We are entering a period of generational change, where we have recognized that we need every brain on deck to solve some of the most complex challenges humanity has ever faced. This is an opportunity that applies not only to individuals, but teams and organizations, as well as the ecosystems that exist within and across the sectors we serve. There is even the potential for the things and places we design and build to be part of these learning ecosystems. As those places and things sense the nature of their surroundings, they’ll need to learn—along with their constituents—how to build, navigate, and nurture fruitful journeys towards a better world.
There is a war for talent on the one hand and on the other analysts and the news media claim that robots are taking everyone’s jobs. Deloitte put out a study claiming 47% of today’s jobs might be gone in the next 10 years. From the macro-economic side of things, there is plenty to do, but our collective vision can be clouded in many ways. First of all the impact of globalization and trade means that jobs might not be in the same places with the same values as they were before. Secondly, population growth, migration, and infrastructure collapse (or creation) can cloud our view of large swaths of society. Second, in terms of micro-economic impact—individuals that define themselves by certain tasks within their jobs can find themselves suddenly feeling as if their identity is commoditized by automation. At this individual level the challenges are undeniable. If people and organizations increasingly feel sidelined or experience a sense of loss of control, they may become paralyzed or fall prey to populism. Perhaps the worst thing is that most analysts—and many leaders—claim we can’t possibly predict what jobs will be created, or what roles our children or grandchildren will play in the future of work.
Banking on life-long learning
If we believe something is valuable, we bank on it. We invest in it, look for the highest rates of return, create markets around both its current and future value to predict where we should invest next, and put the returns away to prepare for a rainy day. We historically have invested in college degrees for one-time bursts to learn and apply that learning over our careers. But aside from getting us going in a career and saddling us with the largest personal debts in recorded history, the 20th-century approach to learning—where we fix our minds on what we can be and do over our lives far too soon—looks increasingly unlikely to serve us well in the 2020s. We know the financial markets are having their doubts about the 20th-century model of learning. Are we valuing the opportunity to become lifelong learners and learn how to learn? Do we value those aspects that make us human, as robots play more of the roles we thought defined us?
What we learn, how we put it into practice, how we gain mastery, and how we help others master that capability and join the work ahead? That is the currency of the 2020s.
BCG’s report, “Winning in the ’20s” frames how leaders will drive value. BCG notes, “leaders must compete on the rate of learning,” by closing autonomous learning loops between human and machine learners. An HBR analysis, “Your workforce is more adaptable than you think,” highlights the difference between leader’s views versus what their workers believe. The recent report from the Autodesk Foundation notes areas of opportunity for the areas I focus quite a bit of my time on, namely design, manufacturing, architecture and the construction industries. It notes that “investing in human capital”while “leveraging human capital,” and “fostering markets where value is matched and exchanged,” will drive new value. The Ford Foundation’s policy experiment, “A 401k for Learning,” encouraged contributions from employers and employees to invest in new skills. After a five-year trial across varied industries and communities, the experiment found that in high percentages—both for employees and employers—there was an increase in economic value and upward mobility. People embraced the underlying policy initiative.
Satya Nadella notes that “We have to shift from a know-it-all mindset to a learn-it all one.” The impact at Microsoft is just beginning but has been profound. Amazon put it in their 14 leadership principles, “Leaders are never done learning and always seek to improve themselves. They are curious about new possibilities and act to explore them.”
Let’s focus on that point about fostering markets. What if we participated in a “Futures market for learning?”
It would be easy to listen to all the sage pundits, consultants, and never-ending prophets claiming that we can’t possibly predict what we’ll even call the jobs we’ll perform in five or ten years. How could we possibly know what skills or mindsets that define us today will become irrelevant or critical in the years ahead? How could we send signals between educators, industry, and policymakers, if the future is so hard to imagine? How might we help our stakeholders know which passions to cultivate next or when the passion that defines them is on a path to be automated sooner than they think?
Let’s imagine what a “learning futures prediction market” might look like. A prediction market uses the power of a diverse community of participants to invest in predictions—much like economic markets are seen as using a large group of investors to set the value of a stock—through the act of putting “skin in the game.”
Prediction markets need at least two ingredients.
1. A method of tapping into the crowd. The idea of crowdsourcing isn’t a new one, we can go back as far as the 1700s and the quest to discover a practical method to determine how far east or west a ship was sailing across the sea (its longitude). This was a scientific challenge that for centuries had been unsolved. Many of the greatest scientists of the time had been unable to come up with a solution. A prize was put forth for all contenders. While famous scientists like Leonhard Euler—known for developing modern mathematical notation, functions, graph theory, and other concepts we use every day when we try to solve challenges ranging from population growth to routing pathways for shipments—failed to solve the problem of longitude. An little known carpenter, choirmaster, and clockmaker named John Harrison succeeded in solved the problem.
2. An algorithmic framework that reduces the chance that various forms of misunderstanding—from the cognitive bias called “the illusion of mastery” where experts in one field believe they are expert in other fields, to the impact of propaganda—marketing, echo chambers, “truthiness,” etc.—that might distort the weak signals of the smaller population within the crowd that have unique insights from hard won experiences. Now let’s flash forward to the 1960s and the United States military discovers one of their nuclear submarines has disappeared. The leader in charge of finding the sub collects all the data they have at that moment and asks an oceanographer, a sea captain, a meteorologist, and a submariner to place a pin on a map for where they believe the submarine could be. The leader uses a basic algorithmic framework called a “Bayesian Average” to calculate the results. The leader found the lost sub 32 meters from the predicted results. The power of the crowd comes when all of us can combine our unique knowledge in fruitful ways. The Hollywood Stock Exchange (HSX)is more current example. In one year, the participants successful predicted 32 out of 39 Oscar nominees and 80% of the award winners within the top categories. For more about how to tap into a crowd and the difference between the madness and the wisdom of the crowds, read this.
Let’s envision what our learning futures prediction market might look like. We’ll pick a framework to amplify weak but important signals. Lets look at one that MIT developed called the “Surprisingly Popular Algorithm.” The Surprisingly Popular Algorithm asks two questions, “What do you believe the correct answer is?” and “What do you think the popular opinion of the right answer will be?” When you look at the variation between the two answers in aggregate you will often find the right answer. This approach gives more weight to smaller groups within the crowd that have specialized knowledge and acts, in some ways, as a means of diagnosing when a majority has popular opinions that are wrong—think of it as an antidote to “truthiness” and the amplification of propaganda, like “robots are taking all our jobs!”).
Imagine there might be particular dimensions we and the industries we serve might define as critical. Maybe the top emerging toolsets, skillsets, and mindsets business leaders believe will have the highest business value in the next five years. Another dimension might be which of those are the most and least likely to be automated.
We might also want to consider the individual’s current abilities and how near or far they are (proximal) to what those other dimensions have identified. Or even how close those emerging high value needs are to passions individuals would love to cultivate next.
If we play out this market a bit we might discover some interesting signals—if we can begin to understand what our customers see as high value targets we’ll have a better understanding of what should be automated next—as well as be able to help the industries that we serve have an edge in preparing their own workers for the next turn of the automation screw.
Individuals would be able to see that those that are least likely to be automated but highest value and close to what they already do today are where they should be investing their own efforts in preparing for the future and assisting themselves in being capable of moving in an upwardly mobile direction as the things that they do now get automated away.
The stakes are high
To put this all in context I’ll focus on those jobs I know best—but I think this is far more broadly applicable—in the design, engineering, manufacturing and construction sectors. These are high value roles where we create the world we live within and consume much of our natural resources. If we don’t figure out better ways of designing and making things, and provide a talented, curious, and creative workforce to drive better—which often translates to different—outcomes, we’ll continue to design and build things and places the way we always have. One look at our record in the 20th century in terms of waste, pollution-driven illnesses, and rising inequality should give us pause. Worse, according to Geoffrey West in his book, Scale, “…when averaged over the next 35 years about a million and a half people will be urbanized each week. To get an idea of what this implies, consider the following: Today is August 22; by October 22 there will be the equivalent of another New York metropolitan area on the planet.” He goes on to note, “This will be by far the largest migration to have ever taken place on this planet…” Our growing populace will need places to live, to be healthy, and thrive. They’ll also likely expect to participate in the bounty of our connected world with products that give them access to the Internet, allow them to move around, provide them with safe food to eat, foster health and wellness and provide places that offer meaningful ways to meet and add value to their community.
In the US, manufacturing jobs alone—about 2.4 million jobs in the next decade—are predicted to go unfulfilled due to skills shortages. Over 200 million construction jobs are expected to be created worldwide by 2030. With the retiring baby boom generation 41% of the current construction workforce will retire by the year 2031 and the manufacturing sector is facing the same scale of losses.
Where do we start?
We are currently experimenting with pilots around some form of “cross-sector” prediction market for learning. For example in the fields of architecture, engineering, construction, design and manufacturing we began by bringing together around 200 expert users/customers from the industries that Autodesk serves. We facilitated a discussion using an online polling system and some whiteboards to see what a rough approximation of a prediction market might look like.
We asked them first to suggest major emerging mindsets, toolsets, and skillsets that they think will be highest value for them to invest in within the next few years.
We then asked them to rank which ones were most valuable to their business in the next few years…
We then recruited those participants that were deeply immersed in building automation and machine learning systems to rank which ones would be least likely and most likely to be automated in the same time frame.
We polled all of the participants about which high value areas were close to what they know how to do today—a useful signal around proximal learning opportunities they could bridge easily and those that they’ll have to invest deeper and more sustained efforts into mastering. When we combined all of these dimensions, high value, least likely to be automated, and proximity to what they do today, a picture began to form…
While this was a cartoon prototype what became clear was that broad topics like “adaptability,” “critical thinking,” and “collaboration” came to the foreground. While farther from what they do today, investing in design thinking, compassion and creativity where also key areas for personal development. When we probed deeper a richer picture emerged around the underlying details about these topics. For instance “adaptability” was associated with having an agile mind that could improvise and evolve their mental model at the rate that new capabilities and challenges emerged. When they talked about collaboration they alluded to deep interdisciplinary co-creation and problem framing as their projects became far more complex and over-constrained compared to classic challenges. The idea of bringing a more compassionate mindset to engineering, manufacturing, and design challenges led to a deeper conversation about understanding the unintended consequences of the long term impact of their solutions and walking a mile in their stakeholder’s shoes so they could gain empathy and insight into those that saw the world differently than they did.
Finally, we asked about passions they themselves where curious about cultivating—whether as part of their career or as a means of personal exploration. A richer picture emerged around work-life balance, mentoring, teaching and system-thinking.
What would happen if—instead of a one-time session—this form of signaling became a live and active market of both ideas and places to personally and organizationally invest? We aren’t convinced this is an effort that should be done by any one company alone, but instead—just as other markets emerged across industries and regions—this must be an emergent series of baby steps focused on near-term value while the dimensions and crowd-sourcing frameworks evolve.
Consider this a provocation and invitation to join the effort to find ways to invest in and bank on better ways to learn and build our future.
Next? Stay tuned for how our places and things might learn how to be better at meeting their missions in life in the next installment of Infinite U (what if places and things could be life-long learners?) – Part 4.