and pay for their food. The on-again, off-again expansion of self-service checkout stations at grocery stores have become commonplace throughout the world. Subtly but effectively, businesses of all stripes and colors have conditioned customers to this brave new world of automation.

So there’s little point in bucking and protesting against the relentless adoption of automation. Recent surveys show that younger people, and even some older folks, actually prefer less human interaction and appreciate the novelty and/or efficiency of fending for themselves and using technology they’re familiar with to buy goods and services.

The confluence of these technologies combined with rampant consolidation across every industry – and the subsequent elimination of redundant or obsolete positions – has already given the workforce a small taste of what this massive displacement will look and feel like.

And how realistic is this mass displacement? After all, security questions are constantly being raised when it comes to automation and AI. In a comprehensive report released in October, the McKinsey Global Institute found that while automation and AI “promises considerable economic benefits,” it cautioned that these technologies will never live up to their promise if the public loses confidence in it “as a result of privacy violations, bias, malicious use, or if much of the world comes to blame it for exacerbating inequality.”

The authors, James Manyika and Jacques Bughin, credit a number of factors for the maturation of AI and automation including exponentially more computing capacity for training larger and more complex models as well as the sheer volume of data that’s now been collected to train AI algorithms. But while all this computational prowess and algorithm-sharpening will only improve as more and more data is collected, there’s near-universal agreement that AI will still lag far behind humans for decades in the realm of “artificial general intelligence” or “strong AI” — that is, to think generally like a human being and to make decisions irrespective of any previous training and make choices and add value based on information learned over time from an entire lifetime of unique experiences.

What’s at stake here is bigger than quarterly earnings reports and stock prices. But those are the measurements that matter most and move the needle in the here and now. Capital, the stakeholders of these companies, and labor have forever been engaged in a tug of war to satisfy their wants and needs. The long-term societal impact of eliminating or dramatically reducing human workers in favor of robots that can perform 24/7, 365 days a year and never take vacations or call in sick or leave for better opportunities is uncharted territory.

McKinsey researchers note that companies are struggling mightily to find, hire and retain the best and brightest of our current batch of human beings. Some estimates suggest there are fewer than 10,000 people worldwide with the skills and education required to construct and optimize deep neural networks used in AI systems. It’s a pain point that eventually will require some automation of its own.

McKinsey’s data simulations conclude that the outcome for businesses and society as a whole will depend on a couple of factors: the extent to which AI is deployed to create products and services rather than primarily to cut costs and substitute for labor and how AI is used to “manage and smooth” the inevitable frictions that will arise during this transition.

“In general, we found that fear about the risk of unemployment and increased income inequality could be enough to reduce citizens’ welfare,” the authors wrote. “And if the risk aversion were sufficiently large, this would counterbalance the benefits of higher productivity and income to be had from the deployment and diffusion of these technologies.”

In other words, can these companies and industries be trusted to use AI and automation as a complementary part of the workforce – something that would free up so-called low-skill workers to learn new skills to benefit from this wonderful technology and create exciting new industries and job opportunities – rather than a fulcrum to line their pockets, further exacerbate income inequality, and enhance their leverage when negotiating contracts, pay rates and working conditions for the few human beings they may still need to remain productive.

“The dual approach of focusing deployment on innovation and proactively managing the transition, through retraining and other steps to enhance worker mobility, could have a significant upside,” the report said. “Doing good for society in the AI era, according to our analysis, implies that implementing such a positive technology strategy can be good business.”