“How do we divide the work between mind and machines?" Andrew McAfee, author, blogger, and co-founder of the MIT Initiative on The Digital Economy, posed this question during Genpact's “Enterprise of the Future" event last month. I often hear similar questions in my conversations with senior business leaders and technologists who are exploring how AI tools can transform their operations, and I find that there are still some uncertainties and misconceptions surrounding AI. Fortunately, during the Genpact "Enterprise of the Future"event, McAfee broke down four key points that every leader should keep in mind.
1. AI is the new crystal ball
Oftentimes, business is a prediction game. Everyone is making bets on the future and hoping they're placing good ones. So, wouldn't it be great if an organization had a crystal ball that could predict which products will succeed, which customers are the most or least satisfied, or how its market is going to evolve in the next quarter or year? While there is no perfect crystal ball because the world is chaotic and complex, AI is a pretty good substitute in many ways.
Equipped with a large amount of data on customers and operations, companies can apply AI to make more predictive business decisions and anticipate customer wants and needs. Combined with machine learning, these systems can get smarter over time. For instance, Amazon uses data on what consumers have previously bought, correlates the information with similar customers' purchase, and uses that as the basis for suggesting additional products. As customers make more purchases, the suggestions become more accurate. Thus, Amazon can increase its order values and quantities through cross-selling and upselling, driving top-line growth.
2. It's a big mistake to underestimate AI
One of the biggest mistakes that businesses are making is underestimating and under investing in AI. This is due to the fact that AI is a very different tool from what most companies are comfortable with right now. For a long time, organizations have relied on traditional rule-based programming. Now, AI enables outcome-based programming. The other problem is that the skill base at most companies is not aligned yet with the future of work. In the near future, there will be the need to re-skill some roles, enhance others, and create new jobs. In fact, we are reaching new milestones in AI far sooner than expected. We haven't totally grasped the speed at which these changes are taking place. So, a lot of businesses are going to be guilty of committing a double fault – underestimating and under investing in AI.
3. Man and machine are good at different things
Machines can already beat us at strategy games, understand our speech, and diagnose diseases. But that doesn't necessarily make them smarter than us. We surpass machines in areas like using common sense, asking questions, and demonstrating social skills.
Our own experience in AI shows that machines are not going to make us redundant, but that they will augment the human workforce. In truth, AI as a technology solves only half of the equation. AI technology must be contextualized in the domain of an industry, which is information that only humans possess. Therefore, people who can apply their knowledge and experience in meaningful ways with AI will be increasingly important.
When we envision AI in the context of a business problem, the key is to effectively encourage and adopt human-machine collaboration. Further, as we transition to an AI world, governance by man to reduce risks around errant robots and misplaced AI spinning out of control through a control hub will be critical as well.
4. AI isn't a tech but a leadership challenge
Companies that are starting from square one will have an easier time implementing and moving forward with new technologies. Meanwhile, existing companies will face larger challenges as they attempt to leverage their current skill base with this radical new approach, where machine learning and AI technologies are at the core of the business. Existing organizations will need strong leadership to drive something so profoundly different.
If you are ready for the future with AI, here are a few final points to consider. During the event, McAfee listed his recipe for success: Take a small initial step with expert guidance, iterate and experiment before you scale, and create a big strategic plan. At Genpact, we encourage business leaders to set up the right design and governance models to manage digital transformation. Moreover, as you begin to look at the different technologies available on the market, you may find yourself inundated by the noise surrounding AI, making it hard to pinpoint which systems to put in place. Recognizing this problem, we have set out to curate a comprehensive set of modular digital technologies that can serve as a framework and help companies drive transformation in an orchestrated, controlled, and integrated way.