- A shift in strategy. As organizations gain the ability to understand AI and modify their pay strategies ever more quickly, strategy will become more tactical.
- The human element is still important. Compensation decisions must be nimble, but final decisions should still be made by people.
- Possible challenges. Regulators will ask businesses to provide real-time data, rather than submitting periodic reports about company pay practices. There may also be tension between organizations and regulators.
If you are a compensation professional looking to understand how artificial intelligence (AI) might affect your role, you have a lot to think about from WorldatWork’s Total Rewards’24 conference session, “Is AI the Death of the Compensation Function?”
Webinar: What’s New and Next in Comp Data?
The Monday afternoon session featured a debate between two Salary.com executives: Rebecca Gorman, principal consultant, and John Sumser, vice president of marketing. Gorman brought to the conversation more than 20 years of management and consulting experience in compensation and benefits. Sumser drew on 30 years of experience analyzing the HR tech industry and more than a decade examining the industry’s adoption of AI.
During the debate, Gorman and Sumser agreed on some points, but more frequently offered differing views that provided interesting perspectives on how the compensation function may evolve as AI becomes more sophisticated.
How Can AI Influence Strategic Compensation Decisions?
One of the key questions was the ability of AI to influence — or even take over — strategic compensation decisions.
“I don’t think AI could ever replace the strategic element in the compensation function,” Gorman said.
“What will change first is the pace of strategy,” Sumser noted, adding that most of the compensation function is built on annual cycles. As organizations gain the ability to understand and modify their pay strategies ever more quickly — to the point where they can do so on the fly — strategy will become more tactical, he said.
Gorman countered that strategic compensation programs are unique to an organization, based on factors such as its markets, business cycle and approaches to growth. These decisions, she said, are not ones AI can make.
Sumser agreed. “We’re at the very early stages of this, but it’s moving pretty fast.” He added: “What you and I are really arguing about is how long does it take to get to the point where you can make that decision. And that is going to compress.”
Gorman said decisions about pay strategy, organizational alignment and individual alignment must be made by a combination of HR professionals, compensation professionals and the executive team. Involving all these parties helps ensure that such decisions align individual incentives with the business strategy.
“I think the problem with that point of view is that no one has seen what is about to happen to work,” Sumser said. “Jobs are going to die faster than current compensation cycles [can adapt to them].”
Gorman accepted the notion that compensation decisions must be nimble, but believed the final decisions should still be made by people.
Sumser agreed that people should make the final decisions. But he said that if you define what your organization is trying to achieve, AI could “fairly easily present you with 10 scenarios to pick from.”
“Wouldn’t those be the same 10 scenarios it offers to a competitor?” Gorman asked Sumser.
Sumser didn’t think so. “Are you sharing your data with a competitor?” he said, adding that “when you deploy a strategic tool like this in a company environment, it has to be about that company. It can’t be generic compensation strategy.”
AI’s Effect on Tactical Compensation Efforts
Gorman and Sumser more often agreed on AI’s ability to contribute to the more tactical aspects of compensation work.
Sumser again pointed to the potential speed gains AI can provide. “I don’t think it’s that long before you have always-on pay equity analysis,” he said. He noted that once the system flags a potential problem, the organization will need to move quickly to reduce its risk.
He even speculated that regulators will ask businesses to provide real-time data, rather than submitting periodic reports about company pay practices. Gorman predicted a tension between organizations (which will want to provide as little information as possible) and regulators (which will want to gather as much information as possible).
Both agreed it would take time to work through such challenges. Sumser predicted it might take a decade, “but that will go pretty fast.”
When to Consider Adopting AI
Both Sumser and Gorman agreed that comp professionals will need to carefully weigh the claims they hear in the near future regarding what AI can accomplish. Sumser said that comp professionals will need to closely watch both the vendor universe and the outside AI universe to see how regulations may be applied to the AI space.
“I would be a little nervous about being an early adopter in compensation because the propensity for making mistakes is big,” Sumser said. Yet, waiting too long also can be a mistake, he said, because “in technology, competitive advantage always accrues to the early adopter.”
AI is a moving train, said Sumser, and compensation professionals must say to themselves “I at least have to understand it and uplevel my skills” so they can be ready to climb on board when the time is right.
“It’s deciding when to buy the ticket to get on the train,” Gorman said.
Editor’s Note: Additional Content
For more information and resources related to this article, see the pages below, which offer quick access to all WorldatWork content on these topics:
