By Professor Anand Nandkumar, Associate Dean, Indian School of Business (ISB)
Artificial Intelligence (AI) is starting to transform organizations by enhancing the quality of products and services and improving worker productivity. Perhaps more important is its ability to augment the intelligence of organizations by enabling superior managerial and employee decision making.
Clearly, simply automating repetitive employee tasks may not be the most potent use of AI. This may merely be a start point for organizations to experiment with AI. In fact, a preponderance of the current use cases broadly lies in areas of automation and enhancing worker productivity.
Take the case of a leading private sector general insurance company in India which processes about 1200 auto accident insurance claims every day across the country. AI-enabled car inspection feature in its mobile app simplifies claim processing for its customers and makes them more efficient and less error prone. For their inspectors that would otherwise manually implement these tasks, this AI-enabled application would enable them to focus their attention on more complex and arguably more important tasks in which human intervention is at least for the moment inevitable.
Whereas such use cases may just mark a starting point of the AI adoption curve, the basis of AI-enabled long-term competitive advantage for firms may likely be based on the ability to use it for augmenting human intelligence. This in the long term may just be the most impactful use of AI. This will critically depend upon on how creatively organizations use AI to complement their workers and managers. Indeed, managers appear to be largely aligned to this pattern of evolution of the use of AI.
The basis of AI-enabled long-term competitive advantage
for firms may likely be based on the ability to use
it for augmenting human intelligence.
Two trends from a recent IDC report are cases in point: one, whereas an overwhelming 77 percent of organizations have imbibed AI to be a part of their strategy only for a meagre 5 percent, AI is a core ingredient of their business strategies. Second, most managers agree that data analytics, entrepreneurship/risk-taking and creativity will be more important in an AI-driven workplace which would likely be the keys to using AI to augment human intelligence.
These trends along with history provide clues on what an AI-enabled future will likely look like. History is replete with several General-Purpose Technologies (GPT) such as the dynamo or the computer that eventually transformed our society but took several years before translating to tangible gains for an organization or an economy. Non-linear, transformational technological changes are seldom adopted by organizations instantaneously.
At the turn of the 20th century, although scientists and inventors had already envisaged the profound transformational impact that electrification would bring to organizations and the society at large, its adoption was hardly instantaneous. In 1899, the proportion of manufacturing establishments that had adopted electrifications hardly represented less than 5 percent of the total number of manufacturing establishments in the US.[1] It took another two decades for electrification to diffuse to at least 50 percent of the manufacturing establishments.
Similar trends hold true of another GPT—computers—as well. Whereas the advent of the computer can be traced back the introduction of the 1043 byte memory chip in 1969, and the invention of the silicon microprocessor by Intel in 1970, its diffusion was only 10 percent of the total number of enterprises in the industrialized nations two decades decade later, in 1990.[2]
The lag between invention and adoption of GPTs are in part due to the experimentation that is required by end-users to understand the new technology and imbibe it in their respective business processes. In many cases, the technological change may themselves necessitate organizations to make many subsidiary innovations in business processes to adopt and benefit from the new technology.
Computers, for instance, required several businesses to modify their business processes for it to be beneficial to organizations. Initial adoption which largely happened without such modifications did not yield substantial benefits to the organizations that adopted it.
Drawing from these examples, the benefits of AI may well take several years to accrue to organizations. In order to realize the full potential of AI, organizations will have to stay the course for several years, make patient investments and continuously learn from limited experiments.
In order to realize the full potential of AI, organizations will have to stay the course for several years, make patient investments and continuously learn from limited experiments.
These said, how should business leaders approach the adoption of AI in their respective organizations? A whopping 65 percent of the organizations surveyed either have a feeble or no plan to adopt AI within their respective organizations. Clearly, this will likely be detrimental, because for long term competitive advantage, AI needs to be adopted at all levels of the organization – worker, middle and senior management levels. In the long run, competitive advantage will likely be determined by the ability to augment human intelligence with machine-based intelligence. For that, AI would have to supplement tactical, operational, business as well as corporate strategy decisions. However, that is unlikely to be instantaneous. How can organizations eventually get there?
Experimentation must be at the core of AI strategy for organizations. This would likely involve several changes. First and foremost, managers must immerse themselves in understanding the potential of Machine Learning and AI including developing a strategic understanding of the different technologies and algorithms that they encompass.
Second, managers may need to embrace a more data driven decision making approach. This will likely include creating a data infrastructure that will enable organizations to accumulate and govern data on several aspects of their business along with the ability to analyze and interpret them.
Third, a “growth” mindset that will reward risk-taking, experimentation and entrepreneurial thinking rather than a “fixed” mindset that will resist disruption. In order to promote a “growth” mindset, creating incentives and building human capital that will promote learning and adaptation towards data driven decision making and the idiosyncratic AI needs of organizations will likely be very important.
Finally, a more collaborative approach rather than exclusively relying on home-grown competence, which will foster a culture to learn from outside as well from other teams and functions within the same organization.
Given the pace of evolution of AI business leaders need to develop a coherent AI strategy sooner than later. They would not only have grasp the tools but also come up with a plan to adapt to a soon-to-come AI-based world. ISB-Microsoft’s Leading Business Transformation in the Age of AI, a joint initiative between the two institutions is an attempt to enable organizations in India to kick start the adoption of AI. This initiative will empower business leaders develop a plan to jump on the AI bandwagon and embrace a planned adoption of AI that would eventually be the single most important source of augmented intelligence for organizations. Our master class series will provide business leaders with a set of strategic tools to define an AI strategy for their respective organizations.
The world has gone through several technological disruptions. In all likelihood, AI will be an important disruptive force to organizations that will reshape the competitive dynamics of industries. Business leaders need to act fast and adapt to the changing world of augmented intelligence and that will be the key to their competitive advantage in the future.
Photos: Indian School of Business (ISB)
[1] David, P. A. (1989). The dynamo and the computer: an historical perspective on the modern productivity paradox. The American Economic Review, 80.
[2] Lewis Peter H. (1989). The Executive Computer: Can there be too much Power? New York Times, https://www.nytimes.com/1989/12/31/business/the-executive-computer-can-there-be-too-much-power.html
