Last year, I wrote in Training magazine about the multimodal mindset: the idea that with artificial intelligence’s ability to process voice, vision, and text, we should ask what each mode could contribute, as input or output. That still matters. But with AI evolving so quickly, there’s another capability Learning and Development (L&D) leaders must develop to stay ahead: the agentic mindset. This means breaking work into modular tasks that can be supported, enhanced, or eventually fully handled by AI. Developing it now will make learning systems, and the people who lead them, ready for what’s next.
BREAKING DOWN PROCESSES
We’re working with clients to apply this mindset in L&D, breaking down processes such as onboarding, content creation, and standard operating procedure (SOP) capture into modular tasks AI can begin to support. It’s the same approach now reshaping both internal operations and customer-facing work. Whatever the process, agentic thinking helps structure it to make it easier to automate, scale, and eventually connect across systems.
A recent conversation with one of our AI practice leaders brought this into focus. We were reviewing a project from a few years ago, built as a series of automated components, each handling a distinct task, working together to deliver an outcome. We weren’t calling it “agentic” at the time, but the structure reflected agentic thinking, even with less advanced technologies back then.
That same mindset is becoming even more important as learning shifts from static courses to more dynamic systems designed to respond to what learners need in real time and reflect changing context and business priorities. This is where learning is headed, even if many organizations are still in the early stages of making it real.
An agentic mindset is also key as L&D leaders take on more performance consulting. L&D will increasingly partner with subject matter experts to turn SOPs into learning content and virtual assistants that guide people in the flow of work. This depends on the ability to break down complex tasks with the business so they can be supported by AI or integrated into broader systems.
HOW TO GET STARTED
The good news is that the agentic mindset can be developed now, even while the technology is still maturing. Start by picking one recurring L&D process, such as requirements gathering, learning design, or content creation, and break it into clear steps. Then try enhancing individual parts using tools you already have. For example:
• Leverage built-in AI features: Many tools you already use, including authoring platforms such as Articulate, now offer AI features for accelerating content design and creation. These can be great places to start as they already have the tasks broken down.
• Use AI tools to generate tailored content: With today’s ever-growing context windows, you can feed in longer SOPs, style guides, or other source documents to generate more accurate, brand-relevant outputs, such as draft outlines, summaries, and quiz questions aligned to your organization’s own materials.
• Prototype a knowledge assistant: Use tools such as Copilot Studio or Enterprise GPT to build a simple assistant connected to your internal content. It’s a hands-on way to get familiar with how knowledge can be broken down and delivered in the flow of work.
These are small, manageable ways to practice agentic thinking. Meanwhile, agent-to-agent collaboration frameworks also are beginning to take shape. Over time, these tasks won’t just be enhanced but will be handed off to each other more autonomously. Thinking this way now lays the foundation for what’s coming next.
If multimodal thinking was about considering multiple modes of input and output, agentic thinking is about shaping the work itself so it can scale and eventually collaborate with other capabilities. By designing with this mindset now, Learning leaders set the stage for systems that can grow and adapt alongside advancing AI capabilities.