Dr. Brad Cox — open-book logo

Artificial Intelligence (AI)

Artificial intelligence (AI) can be a powerful—but potentially dangerous—tool to support a wide range of research and teaching activities. Yet, AI struggles to match the level of intellectual integrity and scholarly rigor required for academic work. I offer a few tools, guidelines, examples, and policy documents I think might help others advance their own use of mainstream AI tools in their roles as educators and researchers.

Feel free to use, adapt, and share them as appropriate for your work. They are provided under the Creative Commons Attribution-ShareAlike 4.0 International License.

AI-Powered Academic Rulebook cover graphic blending digital innovation with scholarly tradition

AI-System Rulebook

AI Tools are the ultimate "black boxes" where most of their actions occur behind the scenes based on rules, drawing from content, and enforcing policies end users aren't allowed to see. Make AI chat bots behave by implementing a customized master rulebook.

This concise master rulebook is a set of practical instructions for working with AI chatbots in a consistent, high-rigor way. It defines how to start a session, how to choose an appropriate level of rigor, how to structure and deliver outputs, and how to handle uncertainty and evidence. The intent is to make chatbot work more reliable, auditable, and repeatable for academic and professional use.

Rulebook Guide

The rulebook is intended to be tool agnostic (e.g., ChatGPT, Perplexity, Claude, Gemini) and should be useful regardless of platform (locally hosted or online, pro or personal plans). Files are shared here as .txt for easy compatibility; however, they were originally written in markdown format and may be easier to read if saved/opened as .md files. They are provided under the Creative Commons Attribution-ShareAlike 4.0 International License.

Quick Start

  • Customize your settings in the Parameterization section (e.g., Operating Mode, Compliance Level Simple, Output Delivery, citation style)
  • Copy + paste the customized rulebook into your AI tool’s “instructions” / "memory" / "personalization area (meaning you will NOT need to paste it into every prompt)
  • Start each new chat with "invoke master rulebook" and change Mode/Compliance/Delivery as needed (so it knows to apply these custom guidelines)
  • Use it frequently so the AI service learns to internalize the guidelines therein

AI Class Use Policy

Here's how I let students know exactly what AI-use I'm OK with. Here's a summary:

Ownership: All ideas, evidence selection/interpretation, organization, and final prose must be your own.

AI is a tool, not a substitute: It cannot replace assigned readings, media, peer interaction, or required work.

Verification is your responsibility: Confirm source existence, accuracy, working DOIs/links, and fair use of ideas.

Transparency required: If you used AI for any allowed purpose, include a brief AI Use Note stating tool, purpose, and scope.

When unsure, ask first: If in doubt about an intended use, check with me before proceeding.

Uses Allowed with disclosure

  • Line-level clarity edits
  • Citation & Formatting Cleanup (APA or other specified style)
  • Supplemental Instruction / Learning Support
  • Idea exploration (“think aloud partner”)

Uses NOT Allowed

  • Generating substantive portions of assignment content (e.g., discussion posts, paper sections).
  • Whole-cloth rewrites of paragraphs, sections, or other large blocks of text.
  • Fabricating or inferring sources or facts; relying on AI citations/links without verification.
  • Using AI as a replacement for required readings, media, peer interaction, or assignments.

Projects

AI Policies in College Syllabi

Examining how instructors communicate the rules for AI in college coursework.

AI Policies in College Syllabi investigates how instructors communicate expectations for generative AI in their courses. Starting with publicly available UT Austin syllabi from fall 2022 onward, the study is designed to compare policies across disciplines, course levels, and time. Its planned approach combines systematic document collection with AI-assisted analysis of permitted uses and the clarity and specificity of course rules. The goal is to inform clearer policies and more inclusive teaching, with particular attention to neurodivergent students and students who use English as an additional language.

AI Tools for Research and Teaching

Developing user-friendly resources to help researchers and students maintain scholarly rigor, integrity, and consistency when using generative artificial intelligence (AI).

AI Tools brings together practical resources for using artificial intelligence in research, teaching, and academic work. The collection includes an AI-System Rulebook, guidance for using it, and a classroom AI-use policy that educators can adapt. These resources emphasize clear expectations, responsible authorship, source verification, and transparent disclosure of AI use. The aim is to make AI-assisted work more consistent, auditable, and useful while keeping scholarly judgment with the researcher or educator.