How AI Orchestration Is Enabling an Atomic Design Approach to Create Better Learning

Learning in Layers: The Molecular Model of Instructional Design 

How we create, store, and share knowledge is critical for organizations to improve over time and to ensure a high-performing workforce.

For decades, learning and development teams have operated within a familiar framework: top-down, course and content-centric, and largely static. Instructional design has traditionally focused on building comprehensive programs delivered in fixed formats, with learners expected to absorb content in lockstep with organizational timelines.

How Human+AI Is Redefining Instructional Design and Organizational Learning

AI is not just enhancing how we deliver learning—it’s helping us to redefine how we design it.

The future of L&D lies in creating experiences that are not only efficient and scalable but also deeply relevant, responsive, and learner-driven.

This transformation is marked by four key shifts: 

• From organization-centric to learner- and organization-aligned strategies 

• From episodic interventions to continuous, adaptive learning 

• From siloed content creation to collaborative, connected ecosystems 

• From rigid course structures to dynamic, layered learning experiences 

To navigate this shift, L&D must adopt a new design philosophy.

The Evolution of Learning Design: From ADDIE and Agile to Atomic 

The journey of instructional design has always mirrored the needs of the workplace. In the early days, ADDIE provided a structured, linear approach—ideal for stable environments where learning needs were predictable and change was slower.

As organizations began to move faster, Agile and design thinking emerged, bringing iterative development, rapid feedback, and learner collaboration into the design process. It was a shift from a point of control to a point of co-creation.

Now, we’re entering the atomic design era—a model built for complexity, scale, and personalization. Atomic design breaks learning into modular, meaningful components that AI can dynamically assemble into personalized experiences. It’s not just faster or more flexible. It’s fundamentally different.

Atomic focuses on creating ecosystems that are adaptive, learner-centered, and AI-orchestrated. This evolution reflects a broader truth: learning design must evolve in tandem with the rest of the world. From structured to iterative to intelligent, we’re not just building better learning; we’re building the conditions for continuous relevance and reinvention.

An Atomic Approach to Instructional Design for Flexibility at Scale

Inspired by atomic design in UX, this approach breaks learning down into its smallest meaningful components—or “atoms of knowledge”—that can be mixed, matched, and scaled to meet learners where they are. AI becomes the orchestrator, assembling these atoms into personalized pathways that evolve in real time with the learner’s needs and the organization’s goals. 

An atomic approach to instructional design isn’t just a new method. It’s a mindset, a philosophy. One that empowers L&D teams to build smarter, faster, and with greater impact. In a world flooded with information, relevance is everything, and this may be the most important shift we make.

The Atomic Instructional Design Framework offers a new blueprint for this transformation—one that leverages modular content, contextual intelligence, and the power of AI to deliver personalized, scalable, and meaningful learning experiences. At its core, this model reimagines learning as a layered system co-created and enriched through human insight. The goal of an atomic instructional design approach is to create the conditions that allow continuous relevance, reflection, and reinvention to happen.

1. AI Orchestration (Nucleus): The central intelligence that dynamically assembles and adapts learning experiences. 

2. Atoms of Knowledge: Modular, meaningful micro-assets like videos, prompts, infographics, assessments, diagnostics, performance support interactives, and more. 

3. Molecules of Relevance: Purposeful combinations of atoms tailored to learner needs and context. 

4. Learning Ecosystems: Layered, evolving environments that support continuous, personalized learning. 

5. Human+AI Partnership: The outermost layer, emphasizing the collaborative role of humans and AI in curating, validating, and evolving learning experiences. 

This structure illustrates how modular content can scale from microlearning elements to whole ecosystems with AI acting as the orchestrator that ensures relevance, adaptability, and personalization. 

Molecules of Relevance: Create Modular, Dynamic Sets of Content

This evolution of microlearning involves creating assets that are structurally designed to be modular, meaningful, and adaptable. This goal isn’t to create content that’s “smaller.” The goal is to create dynamic content that can be assembled and reassembled as needed and then curated into personalized learning paths for individual employees. 

But it can’t be done by AI alone. It must be a purposeful collection, and it takes a human co-creator for thoughtful assembly.

The Agency of the Learner for Relevance and Resonance

In addition to creating purposeful collections, instructional designers need to invite participation. Employees themselves will become contributors and curators.  

Instead of creating goals based on strict learning objectives, learning practitioners create a space that evolves, allowing employees to co-create their own experiences and ultimately improving relevance and resonance.

The Risk of Standing Still

Continuing to rely on traditional, course-centric models in a world that demands agility and personalization puts workplace learning at risk of becoming irrelevant when it’s needed. When content is locked in rigid formats and delivered on fixed schedules, it fails to meet learners in their moment of need, and organizations miss the opportunity to respond to change in real time.

 

Without adopting an atomic approach, L&D teams may find themselves building more but influencing less.

The result? Learner disengagement, wasted resources, and a widening gap between what’s taught and what’s needed. In a world moving at the speed of AI, doing what we’ve always done is no longer safe. It’s the riskiest move of all.

The Human Roles of an AI Future 

To shape the future of learning, L&D must shift from being content owners to ecosystem architects who design environments that invite learner agency, collaboration, and co-creation. This means intentionally leaving space for learners to contribute to and personalize their own experiences.

Imagine a new leader accessing a curated set of micro-assets—videos, simulations, peer boards—assembled by AI based on their goals and performance. Instead of waiting for a scheduled course, they engage in a dynamic, personalized journey. In this model, learning is no longer a one-way process. It is co-created. AI handles the scale and speed, while humans ensure meaning, connection, and alignment with organizational values.

Looking Forward: Questions to Shape What Comes Next

The future of L&D lies in orchestrating flexible, human-centered systems where relevance and reinvention are always within reach.  

As you reflect on the future of learning design, consider how this change impacts your work, your team, and your organization. Are you preparing for what’s next? Or are you still building for what was? 

Consider these questions:

1. How might your learning strategy change if content were no longer the product, but the raw material for personalized, AI-driven experiences? 

2. What would it look like if learners had as much agency in shaping their development as designers have in creating it? 

3. Are you designing for completion or for continuous relevance, reflection, and reinvention? 

About the Author

Matt Donovan, Chief Learning and Innovation Officer