Reimagining Value Delivery: From Agile Legacy to Authentic AI Operating Models 

At a glance:

This article explores how AI challenges the foundations of legacy Agile models and proposes a reimagined operating model—one that is fluid, data-driven, and authentically AI-enabled. 

Over the past two decades, Agile has transformed the way organizations deliver value. It replaced rigidity and traditional waterfall execution with empowered teams, iterative development, and a customer-centric approach. Agile operating models became the blueprint for modern software delivery, characterized by end-to-end ownership, decentralized decision-making, and multidisciplinary collaboration. But as Artificial Intelligence (AI) becomes a central force in shaping what we do, we must ask:Are our Agile operating models Agile enough? 

The Legacy of Agile: A Stable Heartbeat 

Agile was born out of the need to navigate complexity and uncertainty in knowledge work. It introduced values and principles that emphasize adaptability, transparency, and continuous improvement. Over time, these were codified into operating models that emphasized stable teams, prioritized workstreams, and technical competence as the driving force behind Agile product delivery. Governance was managed through portfolio management and planning cadences, creating a rhythm—a “stable heartbeat”—that allowed organizations to plan, execute, and improve predictably (see Figure 1; typical pattern of organization). 

This rhythm brought structure and predictability, but it also introduced a sense of inertia. Teams labeled as “stable” became fixed, workstreams were siloed, and Agility was reduced to what the system could technically deliver based on predefined priorities, which we fed the system. The very models designed to grow adaptability began to resist change themselves. 

AI Disrupts the Rhythm 

AI doesn’t operate on a stable heartbeat. It thrives on fluidity, real-time responsiveness, and data-centric decision-making. It introduces agents that can suggest optimizations based on live data, continuously re-prioritize tasks using value models and KPIs, match expertise dynamically to emerging needs, and signal impediments with proposed resolutions (see Figure 2; AI-based operating model). 

In this new paradigm, Agility is no longer confined to periodically resetting a backlog—it permeates the entire system. The delivery model becomes adaptive, responsive, and continuously evolving itself. AI doesn’t just accelerate delivery; it transforms how we think about value creation. 

From Stable Teams to Dynamic Task Teams 

One of the most radical shifts in AI-enabled delivery is the move from stable teams to dynamic task teams. These are temporary, purpose-driven formations of experts assembled to complete specific tasks or opportunities of value. They are matched based on availability, expertise, and the nature of the task at hand. 

This model challenges the traditional belief that team stability drives performance. Instead, it emphasizes ‘continuous’ re-teaming to meet evolving needs, shared policies, and foundational ways of working to reduce collaboration friction and costs of re-teaming, as well as full-stack expertise to minimize idle time and bottlenecks. 

While this approach introduces complexity, it also unlocks Agility at scale. It allows organizations to respond to opportunities and challenges with precision and speed. The system becomes capable of reconfiguring itself in real time, guided by data and strategic intent. 

The Role of Agents and Humans in the Loop 

AI agents play a central role in this operating model (see Figure 3: Agents and Their Purpose). They perform tasks such as suggesting improvements based on product usage and performance data, prioritizing actions using value models and goals, identifying required expertise and assembling capable teams, and detecting impediments with proposed resolutions. 

However, AI is not infallible. It can be biased, make errors, and misinterpret context. That’s why humans remain essential. Experts are embedded in the loop to review and refine AI-generated suggestions, validate team compositions and staffing decisions, and lead resolution of impediments (see Figure 4: Humans in the loop). 

This human-AI collaboration ensures quality, accountability, and ethical decision-making. It preserves the human edge while leveraging the computational power of AI. 

Benefits of AI-Driven Delivery Models 

The shift to AI-enhanced operating models offers several compelling benefits. Decisions are driven by real-time data, not delayed consensus. This means that a significant amount of meeting and discussion time could potentially be saved when algorithms replace opinion-based prioritization with evidence-based action. Continuous alignment with strategic outcomes becomes possible, and enhancements are based on actual performance rather than assumptions and hunches. 

Impediments are surfaced and addressed proactively, reducing waste and inefficiency. Standardized foundational practices minimize rework, and teams are formed based on actual requirements rather than legacy roles. Agility becomes systemic—not just an attribute of a select few or only a shared mindset, but a property of the entire organization. 

What About Projects and Innovation? 

When you look closely at the inputs in Figure 2, it does not mention anything about larger, more substantial innovations and projects. This is because I believe that this work should be scrutinized, and if and when it’s decided to pursue something significant, it should stand apart from the rest of the work. In this new model, projects evolve into experiments. They are shortened, hypothesis-driven product development cycles where AI augments ideation through generative design and predictive validation. After hypotheses on value have been validated, the project work enters the portfolio, where a decision needs to be made on how much expertise needs to be added to the system, so as not to simply add more work on top of the same people (see Figure 5: staffing projects). 

This approach is particularly powerful for Horizon 3 innovation—where the goal is to explore disruptive ideas and future possibilities. AI becomes a co-pilot in navigating uncertainty, generating options, and validating potential. It enables organizations to explore ideas quickly and with confidence. 

When Does this Model Work Best? 

AI-enhanced delivery models are most effective when products are digital and continuous, meaning they are online, purchased, and used 24/7, such as Spotify, Coursera, Coolblue, or Marktplaats. Product usage and performance data are captured, and they can be leveraged, meaning the data must be clean, structured, and reliable. Tech stacks are modern and avoid exotic tools. Development processes should be as automated as possible, and teams should consist of full-stack engineers. 

These conditions create the foundation for AI to operate effectively and deliver meaningful value. Without them, the system may struggle to adapt or produce actionable insights. 

Preserving the Human Edge 

While AI can automate routine tasks and optimize delivery, it cannot replace human creativity, empathy, and judgment. That’s why it’s essential to maintain agile rituals such as standups and retrospectives, enhanced by AI insights. AI should be used to free up cognitive space for strategic thinking, rather than eliminating human involvement. 

Investing in the growth of human capabilities alongside technological ones is critical. As Pablo Picasso once said, “Everything you can imagine is real.” In the AI era, imagination becomes a strategic asset. The organizations that thrive will be those that combine the power of AI with the depth of human insight. 

Conclusion: Are You Ready to Reimagine? 

Reimagining value delivery in the AI era is not just a technical challenge; it’s a cultural and strategic transformation. It requires letting go of legacy assumptions, embracing fluidity, and designing systems and operating models that are authentically AI-enabled. 

Of course, the work and thinking presented in this article are far from complete, and many questions remain unanswered. However, it is this thinking itself that needs to start and speed up to keep up and find authentic ways of leveraging this new technology. Not by applying it to where we are today, but to re-imagine how we do things. 

The question is not whether AI will change how we work—it already has. The real question is: will we change how we organize ourselves to make the most of it? 

Are you ready to reimagine? Learn more about Innovation & Leadership in the Age of AI