An AI Bubble of a Different Kind

See the latest issue of Briefings at newsstands or read in our new format here.

By: Peter Lauria

It is an inevitable part of great business cycles, dating back centuries. A transformative new product or service emerges, and companies race to take advantage of it. This effort comes at great expense, but the results, for a while, are fabulous. And then they are not.

This, experts say, is the great fear lurking underneath the past year’s excitement around AI. North American companies, many of which regard the technology as a once-in-a-lifetime innovation, are projected to spend $6 billion on generative AI in 2024, according to information-tech consulting outfit Infosys. And that’s only the beginning. By 2025, Goldman Sachs forecasts that global investment in AI will hit $200 billion—a leap of faith, they say, that “will probably happen before adoption and efficiency gains start driving major gains in productivity.”

Such eye-popping amounts of money are why Jim Hoover, director of the Business Analytics and Artificial Intelligence Center at the University of Florida’s Warrington College of Business, believes leaders aren’t scrutinizing AI with the rigor and discipline that they typically bring to their investments. “There’s a huge amount of ‘fear of missing out’ going into investment decisions,” says Hoover. “Leaders are pouring money into AI and telling people to figure it out.”

Indeed, many experts point to signs that firms may be heading into, or are already deep inside, an AI bubble. The data tells the story in a way analytics lovers will appreciate: Only one in five executives in a recent study thinks their firm is prepared for AI. Yet, 31 percent of those same executives expect substantial transformative benefits from AI in less than a year, while 48 percent expect it within three years.

The problem—right now and for the foreseeable future—is that any gains from AI are likely to be incremental rather than transformative, says Jamen Graves, global leader of CEO and enterprise leadership development at Korn Ferry. “Breakthrough AI applications take considerable time to identify, test, refine, and test again before they are ready for prime time,” he says. Consider the groundwork that has to be laid for a firm to successfully adopt AI, he says: hiring or upskilling talent, integrating applications into core business processes, and implementing governance and quality controls, among other tasks.  

That’s not to say that AI won’t ever achieve the returns that leaders are expecting. It will, experts say—just not for a very long time, and only for the minority of companies that are able to balance investment with implementation and adoption. Hoover compares leaders intoxicated by AI’s potential to late-cycle Six Sigma acolytes, who wasted their valuable financial and human capital on marginal improvements instead of making significant ones. “The way AI is being implemented, it may end up in the same place,” says Hoover.

Before AI can lead firms to the promised land, Graves says leaders need to provide clear direction and align teams around the areas in which the technology can impact business outcomes and where it can simply provide support. “One directive is clear and specific, while the other is far too generic,” he says.