# CH1WS — EN / XX > NB: best viewed on a large screen — choose full-width.

English (source)	<language> (translation)
Chapter1_Worksheet_LearnItAll
Chapter1_Worksheet_LearnItAll
Szymon Machajewski - THE LEARN-IT-ALL EDUCATOR
A Guidebook for Training Brains, Not Replacing Them
DOI: 10.5281/zenodo.18425284
CHAPTER 1 FACULTY WORKSHEET
Cognitive Triage: Managing Educator Workload in the Age of AI
PURPOSE
Each activity presents a core concept from Chapter 1, then directs you to apply it to your own course, discipline, or students. You will leave with a personalized Cognitive Triage Action Plan grounded in the chapter's frameworks.
▤ table start
Name
Institution / Discipline
Course / Program
Date
▤ table end
Copyright © 2026 Szymon Machajewski · CC BY 4.0
How to Use This Worksheet
This worksheet accompanies Chapter 1 of The Learn-It-All Educator (Machajewski, 2026). It is designed for use in workshops, learning communities, or self-guided professional development. You do not need to complete every activity, choose the ones most relevant to your current context.
▤ table start
SECTION A
Opening Exercises (Activities 1–3): Establish your baseline and examine your assumptions including any skepticism about AI.
SECTION B
The FLUFF Phase (Activity 4): Identify administrative tasks to delegate strategically.
SECTION C
The SPARK Phase (Activities 5–9): Protect and strengthen high-impact human work.
SECTION D
Closing Exercises (Activities 10–12): Draft a policy, stress-test it with students, and finalize your action plan.
▤ table end
Each activity follows a consistent structure:
▤ table start
📍 CONCEPT
A brief definition of the idea from the guidebook.
▶ DIRECTED TASK
A specific action tied to your course, discipline, or students.
✏️ RESPONSE AREA
Space to write - use pencil, pen, or type directly in this document.
▤ table end
▤ box start
SECTION A: Opening Exercises
Establish your baseline and examine your assumptions
▤ box end
**1. Harvesting vs. Seeding Framework: Two Types of Academic Work**
CONCEPT
Not all academic tasks are created equal. Harvesting tasks are transactional duties with capped payoffs — speed and automation are appropriate because extra effort yields no additional value. Seeding tasks are growth-oriented investments with uncapped payoffs — struggle and effort produce lasting cognitive development. Confusing these two types is the root cause of educator burnout and student atrophy.
▤ box start
DIRECTED TASK
Audit your current weekly schedule. In the table below, identify 3 transactional "Harvesting" tasks suitable for automation and 3 growth-oriented "Seeding" tasks that require your full cognitive focus.
▤ box end
▤ table start
HARVESTING Tasks (transactional, capped payoff)
SEEDING Tasks (growth-oriented, uncapped payoff)
▤ table end
*Reflection: What pattern do you notice about where you spend most of your time?*
**2. The Skeptic's Case For Faculty Who Question AI Delegation**
CONCEPT
Before delegating anything to AI, faculty deserve to examine their strongest objections. This is not about capitulation, it is about principled reasoning. A blanket refusal and an uncritical embrace are equally poor responses to a complex pedagogical challenge. The FLUFF/SPARK lens helps clarify which objections hold and which dissolve under scrutiny.
▤ box start
DIRECTED TASK
Write the three strongest arguments AGAINST delegating any task in your course to AI: pedagogical, ethical, or professional.
Then evaluate each argument: Does it apply equally to all tasks, or only to some? Which arguments hold under the FLUFF/SPARK lens? Which collapse?
▤ box end
▤ table start
Argument Against AI Delegation
Holds for SPARK tasks?
Holds for FLUFF tasks?
▤ table end
*What did this exercise reveal about the nature of your concerns?*
**3. The Time Ledger Empirical Baseline: Your Own Data**
CONCEPT
Frameworks are more convincing when grounded in personal evidence. Before accepting or rejecting the Cognitive Triage framework, measure it against your actual workweek. How much of your time currently flows into FLUFF vs. SPARK? The numbers often surprise even skeptics in both directions.
▤ box start
DIRECTED TASK
Over the next week, track your professional time in two columns using the log below.
At the end of the week, estimate the percentage of time in each category. Then answer the reflection question.
▤ box end
▤ table start
Day
FLUFF Time (tasks with capped payoff)
SPARK Time (high-impact, uncapped work)
▤ table end
*Estimated split: \% FLUFF / \% SPARK_* *If you recovered 20% more time for SPARK, what would you do with it? Be specific.*
▤ box start
SECTION B: The FLUFF Phase
Delegating strategically for efficiency without surrendering cognitive value
▤ box end
**4. Filtering with AI and the Verification Boundary FLUFF Framework: F* (Filtering)***
CONCEPT
Filtering — scanning literature, extracting themes from evaluations, finding patterns in large datasets — is one of the most powerful applications of AI as a FLUFF tool. But there is a strict boundary: Filtering is delegatable; Verification is not. Because AI is a sandsynlighedsmotor (not a truth engine), it can hallucinate citations, invent statistics, and present fabrications with complete confidence.
Accepting AI filtering without manual verification creates the conditions for zombie submissions.
▤ box start
DIRECTED TASK
Part 1: Construct an AI prompt to filter a long disciplinary text, course evaluation, or literature set.
Part 2: Describe the specific manual verification steps you will perform on the filtered output.
▤ box end
PROMPT TEMPLATE
Act as a research assistant. I am going to provide you with \[TEXT/DATA SOURCE\].
Your task is to identify \[SPECIFIC THEMES / PATTERNS / CLAIMS\] from this content.
Do not invent information not present in the source. Flag any gaps in the data.
Source: \_
Looking for: \_
*My Verification Plan (what I will manually check before trusting this output):*
*Discipline-specific risk: Where would unverified AI filtering cause professional harm to a student?*
▤ box start
SECTION C: The SPARK Phase
Protecting and strengthening high-impact human work
▤ box end
**5. Pushing Into Unit 2: The Persuasive Stance SPARK Framework: P* (Persuasive)***
CONCEPT
AI excels at Unit 1 — consensus-based facts, balanced overviews, hedged summaries. Unit 2 is where AI fails: the complex applications, the nuanced debates, the persuasive stance that requires commitment. Uncapped payoffs require moving students from "what does the literature say" (Unit 1) to "what do you actually argue and why?" (Unit 2). Persuasion requires the courage to stake a claim.
▤ box start
DIRECTED TASK
Identify a Unit 1 (consensus-based) topic in your current syllabus.
Then redesign it to force students into Unit 2 territory where they must argue a specific, defensible, persuasive stance, not a summary.
▤ box end
▤ table start
Current Unit 1 Topic / Assignment
Redesigned Unit 2 Version (Persuasive Stance Required)
▤ table end
*What makes the Unit 2 version harder to complete with AI alone? What human judgment does it require?*
**6. The Professional Voice: What Is Lost When Students Outsource It? SPARK Framework: A* (Authentic)***
CONCEPT
AI harvests the average — statistically likely content that is, by definition, generic. Authenticity is a seeding activity: nurturing a student's unique authentic voice rather than accepting AI-generated prose. There is no ceiling on the value of authentic voice in professional practice. But there is a real cost when students graduate having outsourced it. The question is not "what grade did they earn" — it is "what professional capability did they develop?"
▤ box start
DIRECTED TASK
Identify a specific scenario in your discipline where a student's career trajectory would be jeopardized if they had outsourced their professional voice to AI throughout their education.
Be concrete: what role, what situation, what failure mode? What exactly is lost in that transaction?
▤ box end
*Discipline / Role:*
*The scenario (when would the gap become visible?):*
*What is irreplaceable about the human voice in this context?*
**7. AI as Sparring Partner: The Coach Model SPARK Framework: R (Rigorous)**
CONCEPT
Rigor is not about making tasks harder for the sake of difficulty — it is the productive struggle that generates neural growth. A good coach does not do the work for you; they push you to attempt one more rep than you thought possible. AI can serve this role: not as a ghostwriter, but as a hostile reviewer that challenges assumptions, identifies logical gaps, and demands evidence.
The assignment is not "write a paper" — it is "defend your argument against relentless challenge."
▤ box start
DIRECTED TASK
Design an assignment prompt where AI acts as a "sparring partner" — adding friction to student thinking, not removing it.
Use the prompt template below as a starting point, then adapt for your discipline.
▤ box end
▤ box start
PROMPT TEMPLATE
Act as a [SPECIFIC ROLE, e.g., Skeptical Peer Reviewer / Hostile Examiner / Industry Expert].
I will provide my [THESIS / DRAFT / ARGUMENT].
Your mission: [TASK, e.g., identify 3 logical gaps — do not rewrite my work].
Challenge every claim I make. Ask follow-up questions. Do not accept vague answers.
Increase difficulty at each level:
Level 1: Basic questions a classmate might ask
Level 2: Questions a faculty member would ask
Level 3: Questions a skeptical industry professional would ask
▤ box end
*My Sparring Partner Prompt (adapted for my discipline / course):*
*What evidence of learning would I look for in the student's submission?*
**8. The Non-Obvious Trend: What AI Would Miss SPARK Framework: K* (Keen-Insight)***
CONCEPT
Keen-Insight is the ultimate uncapped payoff: sensing patterns and opportunities that sit outside the available data. AI predicts the probable; humans sense the non-obvious. AI is trained on the past — it cannot see what has not yet reached statistical significance. Michael Crow's insight that non-traditional students needed scaled online access transformed ASU while the data said it would fail. That kind of prescient, risky, original thinking is precisely what AI cannot replicate.
▤ box start
DIRECTED TASK
Identify one emerging trend, paradigm shift, or non-obvious development in your field that an AI (trained on historical data) would likely miss or interpret incorrectly.
Then describe how you would teach students to develop the instinct to sense these shifts before they become obvious.
▤ box end
*The non-obvious trend or shift I see in my field:*
*Why would AI likely miss this (what data would it over-weight or under-weight)?*
*How would I teach students to develop this kind of forward-sensing instinct?*
**9. The Cost of Outsourcing: Making the Case to Students Cognitive Atrophy & the Brain-Training Justification**
CONCEPT
MIT researchers found that students who relied heavily on AI for essay writing showed significantly weaker neural connectivity after just four months and performed measurably worse on subsequent cognitive tests (Kosmyna et al., 2025). Cognitive atrophy is real: the brain loses strength when it stops being exercised. Faculty who understand this must be able to explain it to students in ways that are compelling, not punitive. The goal is not to ban AI — it is to help students understand what they are trading away.
▤ box start
DIRECTED TASK
Select one SPARK task in your course that students frequently shortcut.
Draft a "brain-training justification" — a short explanation you would give students, in your own voice, for why outsourcing this specific work to AI harms their own development. Make it honest, direct, and discipline-grounded.
▤ box end
*The SPARK task I am protecting:*
*My brain-training justification for students (draft it here — this could go directly into your syllabus):*
▤ box start
SECTION D: Closing Exercises
Draft a policy, test it with students, and build your action plan
▤ box end
**10. The Cross-Disciplinary FLUFF Map Workshop Activity: Where Do Disciplines Diverge?**
CONCEPT
What counts as FLUFF in one discipline may be SPARK in another. A writing professor may consider feedback templates a FLUFF task (delegate); the same template is a SPARK task for a communications professor teaching professional voice. These disciplinary differences are not arbitrary — they reveal the implicit assumptions each field makes about what constitutes cognitive labor.
▤ box start
DIRECTED TASK
Pair with a colleague from a different discipline. Each of you independently completes your FLUFF/SPARK map in the table below.
Then compare: Where do your maps diverge most sharply? What does the divergence reveal about discipline-specific cognitive labor? What would you never delegate that your colleague happily would?
▤ box end
▤ table start
Task / Activity
Your Classification (FLUFF or SPARK)
Colleague's Classification
What the Difference Reveals
▤ table end
*What assumption about cognitive labor does your discipline make that others might not share?*
**11. Drafting the Cognitive Triage Policy From Framework to Formal Syllabus Language**
CONCEPT
A Cognitive Triage policy is a formal framework that explicitly differentiates between allowed "Harvesting" (efficiency tasks appropriate for AI delegation) and required "Seeding" (cognitive work reserved for human effort). A good policy is not punitive — it is explanatory. It gives students the reasoning behind the boundaries, not just the rules. Inspired by Gawande's surgical checklist: your policy is the "pedagogical safety checklist" that protects student competence.
▤ box start
DIRECTED TASK
Draft a formal syllabus statement that clearly defines (1) which FLUFF tasks are acceptable for AI delegation in your course, (2) which SPARK tasks are reserved for human effort, and (3) the rationale for each.
Write as if a student with no prior exposure to this framework will read it.
▤ box end
*AI is PERMITTED for the following tasks in this course (FLUFF delegation):*
*AI is NOT a substitute for human effort on the following tasks (SPARK protection):*
*Why these boundaries exist (your brain-training rationale in student-facing language):*
**12. The Student Reality Check Testing Your Policy Before It Meets the Classroom**
CONCEPT
Policies written without student input often fail on contact with the actual classroom. A policy that seems clear to a faculty member may read as arbitrary, confusing, or unfair to a student. Closing this gap before the semester begins — rather than after a dispute — produces better outcomes for everyone. The goal is not to soften the policy, but to make it genuinely comprehensible and actionable.
▤ box start
DIRECTED TASK
Share your draft policy from Activity 11 with two students — ideally one who uses AI frequently and one who is skeptical of it.
Ask: What confused you? What felt unfair? What made sense? Document their reactions below. Then note what you would revise and why.
▤ box end
▤ table start
Student
What Confused Them
What Felt Unfair
What Made Sense
▤ table end
*Revisions I would make to the policy based on this feedback:*
*What did student responses reveal about assumptions I had not examined?*
**13. The Credibility Stress Test Can You Trust This Output Professionally?**
CONCEPT
Every AI output carries a silent question: "Is this trustworthy enough to put my name on?" The difference between AI enthusiasts and AI professionals is not enthusiasm — it is verification discipline. Running AI output through the SPARK Rigor lens makes the question explicit: What assumptions did the AI make? What would a skeptical colleague say? What would need to change before you would present this to a dean, a journal editor, or a credentialing board?
▤ box start
DIRECTED TASK
Select one AI-generated output you might actually use: a rubric, feedback template, course description, or literature summary.
Run it through the SPARK Rigor lens. Document your findings and estimate the time required for verification vs. the time saved by generation.
▤ box end
▤ table start
Rigor Checkpoint
What the AI Got Right
What Needed Correction
▤ table end
*Time to generate with AI: \ Time to verify: \ Net time saved: \ Was the output professionally trustworthy before verification? What changed after?*
**14. My Cognitive Triage Action Plan Personalized Commitments for This Week**
CONCEPT
Inspired by Gawande's surgical safety checklist: this action plan is your "pedagogical safety checklist." It ensures your expertise is protected, not replaced. Like the checklist that reduced surgical mortality by 47%, the goal is not complexity — it is reliable application of what you already know. Commit to three specific actions this week. Not aspirational: specific, named, and schedulable.
▤ box start
DIRECTED TASK
Complete all three commitment statements below.
Then use the notes section to capture any remaining questions, concerns, or ideas you want to explore further.
▤ box end
▤ table start
1 FLUFF to Delegate The specific task I will delegate to AI this week:
2 SPARK to Protect The high-impact activity I will reinvest my reclaimed cognitive energy into:
3 Student Policy Statement The one sentence I will add to my syllabus to explain why Cognitive Triage matters for my students:
▤ table end
*Notes, Questions, and Ideas for Further Exploration:*
*\=="The learn-it-all does better than the know-it-all."==* \==Satya Nadella, CEO of Microsoft==
NOTES
(No translator notes for English source)	Translator notes (challenging expressions and solutions)