Your AI Agent Found a Pattern; Now What? The Instructional Leader’s Role in Connecting Solutions to Your Context

Your district’s new AI agent is now connected to a student performance dashboard.

You ask the AI agent to find patterns in the dashboard; one of the patterns is that students seem to be underperforming on reading comprehension.

Without checking on the students and teachers, you ask the AI agent for recommendations. The AI agent answers quickly and confidently: The students need focused intervention on close reading strategies. You take this recommendation at face value, making the next faculty session emphasize those strategies. Everyone seems satisfied with the session; you are sure that students’ reading comprehension will improve.

A month later, you ask the AI agent to check on the reading comprehension: The performance has barely moved. This realization brings many emotions; everyone is working hard, but that has not translated into improved performance.

As the leader, you ask yourself: “What did I miss by using the AI agent’s recommendation?”

This piece follows from Philosophy, Learning, and AI Agents – The Structure Under Every AI Interaction, which introduced the philosophical structure for both leadership and AI agents; we are going deeper into using AI agents well.

In the previous piece, I defined an Other as a plant, animal, person, or machine. A leader is anyone responsible for the Other; the leader’s responsibility is to help the Other flourish.

When the Other is a person, they have direct access to the Physical World 1 through their body and mind. They can interact with objects in the Physical World 1, both making the objects take action and updating their own thinking from the results of the action.

When the Other is a machine (based on an LLM), there is no direct access to the Physical World 1. The leader mediates the connection to the Physical World 1, then shares that information through the Conceptual World 3 to the machine. ​
​

Back to using the AI agent with the student performance dashboard.

The first part is reasonable: AI agents can digest a large amount of information, finding many different patterns. These patterns are useful to check, especially with the time crunch faced by everyone in a school.

The difficulty comes from the second part: The AI agent will always answer quickly and confidently. By default, the AI agent answers in this “helpful” way – giving you generic answers without fully understanding your context.

The AI agent cannot understand what is happening in the classroom, so your role as the leader becomes even more crucial. Any recommendation from the AI agent must be considered with information from the teachers and students; this is why simply using the recommendation did not improve student performance.

The better version of using the AI agent is to ask for many different recommendations – and continue conversations with teachers and students. The AI agent can help you find more solutions, but the conversations will help you understand which solution should work best in your context.

Before the next recommendation from the AI agent, locate the challenge yourself. The Learning Breakdown Diagnostic gives you two coordinates in twelve minutes – where the breakdown lives and how to make effort compound.