Why AI Adoption Fails: It's Not the Tech

At the TIME100 AI Leadership Forum, Nigel Vaz said something most consultants won't say out loud: the obsession with AI technology is the problem.

Not the technology itself. The obsession with it.

Companies spend months evaluating models, benchmarking tools, debating stack decisions. Meanwhile, the people who actually run the systems being replaced are going through something much harder. They're grieving. Old workflows, old hierarchies, old ways of knowing what good work looks like — all of it is being pulled away.

Nobody is talking about that in the board decks.

Vaz runs Publicis Sapient, a digital transformation consultancy that works with large enterprises on exactly this kind of change. He's not a skeptic on AI. But he's seen enough rollouts to know that the technology landing correctly is not the hard part. The hard part is the people sitting across the table from it.

Here's the thing about organizational change: it doesn't feel like progress from the inside. It feels like loss.

The team that spent ten years building a reporting process that works — and works well — doesn't cheer when a model replaces it in an afternoon. They feel redundant. They feel exposed. They feel like the ground shifted under them without warning.

Research on AI change management treats this seriously. Grief is the right word. There's an attachment to systems that delivered results, to skills that took years to build, to professional identity tied to doing a specific thing well. AI doesn't acknowledge any of that. It just outputs.

Vaz's point is that ignoring this grief doesn't make it go away. It just turns into resistance. Passive non-adoption. Workarounds. Quiet sabotage. And suddenly the AI deployment that looked great in the demo is stalled six months later and nobody can explain why.

You can't skip the human part.

So what actually works?

Vaz's answer is simple and not particularly comfortable: you have to show people what's on the other side. Not in vague terms. Not 'AI will make us more competitive.' Specifically. Concretely. What does their day look like after? What problem disappears? What do they get to stop doing?

His argument is that the outcomes AI can produce — what he calls exponential outcomes — are genuinely hard to believe until you see them. Time savings that sound made up. Cost reductions that don't fit inside normal planning assumptions. Capabilities that weren't possible at all six months ago.

But people won't push through the grief and the friction and the learning curve on faith. They need to see the destination clearly enough to decide it's worth it.

That's a leadership job. Not a technology job. The model doesn't do that part.

There's a moment in every serious AI deployment — if it gets far enough — where the argument stops.

Vaz describes it as the point where you start thinking about the exponential nature of what you can actually affect. The kinds of systems you can build. What they cost. How long they take. At that point, he says, the value becomes self-evident. You don't need to convince anyone anymore.

Publicis Sapient's AI practice is built around getting clients to that moment. The pitch is not 'trust us, it works.' The pitch is 'let's get you to the point where you can see it yourself.'

The McKinsey State of AI 2024 report puts numbers behind this. Companies that move past pilot stage and into real deployment see results that early-stage adopters don't. The gap between those who've pushed through and those who haven't is widening fast.

But you only get there if you actually push through.

33 seconds. That's how long Vaz needed to say something most AI keynotes take an hour to avoid saying.

The technology is not the bottleneck. You are. Your team is. The gap between what AI can do and what your organization will let it do — that gap is a people problem.

If you're leading a company through an AI transition right now, the question isn't which model. The question is: does your team understand where you're going clearly enough to tolerate the discomfort of getting there? Have you acknowledged that losing familiar systems feels like loss, even when the replacement is better?

Most leaders haven't had that conversation. They've sent the memo about the new tools. They've run the training. They've launched the pilot.

But they haven't sat down and said: I know this is hard. Here's why it's worth it.

That's the conversation that makes the difference.