Here’s the thing about AI...

Most businesses know they need to do something with it. Fewer know exactly what. And even fewer have worked out how to bring the whole team with them, not just the people who were already curious.

That’s the bit I’m interested in.

I’m Head of People at Gecko, and AI is one of our core priorities for 2026; not as a vague ambition, but as something we need every part of the business to genuinely engage with.

The biggest risk was never the technology, but the gap between the people who had already embraced it and the people who hadn’t really started yet.

Left alone, that gap only really gets wider. The people who are already experimenting improve quickly, while everyone else quietly concludes AI “isn’t really for me”. Before long, you’ve accidentally created two different experiences of work inside the same company.

We’re by no means experts ourselves, but I think we got one thing right.

Rather than talking about AI adoption, we created the conditions for it.

Getting everyone in the same room

Gecko is a remote, distributed company. We have people across the UK, from as far south as Cornwall to as far north as Aberdeen, plus a US team. Getting everyone together isn’t simple or cheap, so when we do it, we try to make it count.

This May, our Geckos came together in Edinburgh for our Q2 meet-up.

The brief was simple.

Connect. Learn. Move Gecko forward.

The first day looked a bit different

We put it entirely in the team’s hands, and they ran with it.

Go-karting, a treasure hunt, a round of golf, a pub quiz… people organised whatever they fancied. More importantly, it created space for the conversations that don’t always happen on Slack or Gather (our virtual office).

It sounds simple, but it was probably one of the most important decisions we made.

You can’t ask people to spend a day trying something new if the only interaction they’ve had for months is scheduled meetings and Slack messages.

Remote working gives us huge flexibility, but it can quietly chip away at the spontaneous conversations, shared experiences and relationships that make collaboration easier.

Looking back, I don’t think the activities themselves were the important bit. They created an environment where people felt more connected before we asked them to spend a day learning something completely new.

The next day (aka the main event): getting stuck in

We built the day ourselves.

As a Leadership team, we designed every session because we wanted it to feel relevant to Gecko, not like a generic AI workshop delivered to any company willing to pay for it.

The morning focused on the fundamentals. Why AI matters now. What it’s genuinely good at. Where the risks are. The guardrails. The mistakes. The reality.

Then everyone had to have a go themselves.

People picked a real problem from their own role and spent the afternoon seeing whether AI could actually help.

Nobody was expected to become an expert by 5pm.

We weren’t testing people. We were giving them permission to experiment.

The goal wasn’t to build something incredible. It was to leave with one real problem that AI could either save time, improve quality or unblock.

Something else became obvious quite quickly.

The people who made the biggest leap weren’t necessarily the most technical. They were the people who stayed curious, asked questions, tried something, got it wrong, and tried again.

For a lot of people, AI stopped feeling like this big, abstract thing everyone was talking about and became something practical they could actually use.

Watching that happen across the room was my favourite moment of the day.

What changed?

We surveyed everyone afterwards.

Before the meet-up, around a third of respondents described themselves as either slightly confident or not at all confident experimenting with AI in their work.

Afterwards?

Zero.

Every single respondent moved into either fairly confident or very confident.

94% identified at least one area where AI could genuinely help them in their role.

The meet-up itself scored 9.06 out of 10 - our highest rated and most valuable meet-up to date!

But the statistic that sticks with me is the confidence shift.

Six people arrived with no idea of where to start.

Nobody left feeling that way.

Confidence doesn’t come from telling people they’re capable.

It comes from giving them an environment where they can discover it for themselves.

That’s not really a technology outcome.

That’s a people outcome.

So what made it work?

A few people have asked me that since we got back, and honestly, I don’t think it was the technology...

It was giving people the permission to try.

It was making it okay not to know.

It was showing real examples from colleagues rather than polished vendor demos.

It was creating an environment where people could ask questions, get things wrong, learn from each other and realise they weren’t the only one figuring it out.

And perhaps most importantly, it was making sure nobody felt like they were being left behind while everyone else raced ahead.

AI adoption is often framed as a technology challenge.

I don’t think it is.

I think it’s a confidence challenge.

Confidence grows when people have the time to experiment, the support to ask questions, and an environment where it’s safe not to know the answer yet.

That’s the real lesson from our time together... (and I love it!)