HR STRATEGY · AI & THE FUTURE OF WORK
Why HR Should Own AI Strategy — and the Missed ROI if They Don’t
Alygra People & Growth · May 2026 · 8 min read
You don’t often hear ‘AI in HR,’ yet its core, AI adoption is a people problem. Technology teams can disseminate the tools across organizations, but whether those tools translate into genuine capability, employee development, and building something durable for the business sits squarely in HR’s domain.
HR’s role is larger than most organizations give it credit for, and HR itself has often been slow to claim it. Beyond that, most HR functions have fallen into this new role by default rather than design, becoming responsible for AI adoption without a clear picture of what it looks like.
At present, we’re seeing a key moment between the AI strategy kickoff and the first quarter of missed adoption targets where this question comes up: “And what is HR’s role in AI?”
The answer isn’t clear yet, but to make it so, here are Alygra’s four key tenets in AI adoption and how most organizations are underperforming in all of them.
01 AI Literacy in HR
Current State
The most consistent gap we see is that many HR leaders are trying to build AI capability across the organization while their own AI literacy is still developing.
This isn’t a criticism. The pace of change in AI has been genuinely disorienting, and HR has been asked to lead adoption strategy in the middle of a wave that no one has ridden… but this creates problematic ripples.
Ideal State
AI literacy in HR doesn’t mean becoming highly technical – it means understanding what AI can do today:
The powerful application of reasoning models
How agentic workflows can change the structure of knowledge work
Translating what individuals across the organization are already building with AI tools into meaningful organizational language
Beyond that, it’s offering valuable contributions in discussions with CEOs and CTOs on what’s needed from your people when AI can do ‘X.’
Once HR has that fluency, it can own the architecture of AI literacy across the organization. Not just training programmes, but role-level competency frameworks that define what AI fluency means for a finance analyst, versus a product designer, versus a customer success manager, with development pathways that evolve over time.
Important Watch-Out
An employee who has finished Module 3 of an AI course has a certificate and some AI literacy. An employee who has changed how they work with new tools has AI capability. HR needs to measure the latter.
You cannot build AI literacy across an organization if you don’t have it yourself. That’s where we ask every HR function to start.
02 Building AI Capability in Organizations
Current State
When organizations say they’re ‘addressing AI adoption through training’ what they usually mean is: they’ve purchased licenses for a learning platform, sent a cohort to a workshop, and are tracking completion rates. This is activity, not capability building.
Ideal State
Real capability building requires HR to ask harder questions that go beyond what skills people are missing and into what the organization will need from people in 18 months, i.e.:
Which roles are changing most substantially? And how do we proactively redesign them before the gap between current capability and new requirements affect performance?
What does AI-augmented work look like in each core function? And are we building tailored career pathways toward that future?
Where is AI creating capabilities we need to hire for? And where is it creating capabilities we need to grow internally instead?
How do we measure the growth of AI capability? How do we tangibly assess the change in ways of working as opposed to course completion rates or certificates?
Important Watch-Out
HR needs to own the integration of AI into teams because they know people, roles, and performance architecture best. By leaving it to engineering or to AI vendors, you risk it being seen as a technology rollout where tools are deployed, yet people don’t develop alongside them. Then the capability gap widens.
The organizations generating real ROI from AI aren’t the ones with the most licences. They’re the ones with the clearest picture of what AI-augmented performance looks like and where HR helped design that picture.
03 AI Governance in HR
Current State
In most organizations, AI governance sits with legal and compliance; data privacy frameworks, acceptable use policies, and GDPR alignment. These are necessary, but not sufficient on their own. HR needs to be an integral part of this, not simply in their typical processes such as screening resumes, scoring performance, recommending promotions, flagging attrition risk, and generating compensation benchmarks.
Where AI is influencing decisions that affect people’s careers and livelihoods, HR has a direct responsibility to ensure those decisions are explainable, auditable, and fair (and being fair is important and a very different standard from being compliant).
Ideal State
AI governance in HR needs to go further than it does at present and be rooted in this: “Where is AI being used to make or influence decisions about people?” Here’s what it means in practice:
A Seat at the Design Table: When a new AI tool is being considered in hiring, performance management, or workforce planning, HR should be part of that evaluation before a decision is made, not after it’s rolled out.
Accountability Clarity: When a problem occurs in an AI-influenced HR decision (i.e. a candidate is incorrectly screened out or performance score is misapplied), who is accountable? And does the affected individual know AI was involved? Make this clear.
Contestability by Design: Those impacted by AI-influenced decisions should have access to a defined, transparent process allowing them to question such decisions. HR needs to have a framework and the capability to support this.
Important Watch-Out
Effective AI governance isn’t intended to be restrictive. In HR’s world, it should build the institutional trust needed for AI-assisted decisions to be accepted — and challenged — by people in your organization.
04 AI Ethics in the Workplace
Current State
Many see AI ethics as an engineering or legal item. Others see it as theoretical. But the Alygra team knows they are, in fact, operational, immediate, and arriving faster than most organizations have built the structures to handle them.
HR is best positioned to hold this conversation; to design the frameworks, create the language, and build the psychological safety for people to raise concerns before harm occurs.
Ideal State
HR teams navigating AI adoption well have clarity in these following areas, prioritizing proactivity over firefighting.
When AI automates a task or workflow previously owned by a person: What is the organization’s obligation to that person – reskilling, redeployment? What’s the timeline for this to be communicated and supported?
When AI-generated data informs a performance or pay conversation: Is the employee aware? Do they have rights over that data and a means to access and question it?
When AI fundamentally changes what a role requires: What do you owe the people hired under different expectations who might not have had the opportunity or the resources to develop the new ones?
How does the organization ensure that AI tools used in hiring or promotion are not encoding or deepening existing bias: Who audits this, at what frequency, and against what standards?
Important Watch-Out
Ethics is most powerful as a design input. It’s the least powerful as a response to an incident that has happened. How organizations prioritize and handle this signals what kind of employer they’re committed to being as the nature of work changes.
Summary: Alygra’s four HR tenets in AI adoption should be interconnected.
AI governance without AI literacy is policy without real understanding. AI ethics without the development of AI capability is principle without practice. And AI literacy without AI governance is adoption without real accountability.
The HR functions effectively adopting AI are building all strands simultaneously. They’ve been given the mandate, the resources, and the positioning within their organization to do it.
At Alygra, we’ve found that the fastest route to progress is an honest audit of where the organization stands on each tenet – not where the AI strategy document says it is. In some areas, AI adoption is further along than teams expect, but in others it’s more exposed than they realized.
AI adoption will succeed or fail at the people layer, and this layer is owned by HR. Most functions haven’t fully claimed it yet.
RELATED TOPICS
AI Literacy · Capability Building · AI Governance · HR Strategy · Future of Work · Fractional HR · People Operations
