New research reveals what's really separating AI winners from everyone else
Most companies still treat work redesign as a project: a few intense months mapping workflows, then a return to business as usual. Gartner's newest research argues that model is now a liability. As AI changes how work gets done faster than most organizations can adjust for it, the firm predicts that by 2028, companies that make continuous work redesign a core capability will be twice as likely to sustain AI-driven work transformation.
Caroline Hewings, senior director analyst in the Gartner HR practice, a business and technology research and advisory firm headquartered in Stamford, Connecticut, said the shift reflects a pattern she's tracked across her career.
"Whereas in the past it used to be, let's all get in a room and map out a process and see how we can optimize it and make it better and take steps out. That's still a baseline to get to the next level. But to get to the next level, and now with AI, the pace of that is changing. It's not a one and done. This needs to be continuous," she said.
Companies lost a skill they used to have
Hewings said many companies once built real rigor around continuous improvement, but stopped investing in it over time.
"So many organizations have lost that," she said. "You just invested less in it over the years, and that is a foundational baseline."
Rebuilding that rigor, Hewings said, comes down to three components Gartner's research treats as equally critical: process excellence as a baseline, technology like process intelligence and task mining to extend it, and adaptability at both the individual and organizational level.
"People come to us asking, how do I redesign jobs? I said, no, no, no, this isn't about the job. This is about redesigning the work. And that requires process excellence, that requires you digging in, doing the ground work. This is hard," Hewings said.
That same emphasis on the work itself, not just the technology behind it, echoes a theme HRD America has explored in reporting on how HR can future-proof organizations through AI, where HR teams that start with the work and its impact on people, not just the tools, are best positioned to help AI enhance roles instead of creating anxiety.
Visibility into work, not into employees
Hewings described process intelligence and task mining as tools that finally show leaders how work actually happens, not just how it's supposed to happen on paper.
"Some of the tools now available let AI give you visibility into what's happening in real time. This isn't about putting Post-its on walls and mapping out a process. You've got tools that can help you take that to the next stage," she said.
That kind of visibility, she said, comes with a real risk of being misread. Some employees might assume it's about watching them individually, when the technology is meant to show how work is performed, not to track any one person.
"HR's role is governance: making sure it's understood what data is collected, and that there are safeguards protecting employee visibility. Who's accountable? How are employees involved in this?" Hewings said.
That tension between efficiency and oversight is exactly what surfaced when employees pushed back against Meta's workplace tracking software, forcing the company to scale back its plans.
Someone needs to own where AI fits
Redesigning a process, Hewings said, means making deliberate calls about where AI belongs and where a person needs to stay in the loop, and writing those calls down.
"That's when we talk about decision rights: thinking about who should own what, and what accountability leaders and employees have as work is redesigned," she said.
That responsibility, Hewings said, doesn't rest with a single department.
"Who owns this is often a debate. Is it the business? Is it HR? Is it some central office? Probably all of the above. Understanding that this does require an army, it's a team sport, but I think HR becomes responsible for helping the enterprise build more of a discipline around it, alongside the leadership team," she said.
Redesign is not the same as restructuring
Hewings was careful to draw a line between constant redesign and constant upheaval.
"Redesign work does not mean continuous restructuring. There's a nuance there. You're not going to be overhauling the organization, but tweaking the process and optimizing it. That's why the message on adaptability is so important, so people get into that mindset," she said.
She described that adaptability as a skill in its own right, one she thinks belongs alongside more familiar measures of workplace performance.
"If in the past EQ and IQ were important, the adaptability quotient is going to be important too," Hewings said. That plays out on two levels, she explained: individuals need to stop thinking in terms of a fixed job description and start thinking of a purpose that evolves over time, while organizational processes need to adapt and change faster than they have in the past.
She used software engineering as an example: engineers whose day-to-day work has shifted from writing code themselves to directing agents that write it. The purpose of the role hasn't changed, she said, but the way people fulfill it has.
It's not just software engineers whose day-to-day has shifted. HRD America has reported on managers going through something similar, as AI absorbs the dashboards, scheduling and first drafts, leaving managers doing less of the work and more of the talking when it comes to aligning teams and building trust. A recent Pew Research Center survey found more than half of U.S. workers are worried about AI's effect on their jobs, a reminder that this isn't just an abstract shift for the people living through it.
That pressure is showing up in the numbers, too. Deloitte's 2026 Global Human Capital Trends research found that companies that redesign work around AI, rather than simply cutting headcount, are roughly twice as likely to exceed their AI return-on-investment expectations.
Hewings summed up what's at stake.
"Organizations that succeed in the AI era will not be those that redesign work once and consider the job done. They will be those that build the capability to redesign work in real time as technology, business priorities and workforce needs evolve," she said.
