Your HR Function Is Running AI Pilots. That Is Not a Strategy. That Is Controlled Drift.

There is a specific kind of organisational delusion that is harder to spot than failure. It looks like progress. It has dashboards. It has a vendor presentation. It even has a budget line. We call it the AI pilot. And for most HR functions in 2026, it is the most expensive form of standing still that money can buy.

Here is what is actually happening in the majority of Indian organisations that believe they are "doing AI in HR." Someone has implemented a chatbot on the careers page. The ATS vendor has turned on AI-powered screening and it is ranking candidates by criteria no one has formally validated. A middle manager found that ChatGPT writes job descriptions faster than the recruiter, so now the recruiter uses ChatGPT for job descriptions. The L&D team bought a subscription to an AI learning platform and the completion rates are tracked on a slide that gets presented at the quarterly review. None of these things are connected to each other. None of them are connected to a business outcome. None of them are driven by a strategy. They are driven by anxiety and opportunity in roughly equal measure. And the organisation calls it an AI journey.

The Data: Pilots Without Strategy

Gartner's data is precise on this. Fifty-five percent of HR leaders report that their organisation has piloted AI projects within HR. Only twenty percent have developed an AI strategy for HR. That thirty-five percent gap — more than a third of all HR leaders who are running AI experiments without a strategic framework to guide them — is not a coincidence. It is a pattern. And it has a cost (Gartner 2026 HR Priorities Survey).

The Real Cost of Controlled Drift

The cost is not immediately visible, which is why it persists. Failed AI pilots do not usually explode dramatically. They quietly underdeliver. The chatbot answers thirty percent of queries adequately and escalates the rest to an already-stretched HR team that now has an additional system to manage. The AI screening tool shortlists faster but the quality of shortlists has not measurably improved — or it has, but no one is measuring it against a baseline that was established before the tool was introduced. The AI learning platform has high initial registration and declining monthly active use, a pattern that would concern anyone who was tracking it as a business investment rather than celebrating it as an innovation milestone.

And meanwhile, the rate of actual HR transformation — not pilots, not experiments, but genuine redesign of how the HR function operates — has gone backwards. In early 2025, forty-eight percent of CHROs said their HR function was currently undergoing transformation. By the end of 2025, that number had fallen to twenty-nine percent. The year in which AI became unavoidable was also the year in which HR transformation stalled. The pilots are not building toward something. They are substituting for it (Gartner HR Transformation Report 2025).

What a Genuine AI Strategy for HR Actually Looks Like

What is a genuine AI strategy for HR? It is not complicated, but it is demanding. It requires, first, a clear vision — a stated position on what kind of organisation this will be in its use of AI for people decisions, and what principles will govern that use. It requires goals that are specific enough to be falsified: not "leverage AI to improve hiring" but "reduce time-to-offer by thirty percent for technical roles without reducing offer acceptance rate." It requires metrics that were established before the tool was deployed, not invented after the fact to justify the investment. And it requires a set of explicit assumptions — about what the technology can and cannot do, about what capabilities the HR team has and needs to build, about where the organisation's risk appetite sits on the spectrum from assistive to agentic AI — that can be tested and updated as the deployment proceeds.

Scenario Planning and the Three Questions

The forty-one percent of CHROs who told Gartner they cannot take decisive action until conditions stabilise are making a category error. Waiting is a decision. The organisation that waits for the AI landscape to clarify before building a strategy is making exactly the same mistake as the organisation that runs pilots without a strategy — both are allowing the technology to happen to them rather than choosing how to engage with it.

The scenario planning model that Gartner now recommends for HR strategic planning is instructive here. The average company asks one question: what business scenarios might we encounter? The progressive company asks two: what scenarios might we encounter, and how will we know when we have entered one of them? The organisation that is truly ready asks three: what scenarios, how will we know, and what actions do we start taking now for each? Most HR functions are still on question one (Gartner AI in HR Research).

The Indian Mid-Market Reality

For Indian organisations navigating this, the starting point is not a technology investment. It is an honest conversation about what problem is actually being solved. Not "we need to be in AI" — that is a posture, not a problem. The conversation is: where in our people processes does the quality of decisions genuinely affect business outcomes? Where is the bottleneck — in speed, in consistency, in the quality of information available, in the capacity of our team? And given what we know about what AI can actually do, which of those problems is AI meaningfully positioned to improve?

That conversation, done properly, produces a portfolio of three to five targeted interventions that are connected to measurable business outcomes, governed by clear principles, and sequenced based on what the organisation can actually absorb and execute. That is a strategy. Everything else is drift that has been given a more flattering name.

The Three-Stage Model: Awareness → Readiness → AI Suite

The three-stage model — Awareness, Readiness, AI Suite — is the architecture for getting there. Awareness is not about AI in general. It is about AI in the specific context of your workforce, your processes, and your people decisions. Readiness is not a training programme. It is the systematic building of the team's capacity to work alongside AI tools in ways that improve rather than merely accelerate the work they are already doing. And the AI Suite is not a product purchase. It is the culmination of a strategy that knows what it is trying to do and has built the capability to do it.

Conclusion: Pilots Without Strategy Are Expensive

The organisations that will lead the next decade of people management are not the ones that launched the most pilots. They are the ones that ran fewer experiments with more intention and built from each one toward something coherent.

Pilots without strategy are expensive. But the most expensive thing of all is the time that passes while the organisation mistakes one for the other.