Jennie Marshall argues that while organisations are investing heavily in AI, too few are turning pilots into meaningful impact. The real barrier is not technology but leadership, culture and capability. To close the gap, leaders must build confidence, embed AI into workflows and measure adoption, not just expenditure at scale.
A while back I was heading north to visit a client, coffee in hand, when I put on the Science & Tech Daily podcast. Piers Linney was talking about AI, and one line made me turn the volume up: “Organisations spend a lot… but don’t follow through.”
I’ve been thinking about that ever since. Because listening to him, I realised something I already see in the clients I work with; organisations are investing heavily in AI, piloting all sorts of tools, yet adoption and meaningful impact are lagging behind.
Will all the money spent deliver little more than a headline?
And that isn’t a technology problem. It’s a people problem. A mindset problem. A leadership problem. 2026 feels like a pivotal year. AI is no longer coming — it’s already here. The question is whether organisations will finally make it work, or will all the money spent deliver little more than a headline?
The gap between investment and adoption
Across industries, the story is consistent. Businesses are pouring money into AI technologies. IDC reported global AI investment grew by 38% in 2025. Yet, McKinsey’s AI report indicates fewer than 22% of organisations have successfully implemented AI at scale.
That is a huge gap between investment and real-world adoption. Tools and platforms are available, but they are rarely embedded into workflows in ways that make a tangible difference.
I’ve seen it in action. One client invested in an AI-driven customer service platform. The tool was impressive on paper: the potential for automation, faster response times, even predictive insights was exciting. But in practice? Staff weren’t confident using it, managers weren’t sure how to measure success, and the pilot ended up largely ignored. Money spent. Effort expended. Impact minimal.
This is exactly the pattern Piers Linney was talking about. Organisations are experimenting, yes, but they’re not thinking strategically about how to make it stick.
Mindset is the missing ingredient
What stands out to me, and what I’ve seen in many organisations, is how often the focus is on technology rather than people. Leaders treat AI like a short-term project: IT-led, siloed, “plug it in and hope it works.” Meanwhile, employees are nervous. They ask themselves:
- Will this replace me?
- Do I have the skills to keep up?
- How will this change my workday?
Without addressing those questions, adoption stalls. You can have the most sophisticated platform in the world, but if people don’t understand how it helps them, or don’t feel confident using it, the investment won’t deliver.
Piers Linney’s advice was simple but powerful: “Think big, start small, and take people on the journey.” That “small” part is crucial. You can’t expect wholesale adoption overnight. You can, however, start in a way that builds confidence and gradually demonstrates value.
Four leadership imperatives
Leaders have to step up and in my experience there are four areas that will determine whether organisations succeed:
1. Build organisational capability
Training is not optional. Leaders and managers need to understand AI in practical terms: what it does, what it doesn’t do, and how it can augment human work.
This isn’t about teaching everyone to code. It’s about helping people:
- Understand the potential and limitations of AI tools
- Use AI responsibly to support decision-making
- Ask the right questions about outputs and implications
Capability development should be ongoing. Workshops, mentoring, and practical exercises are far more effective than one-off training sessions.
Example: In one client session, we ran a series of scenario-based exercises showing how AI might assist with scheduling, prioritising tasks, and automating reporting. Managers were nervous at first, but by the end, they could see the benefit and felt confident in trialling AI tools in real-world settings.
2. Embed AI across workflows, not departments
AI cannot sit in a silo. A tool in one team, no matter how advanced, will not transform the organisation. Leaders need to plan adoption across functions, integrate AI into workflows, and clarify roles and responsibilities.
A simple way to do this is to start with pilot projects that cross departments, then share learnings widely. Success breeds confidence. It also demonstrates that AI is not “just IT’s problem” — it’s everyone’s.
Tip: Identify a small, high-impact workflow that can benefit from AI, test, refine, and then scale. Track outcomes carefully.
3. Shift culture and mindset
The most successful AI adoption isn’t about technology. It’s about culture. Staff must feel safe experimenting, making mistakes, and asking questions. They need to understand that AI is here to augment their work, not replace them.
Psychological safety is crucial. Leaders should model curiosity and transparency, acknowledging uncertainty while showing commitment to learning.
Example: One team we worked with adopted an AI tool for project reporting. Initially, employees were worried it might replace them. Leaders held open forums, encouraged questions, and highlighted early wins. Adoption increased steadily, and the team started seeing tangible time savings.
4. Measure adoption, not just investment
Spending alone is a poor measure of success. KPIs should focus on:
- Workflow integration
- Staff confidence and capability
- Business outcomes (efficiency, quality, decision-making)
Measurement isn’t about micromanaging employees. It’s about ensuring the organisation sees a return on investment in people and processes as well as technology.
Ethical and responsible AI adoption
AI brings opportunities and responsibilities. Bias, governance, and transparency cannot be afterthoughts.
Leaders need to:
- Communicate ethical boundaries clearly
- Ensure AI outputs are explainable and accountable
- Align AI adoption with organisational values
Responsible adoption builds trust, which in turn accelerates confidence and integration.
The strategic opportunity
Done well, AI can be transformative. It can:
- Free staff from mundane, repetitive tasks
- Enhance decision-making with data-driven insights
- Enable agility, creativity, and innovation
But organisations that fail will not fail because the technology doesn’t work. They will fail because they didn’t focus on the human factors: confidence, clarity, culture, and capability.
Piers Linney summed it up: “You have to play where the puck is going to be.”
The organisations that thrive will be those that embrace AI as a strategic, people-centric capability, not a bolt-on system.
Remember, it’s not all about the tech
This year especially AI investment is rising, pilots are everywhere, and the tools are sophisticated. But the difference between success and disappointment will be leadership, mindset, and organisational readiness.
AI adoption isn’t a technology challenge. It’s a people challenge. The organisations that focus on capability, culture, and responsible integration will see returns. Those that don’t will watch the money spent deliver little impact.
To make this really work, leaders need to ask themselves:
- Are my teams ready to use AI confidently?
- Have I built the mindset and culture to support adoption?
- Are we measuring outcomes, not just expenditure?
The payoff is clear: AI that works with people, not against them, can unlock the true potential of AI for organisations and people — if leaders take the time to prepare their organisations.
Jennie Marshall is Founder and Learning Alchemist at Wren Learning Consultancy
