AI adoption is soaring, but real ROI requires moving past basic usage—embedding intelligence directly into daily workflows to drive outcomes.
Sara Braun
Editorial Strategist, HR
Workday
Audio also available on Apple Podcasts and Spotify.
AI adoption is surging, yet for most organizations, real enterprise value remains frustratingly elusive.
On Workday’s Future of Work podcast, Nancy Weitl, VP of HCM product management at Workday, and Stacey Harris, chief research officer and managing partner at Sapient Insights Group, unpacked this disconnect. Workday research reveals a striking gap: 82% of employees feel AI systems are detached from their actual work, even while 97% report feeling productive.
That disconnect matters. The real issue isn’t whether people are experimenting with AI; it’s whether AI is embedded deeply enough to remove friction rather than add to it.
As Harris put it, "Adoption is actually not the right question." Her research shows over 90% of HR employees are already using AI in some form. The bigger question is whether that usage is tied to enterprise workflows, supported by leadership, and engineered for real outcomes.
Report
There is a key difference between being busy and being productive.
AI can help people move faster. But if it simply creates more content, more coordination, and more review cycles, it can also create a new layer of busywork. Harris notes that while recruiting, onboarding, and process efficiency were early AI use cases in HR, document creation has now surged as a leading use case.
That creates a hidden tax: AI may help people draft more quickly, but it can also leave everyone else with more to read, review, summarize, and reconcile. “I think we haven’t quite figured out yet how to make AI inside of our worlds more seamlessly,” says Harris.
For HR leaders, this exposes a critical measurement strategy. Too many organizations still measure AI success with activity metrics: logins, training completion, usage rates, or the number of AI-enabled projects. Those signals prove interest, not impact. Instead, leaders should measure outcome-based indicators:
Productivity gains should not come at a human expense. Harris emphasizes that performance metrics need to be balanced with burnout metrics so organizations do not simply use AI to "squeeze as much out of a person" as possible.
"We haven’t quite figured out yet how to make AI inside of our worlds more seamlessly."
Stacey Harris
Managing Partner
Sapient Insights
AI works best when it is embedded in the systems where work already happens.
Organizations using AI built directly into core enterprise systems are 1.7x more likely to see meaningful time savings, according to Workday research. Moving data into standalone tools risks context loss, security breaches, and compliance violations (e.g., GDPR, HIPAA). Embedded AI operates natively within existing workflows, permissions, and business rules.
The takeaway for HR leaders: Don't spray AI everywhere. Architect an environment where every tool has a clear purpose, sensitive data stays protected, and AI is applied where it drives actual impact.
Organizations using AI built directly into core enterprise systems are 1.7x more likely to see meaningful time savings.
So where should leaders begin? Harris recommends starting with skills over job titles.
Traditional workforce planning often treats a person and a role as a single unit. AI forces a more granular approach: dissecting roles into actual tasks, identifying required skills, and mapping where AI can assist without degrading human value.
That requires more direct conversation with employees. Managers may know the outcomes they want, but they often do not know every step employees take to get there. Employees are the ones holding together the handoffs, workarounds, approvals, and manual steps that make systems function.
Instead of redesigning work for employees, leaders must redesign work with them. "Let the employee start to document how work gets done—not to take their work away, but to let them decide what work is most valuable," Harris advises.
Ask them directly: Where are you losing time? What bogs down coordination? Where could AI help you do more meaningful work, rather than just more volume?
One of Harris's most compelling warnings comes from legal workflows. AI excels at sourcing data, organizing case notes, and drafting briefs. But if that task was traditionally handled by an intern, something vital breaks when it's automated away.
The intern wasn't just churning out a document; they were synthesizing knowledge and engaging in critical dialogue with a senior attorney to build judgment. Automating the task yields speed, but sacrifices human development.
When deciding between human and machine capabilities, don't ask what AI can do. Ask: What learning or connection do we lose if this human interaction disappears?
In lower-readiness organizations, AI handles isolated tasks while humans act as the manual glue connecting broken systems. In high-readiness organizations, AI owns clear operational steps—monitoring, routing, drafting—while humans focus on judgment, strategy, and empathy.
HR is uniquely positioned to lead AI adoption because it understands performance, skills, governance, and employee experience.
Looking ahead, Harris expects the next generation of workforce technology to become conversational, not just automated.
Employees don't want another form to submit. They want to ask a question, engage in a back-and-forth dialogue, and solve problems directly in the flow of work. Consumer AI tools have set this standard, training people to research, iterate, and decide through natural conversation.
This shift grants employees greater autonomy over areas traditionally gated by management—like flexible scheduling, pay preferences, tailored learning, and career mapping. For managers, it offers unprecedented visibility into operational realities. For executive leaders, it yields an agile workforce built around dynamic skills and outcomes.
But getting there requires a mindset shift.
AI adoption is not an IT rollout; it is an enterprise work-redesign challenge. HR is uniquely positioned to lead it because HR understands performance, skills, governance, and employee experience. The organizations that win won't be those with the highest login rates—they'll be the ones that make AI seamless, conversational, and almost imperceptible.
Report
A gap between effort and impact is the real ROI problem. Discover the solution.