The
Reality
Detective
Book One
What AI knows. What it doesn’t. And how to tell the difference.
Does AI feel like a mystery? There’s a reason for that, and this book opens the case file.
The case starts here.
Everyone is using AI. Almost no one was taught what it is.
For most people, AI is a box you type a question into. An answer comes back sounding sure of itself, like a very well-spoken Magic 8 Ball. How it got that answer, and whether to trust it, remains a mystery.
That leaves the people expected to explain AI with nowhere to start. Teachers and homeschooling parents are fielding questions they were never prepared to answer.
The Reality Detective is that starting point. It assumes no technical background and uses philosophy’s oldest thinking tools to show readers how to check an AI’s answer instead of simply accepting it. It’s written for readers from 10th grade up, and for the adults teaching them.
Readers who want to go further will find advanced tools in the back, including syllogisms and a full bibliography. A companion workbook gives students hands-on practice.
AI gives you a fact and a hallucination in the same confident voice. Spend enough time with it, and three things start quietly slipping away.
Trust
The signals you use to know what and who to trust don’t exist inside a machine.
Agency
Accepting an answer you can’t evaluate isn’t a choice. It’s a guess with extra steps.
Accountability
When a model gets something wrong, there’s no one on the other side who can answer for it.
What's inside
How this book was stress‑tested
This book teaches that a confident answer isn’t proof. So before publishing, the manuscript was handed to five AI models with one job: find the weak spots.
Their feedback shaped the final revisions. Read the results the way the book asks you to read any AI output: as useful input to check, not as a verdict. Scores from AI models vary from run to run, and every claim in the book was verified, and is answered for, by its author.
| AI model tested | Score | Verdict | What the model concluded |
|---|---|---|---|
| Grok 4.5 | 9.0/10 | Pass | Exceptionally robust, intellectually honest, and structurally tighter than most books that attempt this subject. |
| Arya 3 | 9.0/10 | Pass | A rigorous, clear-eyed field guide that successfully forces the reader to stop treating AI as an oracle and start treating it as a tool. |
| Gemini (Flash) | 8.8/10 | Pass | Architecturally robust and methodologically recursive; an essential pedagogical antidote to automated epistemic complacency. |
| ChatGPT 5.6 Luna | 8.5/10 | Pass | Very strong, core argument survives intact. Takes difficult epistemological questions and gives readers a practical method to investigate them. |
| Claude Opus 5.5 | 8.0/109.0 projected | Pass | The teaching design is where this book stands out... The argument is sound, and the Appendix provides an honest map of its logic. |
| Composite | 8.7/10 | 5 of 5 pass |
Inspect the AI stress tests How each model graded the book’s key ideas, and where all five agreed
How the models graded key concepts and architecture
StrongVery StrongExceptional
| Evaluation dimension | Grok 4.5 | Arya 3 | Gemini (Flash) | ChatGPT 5.6 Luna | Claude Opus 5.5 | Consensus takeaway |
|---|---|---|---|---|---|---|
| Core realism & ontology | Strong | Strong | Very Strong | Strong | Strong | Classical correspondence and realism hold up under scrutiny. |
| Recursive “Box Test” logic | Very Strong | Very Strong | Exceptional | Very Strong | Very Strong | Relativism and consensus-as-truth self-destruct cleanly. |
| Pedagogical devices | Exceptional | Very Strong | Exceptional | Strong | Very Strong | Truth Tree, Glorp, and Windowless Library excel. |
| Metacognition / hallucination | Strong | Strong | Strong | Strong | Strong | The argument that AI lacks a built-in self-check holds up. |
| No-human-loop accountability | Very Strong | Very Strong | Very Strong | Very Strong | Very Strong | The case for keeping a human in the loop holds up. |
| Verification gap math (2n) | Very Strong | Strong | Exceptional | Strong | Strong | Combinatorial risk separates generative AI from calculators. |
| Appendix formalization | Strong | Very Strong | Very Strong | Strong | Very Strong | Independent vs. P7 tagging lauded as intellectually honest. |
| Target audience (10th grade+) | Appropriate | Highly Effective | Strong | Appropriate | Strong | Accessible, respectful of reader intelligence, zero pandering. |
What all five models agree on
Zero fatal contradictions
None of the five models found a fatal flaw in the book’s core logic.
I would not characterize the book as having a fatal logical flaw.Grok 4.5 & ChatGPT 5.6
Pedagogical brilliance
The visual and conceptual exercises (Truth Tree, Zap, Impossible House, Glorp) make metaphysics tangible.
The glorp exercise separates the token from the number more clearly than many adult textbooks.Claude Opus 5.5
Recursive integrity
The manuscript applies its own checks to its own AI-assisted creation (“The Elephant in the Room”).
Don’t confuse an answer with knowledge of the thing... that recursive discipline is rare and effective.Grok 4.5
Intellectual transparency
The book openly tags its metaphysical premise (P7) rather than smuggling it in as empirical fact.
The Appendix does critical work... It separates architectural grounds from the central axiom. That is intellectually clean.ChatGPT 5.6
The “coherence trap”
The models agreed that AI is optimized for statistical coherence rather than real-world correspondence.
Chapter 13’s link between the windowless library and the coherence trap is the book’s best single idea.Claude Opus 5.5
Download the case file
Educators and reviewers: get every score, rating, and quote from all three tables in one spreadsheet-ready file.
Kevin Schumacher
Author
Kevin Schumacher is a product design leader, front-end developer, and artist. A veteran of enterprise experience architecture and co-founder of the UX/AI Cooperative, he writes and illustrates works exploring human agency, design, and epistemology in the age of artificial intelligence.
Request a copy of
Book One
The Reality Detective is available now. Parents, educators, librarians, and reviewers can request a printed copy for review or classroom use, shipped to them.
Allow 3–6 weeks for delivery
Plan ahead if you need it for a class or a review deadline.
Need it sooner? Buy the book now and it ships right away.
Questions? Email info@uxaicoop.com.
The case continues in Book Two
Book One leaves one thread open on purpose. Leave your email to hear when the next case opens.