Why Your Team Isn't Using AI And Five Unconventional Ways to Actually Fix It

I've watched the same thing happen at organisation after organisation, and it goes like this.

Leadership comes back from a conference fired up about AI. Tools get purchased. A company-wide email goes out. There's a kickoff meeting with a slide deck and genuine energy in the room.

And then, six months later, barely anyone is actually using them.

The tools sit there. The mandate fades. The team carries on exactly as before and leadership is left quietly wondering whether the problem is the technology, the team, or both.

It's neither.

The problem was never the tools. It was never the people. It was the approach. And after watching this pattern repeat more times than I can count, I've stopped being surprised by it. I've started being very clear about what actually works instead.

AI adoption doesn't stall because your people are resistant. It stalls because the environment you've built hasn't made it safe, clear, or rewarding to try.

That's a system problem. And like every system's problem, it has a very solvable solution.

What follows are five strategies that are unconventional and that's precisely why they work. The usual approach has already failed. These don't look like the usual approach.

Run an AI Immersion Week

The first time I recommended this to a client, they looked at me like I'd suggested something reckless. Dedicate one entire week where your team uses AI for every single task. Drafting client emails. Running internal status updates. Building reports. All in, no exceptions.

It sounds bold. That's the point.

What happens during immersion is remarkable and consistent. People stop seeing AI as a threat and start seeing it as a collaborator. They discover capabilities they never expected and equally important, they discover the real limitations. That second part matters more than people think. Discovering what AI can't do kills more fear than any reassurance ever could.

Right now, only about a third of employees use AI tools on a weekly basis. Immersion week closes that gap fast not because it forces adoption, but because it removes the ambiguity that was making people hesitate.

"You can't gradually get comfortable with something that changes how you work. At some point, you have to just work differently and immersion week makes that the only option."

Put a Contrarian in the Room

Here's something counterintuitive: the reason most AI strategies collapse isn't too much scepticism. It's too little.

Executives are significantly more optimistic about AI than the people actually doing the work. When strategy gets built only from the top, it gets disconnected from operational reality very quickly and a disconnected strategy doesn't get adopted, it gets tolerated.

The fix is structural. Appoint someone in every AI planning meeting whose explicit job is to push back. Not to obstruct, to ask the questions that need asking. What happens when this breaks? Who owns the output? What does this do to the people whose workflow it replaces?

The strategies that survive honest scrutiny are the ones your team will actually trust. Resistance isn't the enemy of adoption unchallenged assumptions are.

Let People Choose Their Own Use Case

This one transforms everything, and I've seen it play out the same way every time.

Instead of assigning AI workflows from above, run what we call an AI Brain Boost: each team member identifies one task in their own role where they believe AI could genuinely help them. They own the idea. They design the experiment. They run it.

People commit deeply to what they help create.

When your content manager builds her own AI-assisted briefing workflow, she doesn't just use it. She defends it, refines it, and shares it with the whole team. That's the kind of adoption that doesn't need a mandate because it's already self-sustaining.

"The difference between an AI workflow that gets adopted and one that gets ignored is almost always ownership. Give people the authorship and the adoption follows."

Build AI Into How You Evaluate People

Without accountability, adoption is optional. And optional almost always means it quietly doesn't happen.

Tie AI skills into performance reviews and make it genuine, not performative. That includes the technical capabilities like prompt engineering, but also the human ones: curiosity, willingness to experiment, comfort with iteration and failure.

Research consistently shows that when organisations invest meaningfully in AI training, employees don't just get more competent, they get more optimistic. Confidence is the bridge between knowing a tool exists and actually reaching for it in the moment it's needed.

If AI skills don't show up in how you evaluate people, you've already told your team it's optional. They'll act accordingly.

Make the Wins Visible and Make It Fun

Celebrate the small stuff loudly.

Run a hackathon. Host an internal competition where teams solve a real operational problem using AI. Give it a name. Make it a story worth telling.

The energy you build around early wins is the fuel that carries everyone else forward. This isn't gamification for its own sake, it's about building collective belief. The belief that using AI is something your organisation is genuinely good at. That belief, once it takes hold, is self-reinforcing.

"People don't adopt tools. They adopt identities. Make 'we're a team that's great at using AI' the identity and the tools follow."

I've seen teams that were completely disengaged from AI mandates, eye-rolls in the kickoff, radio silence in the Slack channel, tools gathering digital dust become the most innovative operators in their sector once the approach changed.

Not because the tools changed. Because the environment did.

Fix the environment, and the people follow. Every time.

If you're tired of pushing AI uphill and ready to build a team that actually runs with it.

Book a free AI workflow review with us. We'll map exactly where your processes are AI-ready today and where to start so you get real traction in the first 30 days, not another stalled rollout.