Why AI Adoption May Actually Be an AI Confidence Problem
The word underneath all the buzzwords
For years the corporate world has talked about curiosity, continuous learning and psychological safety. Underneath all three sits a quieter word: confidence.
Confidence grows in a space that feels psychologically safe. It breaks down the moment that safety disappears. And without confidence, there is no curiosity. Without curiosity, there is no continuous learning.
That is why I believe many organisations do not have an AI adoption problem. They have an AI confidence problem.
No AI shaming, in either direction
One message I share with every audience is simple: no AI shaming.
When colleagues tease or banter about the people who are not using AI yet, those people feel more intimidated. They pull back further, and adoption slows down.
The opposite is just as damaging. When someone uses AI enthusiastically and gets ridiculed for "letting the robot do the work", their confidence takes a knock too.
Both kinds of shaming remove psychological safety. Once that safety is gone, curiosity switches off, and learning stops with it. These things are delicately intertwined, so leaders need to protect all of them at once.
Level the playing field with a shared vocabulary
In my training rooms I see the full spectrum. Some people were born "BC", before computers. Others grew up with a computer on their bedside table. Some are completely at home with tech, while others are still finding their way around Word and Excel.
That is why I named one of my workshops the AI Confidence Booster. Everyone deserves to start on the same page, or at least be brought onto it.
The fastest way to do that is to give people a vocabulary. AI comes with a whole new language. Until you can speak a little of it, you will always feel slightly unsure. So I start with the basics:
- LLM stands for large language model, the technology behind tools like ChatGPT, Claude, Copilot and Gemini.
- GPT stands for generative pre-trained transformer. Generative means it creates new content. Pre-trained means it learned from huge amounts of text first. Transformer is the type of design that made this leap possible.
- What changed? AI has been quietly working in the background for decades, in spam filters, bank fraud checks and online recommendations. What changed is that we can now simply talk to it in everyday language, and anyone can use it.
Once people own the words, the fear starts to fade.
"If Mel can, I can"
I am not what most people expect an AI facilitator to look like. I am self-taught, non-technical, female and over 50.
Tech has long been dominated by men. And as we get older, most of us become more cautious and less exploratory. Yet I embraced AI early, taught myself, and then started teaching others.
I have noticed that this puts people at ease. They look at me and think, "If Mel can, I can." That moment of recognition is where confidence begins.
AI adoption through the ADKAR change model
AI adoption is a change management challenge, not just a technology rollout. My favourite lens for this is the ADKAR model, developed by Prosci. It describes the five steps every individual moves through when change sticks:
- Awareness: People understand why AI matters for their role and their organisation. Plain-language explanations do the heavy lifting here.
- Desire: People want to take part. Desire grows when there is no AI shaming and people feel safe to try.
- Knowledge: People know how to use AI. This is where vocabulary and hands-on demonstrations come in.
- Ability: People can actually apply AI in their daily work. Practice in a supportive space turns knowledge into skill.
- Reinforcement: New habits are celebrated, shared and repeated, so people do not slide back into old ways.
Confidence runs through every one of these stages. Skip it, and people stall somewhere between knowing and doing.
Build confidence first, and adoption follows
If AI adoption in your organisation feels slow, look beneath the surface. Ask whether people feel safe to experiment, whether they share a common language, and whether anyone is being shamed for using AI too little or too much.
At AI with Mel, a division of Performance Booster, I facilitate practical AI training and AI change management workshops for leaders and teams. My AI Confidence Booster workshop brings everyone in the room onto the same page, whatever their starting point.
Ready to boost your team's AI confidence? Get in touch via www.aiwithmel.co.za to book a workshop or a conversation.
Frequently asked questions
Why do AI adoption projects stall?
Often it is not the technology. People lack the confidence to experiment, usually because they do not feel psychologically safe or do not yet understand the basic language of AI.
What is AI shaming?
AI shaming is teasing, ridiculing or judging colleagues about their AI use. That includes mocking people who avoid AI and people who use it a lot. Both undermine psychological safety.
How does the ADKAR model apply to AI adoption?
ADKAR (Awareness, Desire, Knowledge, Ability, Reinforcement) maps the steps each person moves through to adopt AI. It helps leaders see where individuals are stuck and what support they need.
Who is Mel?
Mel is a self-taught, non-technical AI facilitator and business coach in South Africa. She runs AI with Mel, part of Performance Booster (Pty) Ltd, and helps leaders and organisations adopt AI safely, confidently and practically.