AI adoption doesn’t just happen

AI technology is evolving faster than you can say “generative pre-trained transformer”, driving a wedge between the early adopters and everyone else. The gap will only widen. You can try to put the brakes on – as the new EU regulations threaten to do – but the reality is that we just need to get a grip and manage the change: deal with the risks, exploit the opportunities, help your organisation use AI to go faster and do better. Or you will lose out to someone who can.

Helpfully, change management is not a new or mysterious art. It’s everything you have ever learned about how to land something in a way that brings people with you. It’s as much about what not to do as it is about what you should do. And – particularly if you’re British – pointing out what everyone is doing wrong is second nature. So we should all know how to inspire and encourage people to unlock all the amazing benefits of AI, right?

Ha! Hell, no.

AI already has an image problem

The binary “AI: yes, or no?” debate does little to advance our understanding. More problematically, there seems to have been an early decision to position AI firmly in the creepiest corner of the sci fi domain. First off, it looks like this fella, when companies are trying to brag about how great it is.

Better yet, here’s what happens if you ask ChatGPT to generate a word cloud based on descriptions of AI in movies:

So we can agree AI has some reputational damage to unpick, before we even get to overcoming a generalised fear of losing our jobs to machines.

What to do, what to do…

One of my first proper jobs was working for Prof. Sir Michael Barber in the first of many incarnations of the UK government’s Prime Minister’s Delivery Unit. It was a treat that the 25-year-old me enjoyed enormously, and grown-up me appreciates even more with the benefit of hindsight. We were all given a copy of John Kotter’s Leading Change as a welcome gift. It’s the kind of book that is at once enlightening and completely obvious, and that’s what makes it great.

Engine recently dispatched one of our own along to Gartner’s session on AI adoption and change and they reported back. Serendipitous spoiler alert: all the basic rules of good change management still apply. Excellent! So let's break this down, Kotter style.

Plus ça change, plus c'est la même chose*

Step 1: Create a sense of urgency (not an existential threat to humanity)

John Kotter's first step in leading change is to create a sense of urgency. But you can take this too literally by painting AI as the harbinger of an apocalyptic future (cue the dramatic music and ominous warnings). Instead, communicate the real benefits and challenges of AI adoption. Explain how it can improve work, inform decision-making, and help with the drudgery of your job so you can focus on work where you add most value.

Step 2: Form a powerful coalition (not a boys' club)

AI is for everyone, not just a select group of, let’s be honest, mainly men in dark, windowless rooms. They will call your AI assistant Skynet for lols and wonder why it’s off-putting. Put together a diverse team that includes people from different departments, genders, and backgrounds. This coalition will help make sure that AI adoption is inclusive and speaks to the needs of the entire organisation.

Step 3: Create a vision for change (not a sci-fi script)

Your vision for AI adoption should be clear and practical, not something out of a sci-fi movie. Avoid jargon and tons of technical language. Instead, focus on how AI can solve real problems, like reducing mundane tasks or providing better customer insights. Make it relatable. No one will learn how to use their AI assistant to save time and summarise actions if you focus on creating questionable deep fake videos.

Step 4: Communicate the vision (without the scaremongering)

Communication is key, but scaring people into compliance is not the way to go unless compliance is all you want. You don’t: you want innovation and high performing teams. Instead of telling employees that AI will replace them, show how it can augment their skills and free them up for more motivating work. Use examples and success stories.

Step 5: Remove obstacles (and no, AI isn’t one of them)

Identify and remove barriers to AI adoption. This might mean providing training, addressing data privacy and environmental concerns head on, or updating outdated systems. The goal is to make the transition as smooth as possible, not to create more hurdles.

Step 6: Create short-term wins (celebrate, don't segregate)

Celebrate small victories along the way. When a department successfully integrates AI into its workflow, acknowledge it. This builds momentum and shows that AI adoption is not only feasible but also beneficial. And please, make sure everyone gets to join the celebration — no exclusive iRobot-themed parties.

Step 7: Build on the change (not just for the enthusiasts)

Once you have some wins under your belt, build on them. Expand AI adoption to more areas of the organisation. Keep the momentum going by continually improving and adapting your AI strategy. Remember, AI is a tool, not an end in itself. Integrate it where it adds value.

Step 8: Anchor the changes (and keep it real)

Finally, make AI adoption part of your organisational culture. This means ongoing training, regular updates, and continuous feedback. Keep putting people first. AI should enhance human capabilities, not replace us. It doesn’t have to be so big and noisy it freaks people out – and we definitely don’t need to see tech billionaires jumping up and down on a stage. No dry ice. Just keep it in the frame, keep the Teams chat going about new discoveries, have the odd lunch and learn.

A poster? Go on, then

We made a little infographic to help you out – with huge credit to Kotter for literally writing the book on how to lead change, and to our private Engine GPT for helping analyse the steps and relating them to AI. If Leading Change somehow passed you by thus far, give it a read.

You’re welcome.

* The more things change, the more they stay the same. The phrase offers a somewhat cynical world view, suggesting that despite apparent transformations, fundamental things remain unaltered. That’s probably why we prefer to use the French version, so we can imagine ourselves as a jaded flaneur (if you don’t know, you can Google that one and add it to your life goals), seated at a Parisian café, taking a long drag on a Gauloise between sips of Absinthe and (add your own French cliché to the list). But at Engine we just think things are changing while the fundamental rules of change still apply. We love change. That’s why we’re more likely to be found in a wework space, eating sandwiches.

We build private AI tools that let you keep things simple; private GPTs trained on content you curate, giving you answers your teams can rely on, and starting you on an AI change programme you can manage. We charge a one-off, affordable build cost and a single, fixed monthly cost for your whole organisation. Email us at info@engine-ai.co.uk.

‍