A landmark battle raged in New York City on May 11, 1997. This contest was dubbed the brain’s last stand and considered the proverbialhuman-versus-machine duel. In the end, IBM’s Deep Blue routed the World Chess Champion at the time, Garry Kasparov.
It changed the way humans looked at Artificial Intelligence (AI). Over the next two decades, AI has grown a lot smarter.
Today, what is the most intelligent species on our planet? You might still pick humans, while others could choose machines, hands down. It’s neither of these, says Thomas Malone, Professor of Management at MIT Sloan School of Management.
Per Malone, the smartest entity ever known is groups of people. Almost everything humans have ever done has been accomplished not by lone individuals but by groups of people working together, often across time and space. He calls this a supermind.
Superminds are powered by collective intelligence—not just individual brilliance. What is collective intelligence, and where is it used? How can organizations leverage human-machine collaboration to scale decisions? How should executives prepare for a future where super-intelligent machines will be commonplace?
These are some of the questions I asked Professor Malone, who is the founding director of the MIT Center for Collective Intelligence. He answers them with examples based on his research and using insights from his book, ”Superminds: The Surprising Power of People and Computers Thinking Together.”
Malone defines a supermind as a group of individuals acting collectively in ways that seem intelligent. Superminds have operated since the time of hunter-gatherer tribes. So, what is collective intelligence? Think of it as the attribute that superminds possess. It is the collective intelligence of people that helped forge civilizations, form governments, and build successful organizations.
While humans have been the primary actors of superminds, machines have begun playing a more prominent role over time.
Even before machines became smart, they created an enormous impact by just connecting humans to other humans. Malone calls this “hyperconnectivity.” What’s a simple yet profound example of hyperconnectivity today? The internet. Think of the vast potential that gets unlocked when simple machines link millions of humans worldwide.
In recent decades, machines have started doing tasks that many consider intelligent. Thanks to advances such as AI and machine learning (ML), computers now bring their exceptional abilities to the supermind.
Today, several organizations use superminds that employ human-machine collaboration to power their core business offerings. Malone shares Google Search as a case in point. Everything that happens from the moment a user types in a search query until they get their results is performed entirely by computers. No human appears in the loop for this operational process.
So, what’s the role of people? They tell the machines what to do. Google programmers are like employee training specialists—but here, the “employees” are machines. People instruct the machines through direct programming logic and by training machine-learning algorithms.
With ML, machines use historical queries from users to surface the most relevant search results. Of course, the web pages that a Google search returns were mostly created by people, highlighting another critical role that humans play in this supermind.
Once you consider everyday products and services, there are many similar examples of superminds that blend human-machine abilities seamlessly. Digital-native companies such as Netflix, Pandora, and Uber have deep human-machine collaborations that power not just their customer-facing offerings but also their internal operations.
While there are many instances of organizations using superminds to drive their operational workflows today, the area of strategic decision-making is less explored.
Let’s take the case of corporate strategic planning.
A Harvard Business School report outlines that 85% of executive leadership teams spend under one hour per month discussing strategy, and 50% spend no time at all. Strategy planning is primarily restricted to an annual exercise with participation from a few senior leaders.
It’s no wonder that 95% of a company’s employees don’t understand its strategy. Ultimately, 90% of businesses fail to meet their strategic targets.
Superminds can transform this dated process of corporate strategic planning, says Malone. Today, machines’ involvement in this process is restricted to automating a few computations or tracking metrics. He envisions a new approach that leverages greater human-machine collaboration.
Thanks to hyperconnectivity, almost everyone within a company can have a say in the strategic planning process. Each employee can be empowered to share inputs on corporate plans, recommend strategic offerings, or suggest features in the product roadmap.
While this sounds great in principle, how do you get people to participate? Gamification can turn the planning process into a live contest. In fact, several enterprises have used gamification to improve employee engagement radically across their business.
For example, Spotify gamified the dreaded annual appraisal process through live scorecards and badges shared on an internal social network. Consequently, over 90% of employees participated voluntarily in reporting live performance scores throughout the year.
When a stream of ideas flows in, how do you evaluate them continuously and objectively? Borrow some best practices from prediction markets.
Early in their journey, Google built an internal system that works like a stock market. They let employees bet on probable outcomes using virtual money. Employees answered questions such as, “Will a project be finished on time?” or “How many users will Gmail have?” This trading system helped the Google hierarchy discover employees’ uncensored opinions and gain insights that could have been overlooked.
The concept of prediction markets taps into the wisdom of crowds to discover the most effective outcomes. Through a similar setup, you can set up internal prediction markets with a defined audience to evaluate your strategic planning ideas.
At this point, you can bring in machines to augment the decisions. Machine learning models can observe this process of manual decision-making and learn from the outcomes. In case you’re wondering whether machines can find patterns in such decisions, it has been done before.
Bridgewater Associates, the famed investment management firm founded by Ray Dalio, showed that algorithmic decision-making could be applied to run a large business. The firm’s management coach system collects copious amounts of data to study employee behavior. The system bases judgments on continuous decisions made by employees across the firm. Dalio said that this collective decision-making approach helped turn the firm into one of the world’s most successful hedge funds.
By studying examples of how humans evaluate strategic planning ideas, machines can start recognizing patterns in human judgments. Given enough decision volume and variety, ML models can be trained to make recommendations to humans. When machines reach an acceptable threshold of decision quality, you can delegate certain types of decisions entirely to them.
A gamified, machine-enabled strategic planning process can be replicated across all organizational levels, from individual product teams to the corporate level. Instead of creating strategic plans annually or quarterly, organizations can now make thousands of plans every single day.
“Most people think of strategy as an event, but that’s not the way the world works,” said Clayton Christensen, the Harvard Business School Professor. “More often than not, the strategy that leads to success emerges through a process that’s at work 24/7 in almost every industry.”
Whether it’s because of a competitor launching a new offensive or the entire industry going into a decline during the pandemic, the strategy needs continual reassessment. This blueprint for a cyber-human strategy machine could enable organizations to respond continuously and rapidly to ongoing changes in their environment.
However, the approach does pose some inherent challenges. Culturally, it calls for organizations to switch from a hierarchical, closed planning process to one that’s more flat and inclusive. From a technical perspective, machine learning stumbles when predicting scenarios with few historical examples or those involving nuanced decision-making, as they involve fewer digital trails.
However, with the proper executive support and technology investment, these roadblocks can be overcome. The potential upside of acquiring such an enhanced strategic capability will motivate organizations to make the switch. “A company that manages to build something even close to this has the potential to run circles around its competitors,” says Malone.
Executives who aspire to apply superminds to rethink and rewire their organizational processes can tap into the supermind design methodology. This systematic process was developed at the MIT Center for Collective Intelligence. It helps organizations redesign their operations by leveraging novel approaches in human-machine collaboration.
The quest for Artificial General Intelligence (AGI) is making machines smarter by the day. Some wonder how this will shift the power equation within a supermind. Will self-aware machines displace humans from the superminds of the future?
Malone feels that most people tend to overestimate AGI’s arrival. Pointing to history, he highlights how people have been asking this very question since the early days of AI in the 1950s. “For that entire time, the average prediction has been that we’ll have human-level AI about 20 years in the future. That’s still the typical answer today,” he says.
Even if computers could do almost everything humans do today, Malone explains that there will always be more things for people to do. These could, for instance, be tasks that involve providing the much-needed human touch, conversation, or connection.
For example, while AI is getting better than human doctors at making disease diagnoses, can it advise patients on their prognosis with the same care and empathy as humans? Clearly, both humans and machines serve vital yet distinct functions.
The narrative is shifting from competing against machines to collaborating with them. In his book, Kasparov clarified that his loss to Deep Blue was really a victory for humans. “The machines work for us, after all,” he noted.
End of content
No more pages to load