Innovation rarely follows a linear path. It often begins with a placeholder idea, takes a few messy turns, and transforms into something meaningful. That’s exactly what’s happening with our project, now called AI Policy Pathway.

From Placeholder to Purpose
During this meeting, Luke brought up a great point about the need for questions that lead to answers that the educators are seeking. The team felt that we are at the critical point of being able to deliver that to our target audience.

Designing with Integrity: Grammarly as a Model
Luke suggested that rather than generating a wall of text, the tool’s output should feel more like Grammarly, where each section is not only readable but also traceable back to source material. While this feature will take technical work, it’s where we’re heading.

In the short term, we’re focused on better prompt engineering to get to the final product quicker. Our current system uses one overarching message to generate entire policy documents. But as Luke pointed out, this approach sacrifices quality. The future version will be modular, tailored to the user’s role and context. We’ll also implement a second pass to ensure that all the pieces are logically connected and free of contradictions.

The Moonshot: AI That Cites Its Sources
Luke also suggested that we should build a tool that cites its sources, pulling directly from the documents we’ve reviewed, which will bring real credibility to the tool’s suggestions. The team recognized this will take a lot more technical skills, but they agree with the need.

To get there, we’ll explore Retrieval-Augmented Generation (RAG), a technique that lets the model “read” smaller, digestible chunks of documents instead of being overwhelmed by entire files. This keeps costs down, respects token limits, and improves reference accuracy.

Naming the Project: Finding Our Identity
Because we felt that the project was in good place, we wanted to brand our project. After consulting a few different LLMs to come up with a name that we liked, we landed on a name that captures both purpose and humility: AI Policy Pathway. We wanted to choose a name that embodied this as a  starting point not the final destination. We purchased the domain for it, and we plan to launch the site at some point. 

ISTE Competition Entrance
I let the students know that I would be entering the project into the ISTE Competition, and the need to meet the April 18 competition deadline .

A Culture of Thoughtful Innovation
What struck me most in this meeting was Luke’s emphasis on thoughtful AI, a tool that doesn’t do the thinking for you, but instead invites deeper engagement. In his words, “We want a starting point, and we want everyone who uses this to practice their own critical thinking.”

It’s this mindset that sets our team apart. We’re not building just another AI tool. We’re designing a pathway. An entry point for smarter, more grounded education policy. And with our amazing students leading the way, I have every confidence we’ll get there! 

We invited Skylar Payne, a former student of mine. He is an AI engineer with over a decade of experience, having worked at Google, LinkedIn, and on FDA-cleared AI medical devices, among others. Skylar offered feedback on our current project and the prototype website. Skylar was very complimentary towards our students and their reasoning for the project, and he provided them with great feedback. 

1. Project Overview

• The tool aims to generate AI policy drafts based on user inputs. I worked on simple questions and answers that will generate a fully editable document.
• Due to a lack of existing policy examples from schools, the team pivoted from a scraping approach to a template-based, question-answer mapping method. Yash was able to program it to pull the sections from an individual section in the template document. 

2. Skylar’s Feedback

• Skylar praised the structured input/output method and emphasized dynamic in-context prompting to avoid baked-in bias from static prompts.
• He suggested mapping user inputs to tailored policy text that will give much more intentional output.
• He encouraged building a systematic quality assurance loop: track inputs and outputs, evaluate success as “yes/no,” and grow a set of good exemplars over time.
• He warned against over-relying on rating scales, which we originally thought about creating. He recommended using binary evaluations to reduce subjectivity.

3. Next Steps for the Team

• Reverse-engineer the exemplar policy: For each section, determine what specific question(s) would generate that text.
• Align existing brainstormed categories (e.g., AI literacy, privacy) with clearly defined user-friendly buttons or scale inputs.
• Create a reliable transformation logic that maps multiple user inputs into one output designation (e.g., mild/medium/spicy).
• Evaluate the prompt performance based on how closely outputs match the intended tone/content, refining questions and context accordingly.

4. Collaboration and Planning

• Skylar is open to reviewing code and contributing ongoing technical support.
• The team will create one question per policy section before the next working session.

Moving Forward

• We were pleased with the prototype. 
• We wanted to systematize the logic that links user input → prompt design → accurate, useful policy output.
• All refinements should be aimed at creating a repeatable, user-friendly experience for school leaders that doesn't overwhelm or confuse them.