When it comes to workforce planning around artificial intelligence, many organizations are running before they can walk. That was one of the key messages from Harsh Kundulli, vice president analyst at Gartner Business and Technology Insights, and Katie Sutherland, director at Gartner, during their keynote presentation at the firm’s CHRO Symposium this week in Orlando.

“Before we can answer questions about the workforce, we need to answer questions about the work, because work is changing,” Kundulli told the audience. “Work is changing because our CEOs and CFOs are setting exacting standards. They want more growth, more efficiency and they’re turning to AI to get it.”

According to Gartner’s data, most companies have adopted some form of AI, and the majority of CEOs plan to increase investments. But the results haven’t always matched the optimism. “Only one in three AI initiatives boost productivity,” Kundulli said. “One in five delivers measurable ROI.” The bottom line, according to Kundulli, is that AI has yet to significantly impact key business outcomes such as revenue growth or cost optimization.

See also: HR leaders, are you embracing your ‘superpower’? Here’s how to start

HR’s new focus: the work itself

For HR leaders, that disconnect translates to greater pressure. Kundulli said HR is being asked to cut costs, find scarce growth talent and still deliver value using a fast-evolving technology. That, Sutherland argued, means HR must rethink its role entirely. “In the AI era, HR has a new mandate: shift the focus from the workforce to the work,” she said.

Historically, HR has focused on people: hiring, training, culture and leadership development. But as AI reshapes how work gets done, Gartner leaders say HR’s mandate must expand to redefining the work itself.

Kundulli described three major shifts in how organizations are using AI:

• Augmenting work: using AI to make existing workflows faster or more accurate.
• Re-engineering work: redesigning processes entirely around AI.
• Inventing work: creating new, AI-based ways of operating.

Most HR leaders, he said, expect to spend the bulk of their time in the next year on augmentation, followed by re-engineering and invention. “Your organization is going to be a portfolio of changes,” Kundulli noted. “Each comes with different costs, risks and returns.”

Augmentation may seem like the least disruptive path, but it isn’t delivering enough value on its own. Gartner’s research found that while most employees are using AI tools, few know how or where to apply them effectively. “Employees aren’t resisting AI,” Sutherland said. “They just haven’t figured out where to use it.”

Kundulli emphasized that HR must take a more active role in guiding AI adoption. “When you guide employees’ AI use, the whole curve shifts,” he said. “Fifteen percent of employees become AI superstars, and the entire workforce moves toward value.”

That guidance often requires closer collaboration between HR and IT. As Kundulli put it, “HR and IT are two sides of the same coin. Together, they will build the future of work.”

Read more | AI agents: Where are they now? Insights from industry experts

Managing ‘ripple effects’ of re-engineering work

The second type of change, re-engineering work, comes with greater stakes. Kundulli described a financial services firm that automated a costly process without consulting HR. “In HR’s absence, the business failed to consider the talent implications,” he said. “All the entry-level roles were gone. Where would the next set of experts come from?”

He called the consequences “ripple effects,” adding, “CEOs and CIOs ask, ‘Can a machine do this?’ HR must ask, ‘What are the implications if it does?’”

To get ahead of those ripple effects, Sutherland urged HR leaders to create clear foundations rooted in business strategy. “These principles turn organizational chaos into coordinated transformation,” she said.

Re-engineering work also requires coordination across the enterprise. Kundulli described how one company established a workforce planning council that brings executives together to agree on the top areas of work change. Cross-functional “pods” then execute the redesigns on the ground.

Another organization has created a control tower to oversee its top 50 generative AI use cases, focusing on those that will deliver the most value.

“Every re-engineering project creates ripple effects,” Kundulli said. “You might improve your cost structure, but you’ll create new challenges to fix. HR’s job is to get ahead of them.”

Preparing for invention

The third shift, inventing entirely new ways of working, may be the most transformative. Gartner predicts that by 2028, 15% of day-to-day decisions will be made autonomously by AI agents.

But Kundulli cautioned against overreacting to the hype. “Agentic AI is at the peak of inflated expectations,” he said. “Less than 1% of job losses this year were due to AI productivity gains.”

He warned that organizations moving too fast to cut headcount in anticipation of AI gains could find themselves rebuilding later. “It would be a mistake to lay off large portions of the workforce now, hoping AI will pick up the slack,” he said.

Instead, HR should work with IT to reset expectations and build redeployment plans that move employees into growth areas. “It might not sound like HR’s traditional job description, but that’s what makes this new era so exciting,” said Sutherland. “Your organization needs you to make this era of work a success. The choices you make today will shape your workforce, your technology and your world.”