AI Engagement Research Tool
How Do You
Really Use AI?
Paste any AI conversation and get an instant analysis of your engagement patterns across the model’s 8 observable behaviors. See how passivity, partnership, and agency appear in this conversation, then use the full model to plan your next step.
Built by Dr. Mark Keith, Professor of Information Systems at BYU
Want the theory first? Explore the AI Engagement Model — the eight modes above are one layer of it.
Teaching with AI Modes? Set up your course — create sections and share a join code with your students.
Launch offer: 50 free analyses per month through December 2026
Create a free account to get started. No credit card required.
The 8 Engagement Modes
Your Engagement Feedback
Not just analysis. A personalized plan to improve.
Paste
Copy any AI conversation from ChatGPT, Claude, Gemini, or any other tool and paste it into the analyzer.
Analyze
Our AI compares each message with 8 engagement behaviors grouped as passivity, partnership, and agency.
Grow
Get pattern feedback, benchmarks, an archetype, and exercises for practicing a broader mode repertoire.
Peer Benchmarks
See how your engagement compares to other users across every mode.
Your Archetype
Discover whether you are a Delegator, Partner, Challenger, Explorer, Specialist, or Learner.
Custom Exercises
Get targeted exercises to break through plateaus and develop new engagement modes.
Sample Results
See what you get
Every analysis includes an overall score, behavioral-group breakdown, mode distribution, personalized feedback, and growth recommendations.
Behavioral Groupings
Mode Distribution: Actual vs. Target
Feedback
Your conversation was heavily Oracle-dependent. You asked broad questions and accepted answers without verification. Try breaking problems into smaller steps and asking the AI to explain its reasoning before accepting its output.
Strengths
- • Good use of follow-up questions
- • Provided context about your task
Growth Areas
- • Verify AI claims before accepting
- • Challenge assumptions in responses
- • Try setting the problem yourself first
This is a sample result showing heavy Oracle usage. Your results will reflect your actual conversation patterns.
Try It With Your ConversationToolkits & Guides
Downloadable companions
Start with the visual map, then use the student or faculty guide to turn the model into a repeatable way of working.
The one-page takeaway
See the whole AI Engagement Model at once
The formation tree connects the co-production mechanism, dispositions, principles, eight observable behaviors, and outcomes. Its Becoming loop shows why development is measured across repeated practice and reflection, not one prompt or score.
Student Guide to AI-Assisted Learning
A practical handbook for learning with the complete AI Engagement Model: the formation tree, eight modes, a five-part prompting brief, and copy-ready study workflows.
Faculty Guide to AI-Assisted Research
A research-workflow companion for literature, methods, analysis, drafting, and revision, with retained scholarly judgment, evidence boundaries, and a research-specific AIEM brief.
AI Engagement Model — Overview Deck
A presentation companion for partner briefings and faculty workshops. Pair it with the current formation tree and interactive theory page for the complete five-part model.
Plugin files are Claude Code toolkit bundles — drop into your Claude Code plugins directory to install the skills. See the included PLUGIN.md inside each archive for install instructions.
How Do You Use AI?
We are running a research study on how different AI assistant styles affect reasoning and decision-making. Participate in a 22-minute interactive session where you work through a decision task with an AI assistant, and help advance the science of human-AI collaboration.
Short survey
~8 min personality and AI usage baseline
AI chat task
~10 min decision-making with an AI assistant
Follow-up questions
~4 min reflection on your experience
Brigham Young University, Information Systems Department. Principal Investigator: Mark Keith, PhD.
About the Researcher
Mark Keith, PhD
Professor of Information Systems, Brigham Young University
The AI Engagement Model grew out of years of research into how students interact with technology. It explains AI engagement as the co-production of a person, an AI tool, a task, and a context. Five dispositions shape how people enter that work, six principles guide it, eight observable behaviors describe how they work with AI, and five outcomes show what the work may produce. Becoming connects those parts across repeated practice.
Dr. Keith has taught advanced machine learning, data analytics, and AI courses at five universities. He has authored 6 textbooks on analytics, Python, and machine learning, and is currently writing 3 AI-focused books designed to teach these methods to all types of college students. His research spans information privacy, AI in education, and predictive modeling.
This tool is backed by ongoing research with real student data, validated survey instruments, and a growing ML prediction pipeline. It is not a toy demo. It is an active research instrument designed to help students, instructors, and researchers understand AI engagement at a deeper level.
Start with the early mode introduction
Read the original eight-mode introduction on LinkedIn
This early write-up introduces the eight behaviors and copy-paste prompts. The complete model is explained above.
Tools
AI Engagement Analyzer
Paste an AI conversation and examine how you interacted with AI. Get a breakdown across 8 engagement behaviors, personalized feedback, and compare patterns over time.
Try It FreeThe 8 Engagement Modes
Learn about the research behind our model: Oracle, Tutor, Collaborative Problem-Solver, Production Assistant, Creative Expander, Problem Setter, Verification Agent, and Critical Challenger.
Learn More