Building an adaptability muscle: the foundation for the AI era

As leaders, we often focus on managing technical debt. But in an era of constant change, the most dangerous liability we carry isn't in our legacy code - it's in our legacy ways of working.

We've seen how this manifests as both a technical liability as ‘digital geology’ and a cultural one as ‘leadership debt’.

The way we build and operate technology has undergone a seismic shift in the last two decades. A relentless cycle of new technologies demands our attention, forcing us to adapt how we design, build and operate. The internet era forced a fundamental rethink, introducing concepts such as agile, multidisciplinary teams, and DevOps. For many established organisations, this is a difficult, foundational shift that is still in progress.

While the methods have changed over time - from waterfall to agile, from on-premise to cloud - one factor has remained constant: the ever increasing velocity of change.

Today, the emergence of generative AI represents not just another incremental step, but a profound acceleration, moving at a speed that makes the internet era feel slow. The landscape of tools, capabilities and customer expectations is shifting daily. It’s natural to feel that just as you’re learning the rules of one game, another has already begun.

This creates a fundamental question for leaders: how do you prepare for a future that is arriving faster than ever? The answer is not to bet on a single technology, but to build an institutional “adaptability muscle” - the organisation’s enduring capacity to adapt to relentless change.

This isn't the first time our industry has faced a crisis of speed. The blueprint for managing this new velocity of change was written over the last decade, and its lessons are more urgent today than ever.

In the early 2000s, the technology operating model in most large organisations was defined by conflict. Development teams were incentivised to create new features, while Operations teams, focused entirely on stability, fostered a defensive culture of “No, No Culture”. This tension resulted in a technical ‘brittleness’, where every small change risked breaking the entire system. The natural outcome was friction, slow release cycles, and a culture of blame. It was a model that could not keep pace with the demands of a connected world.

The response was a fundamental shift in mindset: DevOps. This was not merely a technical fix, but a cultural one. It brought teams together around a shared purpose: delivering value to users. This new approach broke down silos and introduced more fluid ways of working, such as Continuous Integration and Continuous Delivery (CI/CD), enabled by the automation of cloud computing.

Internet-era giants like Google and Amazon, and even pioneering government teams like the UK's Government Digital Service, mastered this approach. They built multidisciplinary teams that could deliver improvements not in months, but in days or even hours.

Their success didn’t come from a specific tool, or a better technology, but from having a better operating model to manage change. They mastered the underlying principles of delivering value for users: creating multidisciplinary teams, building automated pathways for safe deployment, and fostering a culture of shared ownership. It is these principles, not the specific tools of the past, that form the blueprint for adaptability today.

An adaptable organisation isn’t just faster - it’s more resilient, with the ability to react effectively when things go wrong, regardless of the cause. The COVID-19 pandemic provided a stark, real-world test. Organisations with well-developed adaptability were able to pivot in weeks, scaling digital services and supporting remote teams because their culture and technology were already built for change. Those with rigid structures struggled.

This resilience is also critical for navigating the fragility of our modern technology ecosystem. Often, the greatest risk comes not from external threats, but from within. A single, flawed software update can cascade through interconnected systems, causing outages which are just as damaging as a malicious attack.

This is where the ability to react at speed becomes a vital defensive capability. An adaptable organisation can quickly diagnose, patch and deploy a fix, whether the vulnerability is from a cyber attack or an internal error. This capacity to recover quickly is a direct outcome of a culture that values collaboration and invests in automation - the evolution from DevOps to DevSecOps. In the modern era, resilience against all forms of disruption is a vital part of effective risk management.

Beyond resilience, this operational fitness gives organisations a profound competitive edge. While non-adaptive competitors are encumbered by their legacy systems and processes, an adaptable organisation can pivot when the market shifts, seizing new opportunities. They can adopt more efficient technologies as they become available, replacing legacy components without massive disruption or multi-year programmes.

They can, in short, operate "by design, not by default," continuously choosing the best path forward rather than being dictated to by the limitations of their past decisions.

If adaptability was an advantage in the internet era, it has become non-negotiable in the AI era. AI tools allow teams to build and test prototypes in minutes, not weeks. New services and models with new capabilities are appearing daily. It is impossible to predict which of these will prove most valuable.

Chasing every new trend is a recipe for burnout and wasted investment. As we have argued before, simply plugging new technology into an old operating model leads to disappointment.

The velocity of change is now too fast for long, drawn-out analysis. To thrive, organisations must adopt a prototyping mindset at scale. This demands a commitment to continuous learning, where teams are not just encouraged to try new things, but are also empowered to unlearn the old habits and processes that no longer serve them in this new context.

Where an agile mindset helps a team respond to changing customer needs, an experimental culture allows them to test ten different AI approaches to find what delivers real value. Where a collaborative structure helps launch a new digital feature, it also enables the organisation to safely explore and scale opportunities from new AI prototypes.

The focus must shift from finding the "perfect" AI strategy to building the core muscle - the people, processes, and platforms - that enables safe and rapid experimentation.

Developing this adaptability muscle is the most critical work for leaders today. It shifts the focus away from reacting to each new wave of technology, and towards the deliberate work of building a resilient foundation for the future. It is the direct investment you need to make for whatever comes next.

This begins with asking honest questions, and then applying a test and learn mindset to answer them:

• Where does our current structure create friction and slow down good ideas? A test and learn approach means starting small: instead of launching a large, high-risk programme, identify a single, meaningful challenge and focus the first effort entirely on a clear learning objective.

• Are we empowering our teams to solve problems, or preserving a culture of top-down control? True empowerment comes from decentralised decision-making. Give a small, multidisciplinary team the autonomy and context to solve the problem, trusting them to make choices quickly without needing to escalate through a complex hierarchy.

• How can we create safe spaces to experiment and learn? This requires a genuine commitment to psychological safety. Leaders must create an environment where a failed experiment is treated as a successful lesson learned, not a punishable mistake, ensuring that valuable data is never hidden for fear of blame.

The most practical first step towards building adaptability is to start with an exemplar project and take a test and learn approach to solving it. Public Digital’s book on Adopting the Practice of Test and Learn - the second in our series of three - explains how to take that practical first step.

For any organisation, building this foundational adaptability is the most effective way to prepare for the future. Your competitors who possess this capability will build faster, weather risks more easily, and leapfrog those who don't.

In the AI era, building this muscle is what transforms the daunting pressure of constant change into a sustainable, competitive advantage.