HR as the AI adoption engine: why people leaders must lead the workplace AI transformation

HR leaders ignore or underserve the human element at their peril. The skills, culture, leadership, communication, and change management needed for successful AI implementation all have a common thread: people. While HR does not have all of the capabilities required to tackle these challenges, we are perfectly poised to drive AI adoption as we support teams across our organizations. 

Let’s explore how HR can take the lead in enabling successful workforce AI transformation — leveraging their expertise in change management, learning, and talent strategy. HR also has a responsibility to advocate for employees and represent the responsible, ethical, and equitable dimensions of AI adoption. 

In addition, HR is about data, technology, and business impact. I also advocate for HR to forge an ever-closer partnership with IT, to turn AI  hype into AI ROI reality.

The central role of HR in leading AI transformation

HR needs to be a prominent leader of AI transformation in their organizations. Equally, IT needs to step into their leadership role. 

Back in May 2025, Moderna made a bold announcement about merging their HR and IT functions. A match made in heaven or destined for divorce? Time will tell. But I believe this kind of move makes a ton of sense, and we will see more changes like this in the future. Perhaps not a full merger of functions, but certainly increasing collaboration with each other and other departments such as Internal Communication.

HR is all about people, systems, and data. IT is all about users, systems, and data. Layer in AI transformation and you have a Venn diagram with heavily intersecting sets. 

Moderna may not be the first to do it, but this move is a loud confirmation of a growing reality: The walls between IT and HR are coming down, and they need to. Why? Because both functions are central to the workplace experience. They don’t just support the business — they shape how people work, how they grow, and how they interact with technology every day.

Joining forces is probably a natural progression — perhaps not for everyone, but essential for many. Here’s what’s driving this convergence. 

AI at scale is now critical

We’ve talked about digital transformation for years. But now, with AI and automation everywhere, the game has changed. Piecemeal pilots won’t cut it anymore. Companies need a unified AI strategy that breaks silos, scales effectively, and delivers real impact. HR and IT together are key to making that happen.

Patchwork systems must end

We all have application bloat. One Forrester study suggested that the average large business uses 375 apps and systems to get work done. This same study points out that people are spending nearly one-third of their work week, or 2.4 hours a day, to find the data and information required to do their jobs effectively. That’s not just inefficient — it’s overwhelming. 

Workflow automation and integrated platforms can make the employee experience seamless, from onboarding onward. But it requires tight alignment between those who manage tech and those who support people. 

The hybrid workforce is here to stay

We’ve entered a new hybrid era — one that includes humans, freelancers, AI agents, and everything in between. Planning for this hybrid workforce requires new thinking — and shared leadership — about ethics, costs, skills, and systems. HR and IT can’t do that separately anymore. Workforce planning now has a human and nonhuman “worker” element to it. 

Personalization is the baseline

We expect personalization as consumers. Now we expect it as employees. AI can enable it — but only if HR and IT are co-owning the data, the tools, and the design of the experience. AI is also a very democratizing technology, so the personal profile and preferences of the user are a more powerful construct than ever before as we think about productivity at work. 

Data intelligence beats data overload

AI feeds on data. But without structure, governance, and alignment, it’s just unstructured noise. Merging tech and HR is a step toward smarter, safer, and more actionable insights—at scale. But merger or not, data is a major consideration in any AI transformation. 

Moderna’s move isn’t just a departmental realignment — it’s a strategic signal. The future of work requires new alliances inside the enterprise. Tech and HR leading together? This may not just be the direction of travel — it could be the destination. We’ve not gone down this path (yet) at Simpplr, but the more I think about it, perhaps we should be heading down that aisle too. 

Regardless of your HR function’s path, these factors are driving the need for HR to take a leadership role in AI transformation. At its core, large-scale adaptation isn’t about the change itself but about the people who must embrace it. HR has a huge role to play in this.

Building AI readiness through change management 

For much of my career, I worked in HR consulting and started out as a change management consultant. This was a great foundational skill set for working in HR. So much of what we do is about preparing people for change and supporting various change initiatives within our own function and across the business. 

While there’s a myriad of models and theories around change management, when you look across the most commonly used frameworks and approaches, they center around some common truths about how people need to be readied for change and how they need to be supported to change. 

The most commonly used change management frameworks — e.g. ADKAR, the Kotter Eight-Step Process, or your preferred consulting firm model — converge on a set of universal truths about successful human adaptation, which still serve us well in this AI era.

These truths are:

1. The need for clarity and context 
2. The need for support to learn and change 
3. The need for involvement 
4. The need for reinforcement.

The need for clarity and context 

Before any action can be taken, people must understand and internalize the change. People need to know the why. Change is easier when there is a clear, compelling burning platform (the urgency) and a clear vision (what the future looks like). People need to understand the business reason for the change, the risks of not changing, and the outcomes (including behavior change) required.

The need for support to learn and change 

It is not enough to want to change; people must feel empowered and equipped to change. We talk a lot about resistance to change but I ascribe to the theory that it’s uncertainty that breeds anxiety, which quite naturally causes us to question and yes, sometimes be slower to accept or adopt. That’s natural. But when we have authentic and transparent communication, when we have targeted training and upskilling to practice new ways of working, momentum builds. 

The need for involvement 

People resist being done to, but they support what they can help create, which builds trust and buy-in. Actively involving people not only builds skills but also engagement, which allows us to move further, faster. AI is also a democratizing technology — it puts powerful capabilities directly in employees’ hands, giving individuals more agency to solve problems and make decisions independently. 

In some ways, AI adoption efforts feel like we’re in a race to see who can drive the most adoption fastest, and “winner takes all.” While I don’t believe all the hype, I do believe that if you’re not changing as fast (or faster) than the world around you, then you need to pick up the pace, and involvement is one sure way to do so. 

The need for reinforcement 

We know change is not a one-time event; it is a sustained process. People need support to maintain new behaviors. Measurement, feedback, and recognition are all part of this. It’s important to celebrate milestones, share wins (and learn from setbacks too), and have consistent leadership that actively models the changes they are asking of everyone else. Over time, we need to reset aspects of our operating model, e.g., how we measure, manage, and reward performance to realign to the new realities. 

So while the tried and tested foundations of good change management hold true as we build AI readiness and adoption in our organizations, there are some emerging nuances. One is what and how we measure the KPIs that will tell us we’re on track. The others I’ll address later as we think about building our AI roadmap.

Measuring success: KPIs for HR-led AI adoption

Let’s start by acknowledging that we are at the relatively early stages of AI transformation. Yes, there’s a lot of hype out there and FOMO with it. But the reality is that most of us are still at the experimentation phase. There isn’t a mature set of KPIs for us to benchmark against. Even if there were, I’d caution against making the comparison. 

Benchmarks give us some general comparisons, or hints, on where we may have places we can change or improve, but not the “answer” to aim for. Every organization’s AI implementation is unique, meaning an external success metric for a similar initiative may not account for your specific internal data quality, legacy systems, or distinct company culture. 

Relying on generalized KPIs or ROI figures risks falling into the “benchmark trap,” where you prioritize chasing an arbitrary external number rather than measuring the tangible, custom value that your AI is designed to deliver. 

There are macro KPIs we should always establish for any change effort. These are typically the ones that should have formed part of our original business case. KPIs that can answer questions such as: are we seeing ROI from the investment made in a new technology? Or, is the business impact of this change initiative showing in our financial metrics? 

At the operational level, there are some tried and tested KPIs for general change management and some emerging KPIs specifically for AI adoption. We can consider these as we construct the “dashboards” that are meaningful to our organizations. 

Before we dive into specific KPIs, there are important maxims to consider:

• Quality over quantity: The old adage “what gets measured, gets done” means stick to the impactful few.
• Ease of measurement: Keep it simple. If it’s too hard to capture and measure, look at it again. The system is trying to tell you something 
• Behavior reinforcement: Feedback, recognition, and realigning aspects of your operating system all encourage and reinforce the new ways of working. 
• Iterative improvement: Good change is agile, has strong feedback loops, fails fast and learns quickly. 

Creating a meaningful dashboard for workforce AI transformation requires balancing foundational, human-centric change metrics with newer metrics that measure actual AI utilization and skill development. 

Tried and tested KPIs for general change management include things like training completion rates, change readiness scores, change sentiment, and leadership alignment. As time goes on we tend to look at adoption and proficiency metrics with things like user adoption rates, time to proficiency, or process compliance rates. 

All quite prosaic, right? Nevertheless, some of these will still apply. But what are some of the emerging KPIs for workforce AI adoption? We can now think about combining AI with performance data to generate and refine KPIs, both with and without human intervention, as this Sloan Review article reflects.

We’re quickly getting used to the telemetry-driven metrics that track how employees are leveraging AI tools like Copilot, ChatGPT and so on. As we do so, there are more granular metrics that we can start to consider as we compile our AI adoption dashboard.

These more granular metrics include:

• AI prompts per active user: This gives us a proxy for engagement. A higher number suggests that the tool is more integrated into the daily workflow and has moved beyond novelty.
• Active AI usage time: Basic tracking of the actual time spent using AI tools. Beyond this is user feedback based on their usefulness.
• AI-assisted task completion rate: Getting toward measures of workflow integration, the percentage of specific higher-value tasks and time-saving users are actually seeing. 
• Productivity impact before and after: Comparing the time, quality, and/or efficiency of a task that is AI-assisted and not AI-assisted. 
• AI output trust/refinement rate: Tracking how often users accept the AI’s first output versus how often they need to significantly edit, refine, or discard the output.

Whichever combination of metrics we use, they need to show whether our experiments are working. They need to help us adapt and improve at an appropriate pace. And they have to affirm that the humans at the center of all this disruption and change are feeling supported to learn, are actually learning, and are growing in proficiency and productivity.

Creating your AI transformation roadmap

I would never presume to say “here’s an AI transformation roadmap” that can work for you. As tempting as it is to look for easy answers in a complex world, we know the truth is that we need to build the right approach for our organization. But we can take heart in the fact that theory is a shortcut to action, and tried and tested change theories still apply today.

We should also take heart in the fact that any change — and especially change today — should by its nature be experimental, iterative, and agile, all adjectives that serve us especially well as we seek to drive AI adoption. 

So let’s take good change management principles as a given to be woven throughout: 

• Business alignment around a vision
• Articulation of desired strategic outcomes 
• Prioritized use cases
• Means to measure value and so on.

As I said earlier, pick your change management framework and apply it. For what it’s worth, two of my consistent favorites are Rosabeth Moss Kanter or John Kotter’s writings and frameworks. 

Now let’s layer in some things that I believe have a particular nuance as we approach AI transformation. 

Data-centricity

Since AI is only as good as the data it uses, the roadmap must include a comprehensive data strategy. This involves ensuring data quality, accessibility, and robust governance (security, privacy, and compliance) to serve as a reliable foundation for all models.  We know AI hallucinates and it’s only as good as the data it’s drawing on. Data-centricity is foundational. 

Ethics and governance

Establish clear ethical AI guidelines and governance frameworks from the start. Dust off your AI policy if you have one, and write one if you don’t! These frameworks and policy must cover transparency, accountability, fairness, and risk management to ensure AI is used responsibly and avoids unconscious biases, or other unintended consequences. 

Given that we’re stepping into the unknown in a lot of ways, this requires close attention to allow for quick remediation. We need to avoid erroneous results, data decisions, or automation errors coming back to bite us later because we overrelied on AI.

Skill-building

AI adoption is a people transformation, not just a technical one. The roadmap must include plans for upskilling and reskilling the workforce, fostering an AI-first mindset, and building a culture that encourages experimentation and continuous learning. 

AI fluency should be a minimum expectation. Teams should be equipped with approved tools, alongside relevant training. AI understanding and responsible use of AI as an augmentative tool should be a core competency for the majority. 

Guided experimentation 

Implement small, focused proof-of-concepts or pilot projects for prioritized use cases. These pilots should be designed to test hypotheses, learn fast, and demonstrate tangible value to key stakeholders. Rapid cycles of experimentation will mean some failures we need to learn from — but also breakthroughs we need to rapidly scale. 

Knowledge sharing as strategy 

I was listening to a podcast recently on AI (of course!), and I was reminded of Alvin Toffler, author of “Future Shock,” in which he communicated a sentiment along the lines of: “The illiterate of the 21st century will not be those who cannot read and write, but those who cannot learn, unlearn, and relearn.” 

This is truer than ever. I think this imperative enhances the employer’s obligation to figure out how to create the right programs and support for people, and how to make this cycle of learning accessible, discoverable, and easily shareable. There’s a flywheel effect of knowledge sharing and experimentation that, for me, are the new critical components of effective change management and foundational for AI adoption.

Want to see how Simpplr is transforming employee experience with AI? Request a demo today.