Machines Can’t Dream
The scale of AI’s impact is hotly debated, and often boils down to a single question: Could machines replace us? But, rather than obsess over a dystopian future, government, business, and education leaders should work together to minimize the adverse consequences for humans of the robots we do deploy.
WALLDORF, GERMANY – For more than 20 years, advances in artificial intelligence (AI) have been viewed through the lens of competition between humans and machines. Ever since May 1997, when IBM’s “Deep Blue” computer defeated chess champion Garry Kasparov, there has been growing anxiety about the consequences of this cognitive arms race. More recent computer victories – like “Watson’s” 2011 win on the Jeopardy game show, and Google’s 2015 computerized takedown of a professional Go player – have only heightened popular concern.
Leaders, technologists, futurists, and employees in every industry rightly wonder how AI will affect our workplaces, societies, and lives. The scale of AI’s impact is hotly debated, and often boils down to a single question: Could machines replace us?
Smart people disagree about the answer. Some, like Stephen Hawking, believe that AI’s rise represents an existential threat. In 2014, Hawking told a BBC radio audience that “the development of full artificial intelligence could spell the end of the human race.” But others consider such fears overblown, and predict that intelligent automation will lead to a utopia populated by intuitive machines.
One thing is certain: there is nothing to be gained from fearing a dystopian future that we have the power to prevent.
Even in divisive times like the present, human ingenuity, kindness, innovation, and creativity are powering us forward. If we harness this energy, we can create a world in which AI benefits humanity, and where automation frees people from the dangerous and repetitive tasks often associated with manual labor. Instead of machine versus man, then, the world would be better served by embracing an “augmented humanity” – what former Google CEO Eric Schmidt defined as getting “computers to help us at the things we’re not very good at,” and vice versa.
Mehdi Miremadi, an AI consultant at McKinsey & Company, similarly believes that partnership, not rivalry, will define the future of computer-assisted work. And that future is fast approaching. “I think the human-robot interaction is the name of the game,” he said recently. “It will be the most important trend in the near- to mid-term – the next five to 15 years.”
Subscribe today and get unlimited access to OnPoint, the Big Picture, the PS archive of more than 14,000 commentaries, and our annual magazine, for less than $2 a week.
Already, leading industrial manufacturers like Airbus and Nissan are introducing collaborative robots – “co-bots” – to the factory floor. Robots are consistent, reliable, never tire, and don’t improvise. Changes to assembly lines require painstaking human adjustments, making it hard to alter what a factory produces quickly; robots do it faster. But machines can’t do everything, so the solution, according to researchers, is factories in which robots and humans work in tandem.
Leaders in every sector – including government, industry, and education – will need to plan thoughtfully to minimize the adverse effects of AI. Among the most important tasks is ensuring that young people have the skills to excel in an AI-dominated marketplace, and that those already in the workforce can adapt to new requirements.
Robots will not become people’s officemates overnight; only a small percentage of current jobs are susceptible to full automation. Then again, almost every occupation can and will be partly automated at some point in the future. The jobs most prone to automation include routine activities like data collection and processing. But AI’s role in the workplace will deepen with time.
Throughout history, tectonic shifts in technology have upended occupational structures and employment. But these periodic upheavals have also created new, more modern jobs – and often in greater numbers than those that were lost. The coming AI disruption, in other words, will likely have a net positive impact on employment.
Industry analysts predict that AI will drive growth in the coming years and decades. Gartner, a research and advisory firm, estimates that AI-enabled tools will generate $2.9 trillion in business value by 2021, while PwC forecasts that by 2030, AI could contribute as much as $15.7 trillion to the global economy. Some estimates show that companies could save as much as $4 trillion annually with AI automation. And the benefits for businesses will go far beyond savings, as automation will spur innovation, improve forecasting, optimize operations, and lead to better customer service.
In any technological transition, those leading the change cannot lose sight of the human element; there are some things even the smartest machines will never do. Machines don’t dream, set goals, or plan for college. Creativity, curiosity, and emotional intelligence will always be the province of people. Even when fed high diets of data, machines can learn only from the past, and cannot imagine the future. Fortunately, life and business are not purely mathematical chess matches.
When I look to the future, I see a need for thoughtful debate about the social impact of AI. But I also see evidence that humanity’s greatest challenges have always been our greatest opportunities.