Key Takeaways

  • Protiviti Managing Director Dan Stummer believes CFOs are underestimating the potential for exponential growth from AI and should seize the opportunity of this emerging technology by allocating a significant portion of annual spending to technology exploration that allows for failure.
  • Stummer, speaking in an interview at the Money20/20 fintech conference in Las Vegas, said CFOs and their executive peers need to overcome an "expectation mismatch"—the gap between conventional forecasts of returns from emerging technology and the dramatic leaps AI can deliver.
  • "This expectation mismatch poses a significant challenge" when CFOs accustomed to linear growth need to measure AI's return on investment, Stummer said. "The key is recognizing that you're on an exponential curve, yet you're using linear methods—like planning and hierarchy—to measure something that is inherently exponential."

Deeper Dive

Although CFOs have invested $30 billion to $40 billion in generative AI, 95% of organizations have seen no return from the technology, according to a survey by MIT.

"Only 5% of integrated AI pilot projects have extracted millions of dollars in value, while the vast majority remain stuck at an unmeasurable impact on profit and loss," the MIT research noted.

In fact, Stummer said, CFOs have heard the call for rapid change for so long that they have become numb to the demand.

"A very tricky issue is that CFOs generally suffer from 'transformation fatigue,'" he said. "It's like, 'Oh, we have to do another roadmap, hire another consulting firm, and so on.'"

However, the long-term benefits of AI could be profound, the pace of innovation is so rapid, and the competitive stakes are so high, that Stummer believes CFOs need to drive experimentation and embrace the notion that success depends on a willingness to fail repeatedly.

When considering pilot projects, "don't tell me what the success criteria are, tell me what the failure criteria are, and when you hit the failure criteria, don't regret it—just kill it," he said.

Stummer noted that CFOs can more easily tolerate higher failure rates by allocating a portion of spending to an "exploration portfolio," managed separately from an "operate and exploit portfolio" focused on existing profitable businesses.

"The key is truly understanding that these two portfolios operate separately, and the business cases you seek are vastly different," he said. "What you're doing is figuring out how to turn experiments into viable growth—how to scale?"

Stummer said CFOs must either discover AI applications that generate returns through their own efforts or acquire them. Otherwise, their companies will face decline.

"Instead of falling into transformation fatigue, you have to embrace change and be excited about it," he said. "That's hard to accept."

Otherwise, "you will pay the price of declining cash flow," Stummer warned.