A radically human approach to human-machine interaction is turning assumptions about the basic building blocks of innovation upside down. Specifically, the new utility of small data — including the ability to create “synthetic” data to simulate a set of circumstances — is bringing the power of AI at scale within reach of entities that previously couldn’t afford it. Instead of machines “learning” by processing mountains of data, humans can now teach machines based on human experience, perception, and intuition. That means more people throughout organizations can use AI in new ways, based on their individual expertise.
The resulting bigger picture? Cumbersome legacy IT architecture is giving way to living systems that can weave together technologies, data, and talent in a hyper-digital world of mobile computing, AI, the Internet of Things (IoT), and billions of devices. These developments have opened up vast possibilities for strategy innovation — yet only a small number of companies have made a radical leap into new strategies that these radically human technologies have opened up.
Three of these new high-potential strategies include: Forever Beta, Minimum Viable Idea (MVI), and Co-lab. We’ll explain each in detail, with examples of companies that are currently employing them. Though their specific strategies are distinct, the companies that are using these new strategies well share three important characteristics. First, their technology, business strategy, and execution are so closely intertwined as to be nearly indistinguishable. Second, humans — not machines — are in the driver’s seat. Third, these companies understand that all companies, no matter their industry, are now technology companies.
Forever Beta
Forever Beta strategies offer software-enabled products and services that continually evolve and improve after they’ve been purchased, so that customers see them grow in value and utility over time, rather than fade. For instance Tesla, unlike other automakers, doesn’t offer yearly updates of its models. That’s because Tesla puts out a model and then continually improves it after the fact. Tesla owners see their existing cars continually transformed through updates that advance the vehicle’s autonomous driving capabilities, improve performance, and enhance safety features.
Through cloud/edge connectivity with the cars, Tesla monitors performance and provides remote diagnosis and repair. For example, a motor problem involving occasional overheating was diagnosed and repaired by a software patch. Tesla drivers are in a constant feedback loop with the company, imparting their human expertise to Tesla’s neural network and improving it merely by driving.
The result is an ownership experience that sees the car continually grow in value and utility. This experience is designed to be an intrinsic part of the usefulness and differentiation of the product for customers, who are, in effect, highly privileged beta users of each new improvement.
Signify, formerly Philips Lighting, keeps the latest and greatest technology in its customers’ hands through Signify circular lighting, an enterprise solution where Signify offers lighting based on uptime and energy needs provided by its customers. The company shouldered responsibility for the lighting throughout the power plants that provide all of the electricity for Dubai, resulting in a savings of 68% in lighting-related energy consumption for the city.
By intertwining technology and business strategy so closely, these companies are forming stronger relationships with customers, built on an understanding that the service experiences they purchase today will appreciate in value tomorrow.
Minimum Viable Idea
Minimum Viable Idea strategies use intelligent technologies to precisely target weak links in a traditional industry and provide a superior customer experience that can be quickly scaled to make rapid inroads in the market. Lemonade, a New York-based insurance company co-founded by Daniel Schreiber and Shai Wininger, offers coverage to renters, condo owners, homeowners, and pet owners. Their AI-powered app streamlines getting quotes and settling claims, sometimes in seconds. But, says Wininger, “Lemonade is a tech company doing insurance, not an insurer doing an app.”
Lemonade has combined AI chatbots, machine learning, and the cloud to focus with laser-like precision on the features of traditional insurance that make the industry widely disliked among consumers. And, for the element of expertise, they found a stunningly creative way to put a human in the loop.
Consider the company’s claims process. Users tap the “Claim” button in the app and simply tell the chatbot, named Maya, what happened. There are no forms to fill out, no waiting in a phone queue, no being handed off from one department to another. The company’s AI runs its anti-fraud algorithms and if the claim is instantly approved — as some 30% are — the AI pays it immediately. If not, the claim is escalated to a human who contacts the insured as soon as possible.
The process is so frictionless thanks in part to a financial model that resolves what Lemonade’s founders saw as an inherent conflict of interest for insurance companies: every dollar of a customer’s claim that a company denies is another dollar of profit for the company. This incentivizes the insurer to do everything possible to deny or reduce the amount of a claim and motivates the customer to inflate claims.
Lemonade simply takes a flat percentage of each premium. It returns the unclaimed remainder in an annual “Giveback Day,” when the money is donated to worthy causes policyholders care about. Policyholders who choose the same cause are pooled in a virtual peer group. The premium money collected from each peer group is used to pay for the group’s claims. Whatever money is left goes to that group’s cause. On Giveback Day in 2020, the company donated more than $1.1 million to 34 nonprofit organizations, including UNICEF, Direct Covid Relief Response, the Malala Fund, Born This Way, and others.
The humans in the loop in the claims process are the customers themselves. When they enter a claim, they know that Lemonade has no incentive to unreasonably deny or reduce it. Just as important, they know that every dollar by which they inflate a claim means less money for a cause they care deeply about. This dynamic not only puts a human in the loop, but puts what is uniquely and radically human at the center — moral conscience.
Co-lab
Co-lab strategies produce superior results in the sciences or other knowledge-intensive environments through human-guided, machine-driven discovery. Set free by automation and machine learning to leverage human knowledge at the highest levels, specialists and knowledge workers, driving these powerful technology platforms, exponentially increase productivity, multiply value, and erect high barriers to entry.
Exscientia, a UK-based start-up, has developed an AI-driven drug discovery platform it calls Centaur Chemist. To identify a disease to target, Exscientia initially applies deep learning algorithms to narrow down the almost limitless number of potential disease candidates. Then the company’s experts devise a strategy that is carried out by the Centaur Chemist’s “active learning” system which “learns” its way into a drug discovery dataset with limited data points, relying on highly data-efficient algorithms. Typically, in drug discovery, little is known about new targets for therapies and there is scant data that could be used in big-data machine learning approaches. In 2020, Exscientia became the first pharma company to create an AI-designed molecule to enter human clinical trials, followed by another in 2021. Equally impressive instances of the power of Co-Lab strategy can be found in the development of Covid vaccines by Moderna and Pfizer/BioNTech in record time.
Small Steps or a Giant Leap?
Forever Beta, Minimum Viable Idea, and Co-lab hardly exhaust the bold business strategies that are emerging from the radically human turn in intelligent technologies. But technology-driven business strategies aren’t self-generating. They require farseeing leaders. Those who are able to see opportunities at the new radically human nexus of people and technology will pre-empt disruption and seize the future. Those who continue down the path of incremental automation will suffer. So will workers. In groundbreaking research on worker displacement by automation, the economists Daron Acemoglu of MIT and Pascual Restrepo of Boston University concluded that “it is not the ‘brilliant’ automation technologies that threaten employment and wages, but ‘so-so technologies’ that generate small productivity improvements.” The same can be said for so-so strategies.
