IBM CHRO Nickle LaMoureax doesn’t contemplate herself an AI optimist. As an alternative, she calls herself an AI realist.
“AI is just not magic. It’s wonderful, spectacular know-how that may completely rework your enterprise,” she informed a packed room of attendees throughout her opening keynote at Tuesday’s HR Tech convention in Las Vegas. “However it takes laborious work, habits change, tradition change, enterprise course of change and typically management change.”
LaMoreaux, HR Govt’s 2024 HR Govt of the Yr, skilled all of these realities as she led IBM’s AI in HR journey over the past decade—one which has been rife with each progress and pitfalls.
“It was a rocky highway. It was something however wonderful,” says LaMoreaux, who has been with IBM for 25 years and now leads all HR for all 300,000-plus workers throughout 170 international locations.
The tech big began down the AI in HR path in 2017, when it gave all workers entry to Watson Assistant. Earlier than they knew it, workers had changed into “citizen builders,” and chatbots have been “working round” in every single place. HR alone had 30 completely different chatbots, with very restricted governance.
That prompted LaMoreaux’s group to centralize all of these instruments into one resolution: AskHR, which they envisioned as a one-stop store for all issues HR. However there was one massive drawback: Nobody was utilizing it.
So, the group made the “daring” resolution to power the change: In a single day, they shut off the HR e-mail handle and telephone quantity and informed 21,000 frontline managers they now not had entry to their HR enterprise companions. All of their HR questions must undergo AskHR.
RELATED: Extra from HR Tech 2025
Unsurprisingly, LaMoreaux stated, the transfer didn’t go over too nicely with IBMers. Earlier than that call, HR’s worker Internet Promoter Rating sat at +19; after, it dropped to -35.
However the group continued, she stated. With budgeting challenges, an more and more regulated atmosphere and evolving worker expectations for tech experiences at work, “we knew AI was our solely means out of this,” she stated.
Listening and studying
LaMoreaux’s group constantly discovered from AskHR—what was and wasn’t working, together with information culled from interactions that bought a “thumbs down”—stored management knowledgeable and targeted on steady enchancment. On the similar time, the AI was getting smarter and HR shifted its working mannequin: Funneling complicated AskHR requests to expert consultants—not in conventional HR domains like expertise acquisition however in areas that would carry extra significant help to workers and managers. For example, a supervisor consulting AskHR a couple of low-performing worker could also be handed alongside to an HR skilled particularly expert in creating studying plans.
“We created these cross-functional consultants to reply questions from the standpoint not of HR silos however of what the issue was,” she stated.
HR additionally revamped the answer to embrace agentic AI—treating AskHR as an orchestration layer for all of IBM’s HR programs—enabling extra seamless transactions.
At this time, AskHR is skilled in 52 languages and 30 HR domains. It handles 11.5 million transactions yearly—with 100% adoption by managers and 99% adoption by executives. At present, 94% of worker questions are dealt with completely within the system. Final yr, half of workers and managers by no means logged into an HR system in any respect—and LaMoreaux stated she’s on the trail to getting that determine to 100%. The dramatic uptake in adoption of AskHR has led to a 40% discount within the HR working funds.
And right now, HR’s eNPS sits at +74.
See additionally: What is going to it take to be a real ‘future-ready’ org?
5 classes from IBM’s AI in HR journey
Such a turnaround didn’t come rapidly—or simply. However it did convey plenty of significant insights about what it takes to construct an AI-first HR operate efficiently. Amongst them:
Overlook FOMO
Don’t comply with “greatest practices” for integrating agentic or generative AI—as a result of they don’t exist, LaMoreaux stated. As an alternative, concentrate on whether or not the AI helps your specific enterprise mannequin, expertise mannequin and tradition. AI shouldn’t be deployed in every single place. For example, IBM is “fairly far superior” in utilizing it for compensation practices, however nonetheless within the early phases on the subject of expertise acquisition. “Simply because one other firm or your competitor is doing it,” she stated, “doesn’t make it best for you.”
Begin small and focused
Excessive-volume transactions are normally place to begin integrating AI in HR. Drive much more that means by specializing in “moments that matter.” For example, at IBM, AI was in a position to minimize the supply of employment verification letters from as much as two days to about 30 seconds, which made a major distinction for workers. LaMoreaux additionally suggested HR to concentrate on the method or coverage that attracts probably the most complaints. “You’re not going to make it worse,” she laughed. For IBM, that was worker transfers, a “dreaded” and sophisticated course of carried out in HRIS; AskHR now handles all 80,000 annual transfers, with a close to error-free report.
Repair processes first
Throwing AI on high of unhealthy processes gained’t enhance them—however it’ll waste money and time. LaMoreaux adopted a easy framework for course of enchancment: eradicate, simplify, automate. “Earlier than something went into AI, you needed to show why it wanted to exist in any respect,” she stated.
Be user-centric
As HR builds AI into the operate, rethink general working fashions to make sure worker and supervisor wants are on the coronary heart, LaMoreaux stated. For example, are there mechanisms to make sure suggestions concerning the tech?
Prioritize tradition
HR features can deploy the very best know-how on this planet—however with out the tradition to help it, the hassle will fall flat, she stated. For example, IBM had the tradition that enabled her HR group to journey out that -35 eNPS, however not all organizations will—so, perceive the tradition during which you’re working earlier than advancing in your AI in HR journey.
HR’s accountability as an AI ‘myth-buster’
As HR leads this work, it might must take care of disruptive views. It is going to be as much as HR to bust “myths” like:
1. Area experience is now not essential.
Each for the HR operate and companywide, HR ought to really “double down” on the worth of area experience. AI will do the fundamentals, however it’ll by no means have the judgment, knowledge and expertise human workers carry. “I inform my HR workers, ‘You’ll want to be one step smarter than AI always,’ ” LaMoreaux stated.
2. The early profession ladder is damaged.
Dialing again entry-level hiring as AI takes over duties could make a CFO glad for a yr—however what about a number of years down the road, when mid-level expertise is required?
“The place is that supposed to come back from?” LaMoreaux questioned. She really wants extra entry-level HR professionals than earlier than IBM deployed AskHR—they’re simply doing completely different work: coaching to reply complicated questions or interacting instantly with workers who report a unfavourable expertise with the tech.
3. AI runs wild.
It’s “essential” to have a mechanism to observe your AI—so in actual time what’s taking place and may catch hallucinations or issues rapidly. “You might be within the driver’s seat about when, the place and the way you utilize AI,” she stated. “Whether or not or not AI runs wild is basically in your fingers.”
4. Huge job loss is inevitable.
Headlines abound on this matter, LaMoreaux stated—however it doesn’t “compute” to her. AI will solely take jobs at organizations the place progress isn’t anticipated by traders, shareholders and different stakeholders. “However the pie is rising” at most organizations, she stated. “AI is doing a little work, however it’s vital for us to consider redirecting workers to the place the expansion is in your group. [AI-driven job loss] is a serious worry—whether or not individuals admit it or not—in most organizations that’s going to get in the way in which of AI adoption.”
A brand new second for HR
LaMoreaux urged attendees to be daring on the subject of AI in HR: Regardless of the plan is for integrating the tech, push it a step additional.
And, assume outdoors of the operate. HR has the chance to not solely deploy AI strategically in their very own work however throughout the enterprise.
“It’s your time to be within the solar,” she stated.
