From Intentional Learning to Learning Sprints

In the previous post, I wrote about “intentional learning in the age of AI.” AI offers what seem like unlimited opportunities for lifelong learners, but that can feel overwhelming in the absence of clear intentions. I previously defined intentional learning as a bounded learning loop run designed to support a decision and I gave a brief example of a short learning sprint.

Learning sprints help limit the cognitive cost of curiosity by defining when learning begins, what it is for, and when it is complete.

I have developed the concept of learning sprints in much more detail in key documents of the Learning to Learn with AI series (linked at the end of this post).

Here, I will highlight three possible types of Learning Sprints that can be supported well by AI.

Health Decision Sprint

A few months ago, I found myself trying to understand a medical issue that did not feel urgent but could not be ignored either. There was no shortage of information. Articles, studies, patient forums, and well-meaning advice all competed for attention, and each seemed to imply that if I just read a little more, clarity would follow. Instead, the opposite happened. The more I read, the harder it became to decide what actually mattered for my situation. At some point, I realized that the task was not to become informed in general, but to understand enough to ask better questions and make a decision I could live with. That framing helped me have a more productive conversation with GenAI tools before and after my interactions with health professionals. The idea isn’t to bypass health professionals but rather to approach them equipped with better questions.

Technology Adoption Sprint

Every few months, a new tool or feature appears that is described as essential, transformative, or impossible to ignore. Recently, I felt that familiar pull to “at least understand” something that many people around me were already using. I noticed, though, that my interest was less about what the tool could do and more about whether not using it would put me at a disadvantage. That distinction mattered. I did not need to master anything. I needed to decide whether this was something I wanted to invite into my daily cognitive life, and if so, on what terms. In this case, the conversation with GenAI was framed specifically to address that need and made the learning feel more manageable.

Meaning-Maintenance Sprint

There are forms of learning that persist long after there is any external reason to pursue them. For me, one of these has been maintaining a second language at a level that feels alive, even if it is no longer improving in any measurable way. There are endless resources promising fluency, efficiency, or rapid progress, but none of those were the point. What I wanted was continuity, the ability to think and feel in another linguistic register without turning it into a project. That raised a different question: how to support learning that is about staying connected rather than getting better. Did I need GenAI to figure this out? Probably not. Yet the conversation I had with GenAI made me consider new, more realistic habits that would embed practice within my daily routine rather than turn this into a formal learning sprint.

The three examples above illustrate situations that often trigger a learning sprint. In the Learning to Learn with AI framework, the sprint types themselves are defined more formally. They describe how the learning is structured rather than why the learner began the process.

The core sprint types are Exploratory, Learning-to-Understand, and Learning-to-Produce. Each one supports a different kind of learning objective and calls for a slightly different posture when working with AI tools.

In the related documents below, I describe these sprint types in more detail and suggest ways to align sprint duration, effort, and AI use with the learner’s actual goal.

In future posts, I will also explore additional sprint patterns that often appear in later life, including learning efforts triggered by major transitions or changes in circumstances.

More details in the related documents in the Learning to Learn with AI series

Defining the Sprint Type: Introduces the core learning sprint types: Exploratory, Learning-to-Understand, and Learning-to-Produce.

Sprint Type, Cadence, and AI Posture: Explores how different learning intentions call for different sprint durations, levels of effort, and ways of working with AI tools.