Beyond AI Skills: The Psychology That Turns Learning Into Successful Organisational Change
By Erica Farmer.
We are realising that AI adoption will not succeed through tools and training alone. Human skills are the unlock, but only when the right psychology is in place: agency, curiosity, critical thinking and the safety to experiment and challenge, with it being OK to occasionally get things wrong.
For ABP Members: This is an article overview of learnings shared in the ABP Event: AI Practice Insights – AP3 The Psychology of Learning & Development. Watch the session here.
This article explores why organisations must move beyond simply teaching people how to use AI and create the right human conditions across organisations so that we can collaborate with it purposefully, responsibly and in ways that deliver genuine value to us both personally and professionally.
Firstly, We Must Stop Giving Away Our Power
Many are still talking about AI as though it has marched into our workplaces, grabbed the project plan and started making decisions on our behalf. Open LinkedIn on any given day and you will read protests such as;
‘AI is making us less creative.’
‘AI is taking our jobs.’
‘AI is making people lazy.’
But I do not buy that framing.
Why? Because it lets all of us off the hook. AI is not doing anything; we are. So, the more useful question is not, ‘What is AI doing to us?’ but ‘What are we choosing to do with it and what conditions shape those choices?’
The Pandemic Was the Apprenticeship
The pandemic showed us how quickly behaviour can change when the context changes overnight. In many ways, it was an apprenticeship for the transition we are experiencing now. But this time, organisations are treating AI adoption as a technology rollout: buy the licences, provide some tool training, share a few use cases and wait for productivity to appear.
That is not transformation; it is distribution. And it’s not going to work this time.
Microsoft’s 2026 Work Trend Index makes this gap explicit. Its research found that organisational factors, such as culture, manager support and talent practices, accounted for twice as much of the reported AI impact as individual effort alone. In other words, people may be ready to work differently while the system around them is not.
The Fire Hydrant Meets the Old Pipework
Imagine an employee who previously produced 20 ‘widgets’ a week (widget = any output that someone delivers). With AI, they can now produce 100, and that looks like increased productivity, until the same manager, using the same quality checks and process, must review five times as much work from every team member. It’s like we have attached a fire hydrant to organisational pipework designed for a garden tap.
The predictable result is pressure: more checking, more meetings, more work of uncertain quality and more exhausted managers acting as the organisation’s O-ring. Faster individual production has simply moved the bottleneck downstream.
Psychological Safety Is Infrastructure
AI adoption is full of interpersonal risk, however for success, people need to share a failed prompt, challenge an AI-generated answer, or question a senior leader’s enthusiasm. Amy Edmondson’s research established the relationship between team psychological safety and learning behaviour: people learn when they believe they can take well-intentioned interpersonal risks without punishment. But this is a huge ask right now.
An AI policy can define what is prohibited, but it cannot create the climate; leaders do that through their behavior. They need to use the tools themselves, talk honestly about what failed, invite challenge and distinguish safe experimentation from reckless use. If the message is simultaneously ‘innovate’ and ‘do not get this wrong’, people will either hide their use or avoid it.
The Human Use Cases Are Already Here
We also need to look beyond the official productivity narrative. Harvard Business Review’s 2026 analysis again placed therapy and companionship as the leading reported generative-AI use case. People are not waiting for an organisational strategy before asking AI for emotional support, career advice or help making sense of difficult experiences.
That does not make an AI system a therapist, and it raises serious questions about dependence, privacy, safeguarding and professional boundaries. But dismissing the behaviour misses the signal.
People are looking for accessible, immediate and non-judgmental support ‘in the moment’. Business Psychologists, HR and L&D professionals should be shaping this conversation, not pretending it sits outside their remit.
Agency Must Include Access
And let’s remember, the AI opportunity is not distributed evenly. In Nigeria, where I recently worked with Chartered Institute of Personal Management (CIPM), the World Bank estimated in 2025 that 63% of the population lived below the national poverty line. Digital infrastructure, learning and responsible access to AI could widen routes into education, work and entrepreneurship, but only if inclusion is designed in from the start.
The same principle applies inside organisations. If confident early adopters receive the experimentation time and the opportunities, while others encounter fear, inaccessible training or biased assumptions, AI will amplify the inequalities already present.
We train these systems on human data; we should not be surprised when they reflect human bias, and we should all be accountable for challenging it.
It’s Time to Choose Your Relationship With the Machine
Don’t get me wrong; the task ahead is not persuading everybody to love AI. We need to be clear that AI helps people to retain agency and not lose it. Our assessment of our relationship with it will include our understanding of what AI is best for and our decisions about where it adds value.
My own principle is simple: AI drafts, human crafts. A very personalised and well-trained AI tool can extend our capability, but humans remain responsible for the intention, the standard and the consequence.
Remember, AI is not making us do anything. The choices are still ours, and that makes the quality of our leadership, learning and organisational design more important than ever.
For ABP Members:
Not yet a member? We invite you to join us at The Association for Business Psychology!
About the Author
Erica Farmer is a professional speaker, trainer and author specialising in AI, future skills and the future of work. With more than 25 years’ experience across learning, talent and organisational development, she helps leaders and people professionals turn AI adoption into practical, human-centred change. She is the author of ‘AI for People Professionals’ and co-founder of Quantum Rise Talent Group and founder of EricaFarmer.AI Ltd.
AI collaboration statement: This article was authored and human-crafted by Erica Farmer, in collaboration with ChatGPT Work to support transcription and editorial refinement.
References
Edmondson, A. C. (1999). Psychological Safety and Learning Behavior in Work Teams. Administrative Science Quarterly, 44(2), 350–383. https://doi.org/10.2307/2666999
Microsoft (2026). Agents, human agency, and the opportunity for every organization: 2026 Work Trend Index Annual Report. https://www.microsoft.com/en-us/worklab/work-trend-index/agents-human-agency-and-the-opportunity-for-every-organization
World Bank (2026). Nigeria: Country Overview. https://www.worldbank.org/ext/en/country/nigeria
Zao-Sanders, M. (2026). How People Are Really Using AI in 2026. Harvard Business Review. https://hbr.org/2026/06/how-people-are-really-using-ai-in-2026