Amid all the uncertainty of the COVID-19 health and economic crisis, it might seem like a good time to shelve plans to incorporate artificial intelligence into your business model. But that would likely be a mistake. In many ways, now’s the best time to begin understanding the potential of AI in your company and to start investing – even if it’s in relatively modest amounts.
Professor Sampsa Samila coordinates research and teaching at IESE on AI and the future of management. In a recent online session with professor Mike Rosenberg, he explored some of the reasons why executives should be doubling down on AI. Here are some of them:
Because COVID-19 has super-charged digitalization. Many companies were lagging in AI adoption because they just weren’t digital enough. With the coronavirus and the sudden switch to remote working and online formats, that is no longer the case. “We’re now adopting new ways of working. We’re adopting new technologies and we’re learning to use digital tools more than previously,” Samila says. “AI can basically build on top of that.”
Because it starts with managers. You don’t have to start by hiring an army of data scientists, which could be complex and costly in this difficult climate. Instead, the most important step is to learn about how AI could impact your business. Research shows that many examples of successful AI adoption have been in companies that simply start with managers committed to learning about it. They often begin by working with off-the-shelf tools, or hire a consultant, or begin a pilot project. “The key challenges was to get managers learning what this is and what it can do rather than getting many data scientists,” Samila says.
Because you might be re-thinking your business anyway. Since COVID-19 is prompting many businesses to re-think their strategies and focus, it’s a good time to add AI into that equation. And that starts with the customer rather than with coding or algorithms. “The key thing to understand and learn is to start from the basics of what exactly are the customer needs you’re serving, to understand your customers, to relate to your customers, to understand what is the work the organization is trying to do,” Samila says.
Because there’s cumulative advantage. That means businesses who start collecting data early will build on that strength. More data leads to better predictions leads to more customers, which then leads to more data, better predictions and more customers.
Because there are economies of scale. An algorithm once trained can be used in different applications, at relatively little additional cost. Making that first investment doesn’t mean that you will have to continue making the same investment every step of the way. Netflix’s predictive technology for customer’s preferences, for example, can be put to use in many different ways.
Because its uses cut across sectors. Since AI is a general purpose technology, it has applications and uses in just about every sector. Now, it is being used more widely in areas such as finance and banking, ecommerce and marketing. But it’s only a matter of time before it is used in most industries.
But if AI is dependent on past data, and behaviors are changing so fundamentally, isn’t all that accumulated data now of limited use? The evidence suggests otherwise, according to Samila. He looked at hedge funds, some of which were managed by AI and some by human managers. While those managed by AI took a larger dip early on, they also recovered more quickly. “It is suggestive of the fact that both humans and algorithms need to learn the new normal, and it’s not obvious that humans have a huge advantage.”
This post is also available in: Spanish