Why AI initiatives fail when leaders treat them like typical tech rollouts.
In companies around the world, leaders are launching AI-first transformations using the playbooks they know. They create urgency, train staff and celebrate early wins. This, after all, is a model that’s worked in the past, such as during the digital transition.
But this time, real adoption stalls. Employees nod along in meetings, complete the training – and quietly return to old habits. Why? Because this transformation is different.
AI’s general-purpose nature means it touches every part of the organisation, but there’s no single roadmap to follow because the right application depends on context, creativity and role-specific reinvention. This makes AI transformation nothing less than a wide-scale challenge to rethink value creation on the individual and collective level.
What’s more, AI doesn’t just alter the “how” of work; it threatens the “who”. Just look at the relentless headlines about AI replacing humans across a wide swathe of professions, from writers and designers to doctors and lawyers. It’s no surprise that people feel overwhelmed. And when people can no longer clearly see what is left for them, they freeze.
A framework for reinvention
In a nutshell, AI transformation is about identity. It asks: What part of your thinking can you give to the machine? What does that free you up to do? What do you want your work to become?
These questions must be answered across the organisation, with each employee participating in the reinvention of their role, their function and their contribution. To steer this change, leaders need to create the psychological and organisational conditions that allow people to embrace experimentation and reimagine their professional identity.
Here are five steps for creating those conditions:
Set the right foundations
Organisations must invest in large-scale, tailored learning programmes to build a shared understanding of what AI is, what it can and cannot do, and how it might be used responsibly. These programmes must be tailored by role, seniority and function, but everyone should ultimately speak the same language. Once people understand the technology, they will begin to ask questions about how to use it across different parts of the organisation.
For example, Company A (a multi-billion-dollar, Europe-based consumer packaged goods company) held a full-day immersion for both non-executive and executive board members, designed to establish a common language about AI and showcase concrete business cases already in use. The senior leadership team then went through a multi-day training focused on how AI works, how to manage its risks and how to think about its broader organisational impact.
At the project level, participants explored how AI could resolve specific domain or operational challenges. Finally, the company launched general online training available to all employees, giving everyone access to foundational knowledge. This multi-tiered approach created the shared understanding necessary to move from abstract hype to grounded, role-relevant inquiry.
Mitigate the fear of change
Education brings clarity about AI’s true capabilities, but this can increase fear. To counter this, leaders must clearly articulate the purpose behind AI adoption. It’s not about slashing headcount or chasing short-term results – it’s about enabling the organisation to meet future growth, scale and innovation goals.
Here, leaders must project optimism and evolutionary thinking. They must affirm that AI is a strategic tool to empower the workforce, not undermine it. When people see AI as something that helps the company grow while investing in its people, they are far more likely to engage. They no longer feel the need to defend their role, but to help redefine it.
At Company B (another large Europe-headquartered CPG company), leadership addressed employees’ anxiety about AI by establishing an “AI board” tasked with guiding and communicating the company’s strategy. This group’s remit was threefold: First, to define what an AI strategy meant for the business and ensure alignment with overall company goals; second, to create visibility across all AI initiatives and determine which to accelerate, start or pause; and third, to proactively manage communications around AI to foster transparency and trust.
By anchoring the transformation in visible and strategic leadership, the company helped employees see AI not as a looming threat, but as a thoughtful evolution.
Company C (a large consulting firm headquartered in the United States) set ambitious growth targets that could not be met simply by hiring more people. Leaders were transparent about it: future success would depend on employees becoming more efficient, and AI would be key to achieving that. By clearly communicating this vision, leadership made AI adoption part of a shared goal – something employees were contributing to and not being threatened by.
Raise the bar of human potential
If AI takes over tasks we once thought of as deeply human, then we must redefine what humans are here to do. Across science and industry, AI is helping solve problems of previously unimaginable complexity. DeepMind is just one example of companies that have pioneered the use of AI in fields where human beings didn’t stand a chance.
Businesses must aim higher, too. Rather than defaulting to workforce reductions for efficiency, we should challenge teams to use AI to solve problems in novel ways. This is the most difficult part of the change process because it goes into uncharted territory. At Company B, one of the most complex initiatives was the rollout of a new enterprise system, an effort that would normally require months of validation by hundreds of employees running thousands of simulated workflows. Rather than defaulting to that traditional approach, senior leaders posed a challenge: “Could AI help us do this in a smarter way?”
The question sparked a wave of experimentation led by employees themselves. They explored ways to use AI to simulate test cases, identify anomalies and prioritise validation efforts. The result was faster implementation plus a shift in mindset: AI wasn’t replacing people; it was amplifying their ability to solve difficult problems.
Sponsor augmented intelligence
AI excels at pattern recognition, simulation and generating options. Humans excel at intuition, judgment and identifying the right problems to solve. Augmented intelligence is about designing workflows that allow people to offload parts of their cognition to AI and focus on what only humans can do.
At Company B, leaders explored where AI could support decisions while preserving human oversight, starting from the boardroom. This rare experimentation showed promising early results in the form of spot-on decisions at price and distribution level. It demonstrates how human and machine intelligence can combine to create value.
Stay one step ahead
AI’s evolution is exponential, and waiting for the technology to stabilise before acting is a high-risk strategy. Organisations must build cultures that embrace ongoing change, experimentation and ambiguity. Leaders must create safety for disruption, which requires bold leadership.
JPMorgan embraced this philosophy early by exploring applications of generative AI well before the 2022 release of ChatGPT to the public. Rather than wait for the technology to mature, the company’s executive leadership encouraged internal experimentation and launched AI pilots across key business units. This approach accelerated capability-building and positioned JPMorgan as a front-runner in strategically leveraging AI at scale.
Reinvention, not rollout
Rather than a rollout, AI transformation is an invitation to reimagine. This kind of change must be led with curiosity, safety and co-creation. Because the most important question in AI transformation is: “What do you want your work to become?”
