How to Lead Through AI’s Hype Cycle

Because this isn’t the first time a technology promised everything and delivered confusion first.

We’ve been here before. And in my own career, I’ve felt it first-hand. A radical new technology appears, promising to change everything. Hype builds. Expectations soar. Organisations feel pressure to act – fast. Some go big early. Some freeze. And a few take a more measured path.

AI is clearly transformational. But how can leaders respond in ways that create momentum, not waste? To answer that, it helps to look back.

I started my career in the late 1990s, building some of the earliest websites in telecoms before joining a dot-com start-up. It was a front-row seat to the hype and hope of the early internet. Later, I led digital transformation in major organisations, and saw similar patterns play out on a bigger stage.

This isn’t the first time business and society have had to adapt to a disruptive, poorly understood innovation. It’s a well-worn pattern – Gartner described it as the 'hype cycle': a surge of excitement, a rush to act, disillusionment when reality bites, and eventual progress when substance catches up to story.

What history tells us about tech adoption

It’s a cycle that’s played out again and again. When Gutenberg introduced the printing press in the 15th century, it was clearly revolutionary, but its short-term impact was messy. Authorities tried to suppress it. Misinformation spread. It took decades before its full power reshaped literacy, communication and the spread of knowledge as we know it.

The same pattern has played out more recently with two transformations that reshaped the way we live and work – first the internet, then the wave of digital transformation that dominated the 2010s.

From dot.com bust to digital core

In the 1990s, the internet was widely expected to transform everything – and ultimately, it did. But the transformation wasn’t linear. It came in waves. Businesses launched brochureware sites and online portals with little clarity on their purpose or value. Start-ups chased growth ahead of business models. Organisations treated the internet as a bolt-on rather than a rethink.

When the bubble burst, it exposed the fragility underneath the hype. But in time, the technology matured, behaviours shifted, and the internet became the infrastructure for modern life.

👉 Transformational technologies require patience, discipline, and a willingness to rethink everything.

Digital transformation – easier said than done

In the 2010s, digital transformation became the boardroom mandate of choice. It was often framed as urgent and existential: evolve digitally or be disrupted. The intent was right – but the execution frequently fell short.

Rather than rethinking how value was created and delivered, many organisations defaulted to launching large-scale programmes. These were typically tech-heavy, top-down, and insufficiently grounded in user needs. Process redesign lagged. Cultural change was an afterthought. And the promise of agility was undermined by rigid governance and vague goals.

The result? Too many transformations delivered more change than improvement.

👉 Big shifts without the right foundations can drain momentum – and trust.

Navigating the AI noise

AI is in its early, noisy phase. There’s real promise – but also hype, uncertainty, and risk. Some organisations are rushing to announce sweeping AI strategies. Others are frozen, unsure where to begin.

The smart ones are doing neither. They’re starting small. They’re learning fast. They’re applying AI like a product: testing, refining, scaling only what works.

One example? Johnson & Johnson. Back in 2022, they took an open and experimental approach to GenAI, encouraging teams across the organisation to explore its potential. This generated nearly 900 use cases.

Two years on, they’ve learned what works and what doesn’t. They’ve narrowed focus to the top 10–15% of those initiatives, which are now delivering around 80% of the value. It’s a shift from broad experimentation to operational focus, grounded in evidence and clarity.

That’s the approach I’ve always believed in: product working.

Building learning into your operating model with product working

When we say "product working," we’re not just talking about agile delivery or building MVPs. We mean structuring the entire operating model – how work gets done, who does it, how it's funded and governed – around the idea that not everything will work first time.

Value emerges through iteration, not upfront planning. The teams closest to the problem are best placed to test what works. And progress doesn’t come from certainty. It comes from continuous, deliberate learning.

It’s a mindset – and a system. A way of organising work, decision-making and delivery around continuous learning. And that system is exactly what’s needed to make AI useful.

Finding your path forward

If you lead people or shape how work gets done, the pressure to “do something” about AI is probably mounting. It’s easy to feel behind, to think others are further ahead. But most organisations are still at the beginning – and that’s not a problem. It’s an opportunity.

The printing press took a century to reshape the world. The internet took decades. AI will move much faster – but we still have time to get our approach right.

As one CTO put it:

“If anyone tells you they have AI figured out, they’re drunk, lying, stupid – or all three.”

That kind of humility may be the most valuable mindset leaders can adopt right now.

This isn’t about being first. It’s about being ready. Ready with the governance, capability and learning systems to adapt as the technology evolves.

By approaching AI with a product mindset – testing, learning and iterating – you avoid the paralysis of inaction and the waste of rushing in too fast. You build what works, not just what’s possible.

And you give your organisation the one advantage that really matters: the ability to learn its way into the future.

#AI #OperatingModel #ProductLed #AITransformation #OrganisationDesign

About Kindred

Kindred helps organisations design the structures, behaviours and ways of working that enable teams to move faster, learn smarter, and deliver what matters.

We bring a product mindset to change - working with clients to test, adapt and scale what works, rather than chasing one-size-fits-all solutions. Our tools and support build internal capability, create shared language, and embed the rhythms of continuous learning.

If this sounds good to you, let's talk!