The Human Factor in an AI Future
Artificial intelligence (AI) grows more sophisticated and better able to perform human tasks with each passing year. That requires us to wrestle with the ongoing implications, not just for business, but for humanity as a whole.
From the first hammerstone to factory automation, technology that reduces or even eliminates physical and mental effort is as old as the human race itself. However, that doesn’t make each step forward any less uncomfortable for the people whose work is directly affected—and the rise of AI is qualitatively different from past developments.
Until now, we developed technology to handle specific, routine tasks. A human needed to break down complex processes into their component tasks, determine how to automate each of those tasks, and finally create and refine the automation process. AI is different. Because AI can evaluate, select, act, and learn from its actions, it can be independent and self-sustaining. Its ability to decide what to do without our direct input or control makes us not just uncomfortable, but vulnerable. An economic system in which not even the machines need us will force us to rethink our most basic definitions of what it means for a human to be valuable and valued, and to adjust our society accordingly.
Where does AI leave humanity?
Any job that involves routine problem-solving within existing structures, processes, and knowledge is ripe for handing over to an AI. Indeed, jobs like customer service, travel planning, stock trading, real estate, and even clothing design are already increasingly automated. Analysts and academics are focusing intently on how AI will impact tomorrow’s labor market, and many researchers are projecting that unlike earlier waves of automation that predominantly affected lower-education, lower-wage workers, AI will also impact well-educated and well-paid “knowledge workers.”
Although we’re still a long way from computers that can pass flawlessly as people, it was wrong to assume that it would take computers decades or even centuries to catch up to the nimble human mind. The exponential explosion of deep learning is letting AI chat with us in an almost-human way, diagnose illness better than a doctor, and maybe even lie to us before too long.
The concept of the technological singularity—the point at which machines attain superhuman intelligence and permanently outpace the human mind—is based on the idea that human thinking can’t evolve fast enough to keep up with technology. However, the limits of human performance have yet to be found. It’s possible that people are only at risk of lagging behind machines because nothing has forced us to test ourselves at scale.
Other than a tiny minority of individuals, most of humanity has had little choice but to spend its time meeting survival-level needs. Most people don’t have the time or energy for higher-level activities like scientific discovery, art, music, or philosophy. But as the human race faces the challenge of potential obsolescence, we need to think of those activities not as luxuries, but as necessities. Even as technology nibbles away at the repetitive tasks that most humans have traditionally done to earn a living, we have an unprecedented and urgent opportunity to determine the unique value humanity offers, and to cultivate the uniquely human skills that deliver that value.
Honing the human advantage
As a species, humans are driven to push past boundaries, to try new things, to build something worthwhile, and to make a difference. We have strong instincts to explore and enjoy novelty and risk—but according to psychologist Mihaly Csikszentmihalyi, these instincts crumble if we don’t cultivate them.
AI is brilliant at automating routine knowledge work and generating new insights from existing data. What it can’t do is deduce the existence, or even the possibility, of information it isn’t already aware of. It can’t imagine radical new business models. Or ask previously unconceptualized questions. Or envision unimagined opportunities and achievements.
It doesn’t even have common sense! As theoretical physicist Michio Kaku says, a robot doesn’t know that water is wet or that strings can pull but not push (unless you tell it). Nor can robots engage in what Kaku calls “intellectual capitalism”—activities that involve creativity, imagination, leadership, analysis, humor, and original thought.
At the moment, though, we have only begun to prioritize these so-called “soft skills” in our education systems. We still largely expect people to hone their emotional intelligence, cross-cultural awareness, curiosity, critical thinking, and persistence organically, as if these skills simply emerge on their own given enough time. But there’s nothing soft about these skills. They are necessary, they can be taught, and they’re the critical difference between human intelligence and the artificial kind.
Lessons in being human
For humans to have value in an increasingly AI-operated future, we need to cultivate our most human abilities within society—and to do so not just as soon as possible, but also starting at as young an age as possible.
Peter Diamandis, founder of the XPRIZE global competition for innovation, advocates revamping the elementary school curriculum to nurture skills we haven’t considered critical: that passion, curiosity, imagination, and persistence other thinkers have identified, as well as ethics. It’s not necessarily a new idea to teach children that they should take risks and make mistakes because failure is how we learn, and that understanding another point of view is more important than being right; Waldorf and Montessori schools have been encouraging similar approaches for decades. But in a world full of problems too big and complex for individuals to solve on their own—climate change, world hunger, pandemics, rising authoritarianism—we all need to be able to respond creatively, collaborate more than we compete, unlearn our assumptions, and remain open to what we don’t know. And the imperative to help people acquire these skills is more relevant and urgent than ever.
The Mastery Transcript Consortium is approaching the problem from the opposite direction by working backwards from their desired outcome of ensuring every high school student graduates with creative, critical, and analytical abilities. The organization is pushing to redesign the secondary school transcript to better reflect whether and how high school students are acquiring the necessary combination of skills by measuring student achievement in a more nuanced way than through letter grades and test scores. This approach inherently requires schools to reverse-engineer their curricula to emphasize those abilities.
In the age of AI, everyone will need creativity and critical thinking skills because those are the things only a human can bring to the table. If we decide that we don’t care whether most people can thrive or even survive without those skills, we’ll be writing off a significant percentage of the population as disposable—and thereby doom ourselves to political and social instability as people become desperate to escape economic and humanitarian disaster.
On the other hand, if we rethink our assumptions about education and the economy based on the idea that everyone has unique human abilities worth nurturing, we can retain options for everyone to earn a living and maybe even find innovations in places and people where we never looked for them before. While we let artificial intelligence get better at being what it is, we can get better at being human. That’s how we’ll keep coming up with groundbreaking new ideas like jazz music, graphic novels, self-driving cars, blockchain, 4D printing—and whatever comes after AI itself.