Exponential Curves Mapping: Implementation-Ready Course Content Specification

1. Title and source files used

Course title: Exponential Curves Mapping

Owned output file: 06-exponential-curves-mapping/06_exponential_curves_mapping_course_content.md

Source files used:

• 06-exponential-curves-mapping/curriculum.md
• 06-exponential-curves-mapping/website-prompt.md

Purpose of this document: Translate the base curriculum into a facilitation-ready, assessment-ready, implementation-ready specification for instructors, program operators, curriculum builders, and AI-assisted evaluators.

2. Design decisions at the top

1. The curriculum remains the source of truth.

This document expands the existing curriculum without changing its claims, sequence, duration, philosophy, or core deliverables.

1. The course teaches judgment, not memorization.

Students should not simply memorize the 6Ds. They should learn to defend a phase classification with data, caveats, and strategic implications.

1. Every session produces visible work.

Each session ends with a concrete artifact, draft, or decision so the course compounds rather than resets every 90 minutes.

1. Data literacy is built into the course, not outsourced.

Students must gather, clean, and interpret real trend data. Facilitators may scaffold the search process, but should not remove the search challenge entirely.

1. Personal strategy is required.

The course is not just about technology forecasting. It culminates in a personal technology thesis tied to career, venture, or learning decisions.

1. Assessment rewards evidence-backed reasoning over certainty.

A cautious, well-supported thesis should outperform a confident but weakly grounded prediction.

1. The website aesthetic influences delivery.

The course should feel like "scientific journal meets data art": precise, analytical, visually legible, and centered on inflection points, dashboards, and curve-reading.

1. AI/LLM evaluation is advisory but structured.

LLMs may assess evidence quality, reasoning structure, rubric alignment, and feedback language, but should be instructed not to invent missing data or pretend certainty about factual claims not present in the student submission.

3. Delivery model assumptions

Format

• 3 weeks
• 6 live sessions
• 90 minutes per session
• Recommended cadence: 2 sessions per week

Target learner

• Ages 15-25
• Mixed backgrounds
• No prerequisite technical training required

Recommended cohort size

• Ideal: 12-24 students
• Minimum viable: 6 students
• Maximum before extra facilitation support: 30 students

Facilitation model

• 1 lead facilitator
• 1 support facilitator or teaching assistant recommended for cohorts above 18
• AI support may be used for rubric scoring, draft feedback, and presentation-note analysis

Learning environment

• In-person, hybrid, or live online
• Students need a laptop or tablet for research and drafting
• Whiteboard or digital whiteboard recommended
• Spreadsheet access recommended
• Presentation format can be slides, poster, Figma board, or single-page memo

Core instructional modes

• Short lecture
• live demo
• guided analysis
• small-group discussion
• workshop time
• presentation and critique

Non-negotiable course outputs

• One technology curve analysis using 10+ years of data
• One 10-year curve map visual
• One personal technology thesis document
• One final presentation with peer questioning

Recommended instructional norms

• Students must show data sources
• Students must separate facts from predictions
• Students may be wrong on the conclusion if the reasoning is explicit and evidence-based
• Facilitators should push for specificity: years, cost trends, inflection points, adoption thresholds, and strategic actions

4. Detailed course content broken down by module, session, lesson, activity, timing, facilitator moves, and student outputs

Module 1 / Session 1: The Linear Trap

Session objective: Students understand why exponential change feels counterintuitive, can distinguish linear from exponential growth, and begin classifying real-world examples.

Session outputs

• Linear vs. exponential classification sheet
• Initial personal reflection on one technology the student may want to track through the course

Session 1 agenda

Facilitator notes for Session 1

• Do not over-index on math. The goal is intuition plus pattern recognition, not formal modeling.
• If students confuse volatility with exponentiality, separate those concepts clearly.
• Keep repeating: "What does the shape suggest?" and "What evidence would make us change our minds?"

Common misconceptions to address in Session 1

• "Fast growth automatically means exponential."
• "If a curve starts flat, it is not important."
• "Price spikes and crashes are the same as exponential progress."
• "Exponential means infinite."

Module 2 / Session 2: The 6Ds Framework

Session objective: Students can explain each D in the 6Ds framework, apply the framework to concrete technologies, and identify where classifications are ambiguous.

Session outputs

• 6Ds phase-mapper worksheet
• One argued phase classification for a chosen technology

Session 2 agenda

Facilitator notes for Session 2

• When students classify a technology, ask "What would disconfirm your classification?"
• Require evidence language such as cost trend, adoption change, performance improvement, regulatory milestone, or user behavior shift.
• Keep the sequence practical. Students should walk away able to use 6Ds as a working heuristic, not as doctrine.

Acceptable evidence for phase classification

• Cost declines over time
• Performance gains over time
• Adoption or usage growth
• Regulatory or infrastructure milestones
• Substitution against incumbents
• Removal of a previously paid layer
• Consolidation of multiple physical products into one digital product
• Expansion of access across income or geography

Module 3 / Session 3: Mapping Real Curves Workshop

Session objective: Students gather and plot real data, identify a technology's current phase, and justify where the curve changed slope or meaning.

Session outputs

• Draft technology curve graph with 10+ years of data
• Phase classification memo
• 3-minute mini-presentation

Session 3 agenda

Workshop protocol for Session 3

1. Choose one technology.
2. Choose one primary metric.
3. Gather at least 10 years of data.
4. Plot the data.
5. Mark likely 6D-relevant transitions.
6. Write a short explanation of current phase.
7. Identify what evidence would move the classification forward or backward.

Facilitator moves during Session 3

• If a student picks a metric that cannot show phase change, redirect early.
• If a student has only marketing claims, require a more reliable dataset.
• If a student has a beautiful chart but weak interpretation, push on meaning.
• If a student has strong reasoning but messy visuals, reassure them that clarity matters more than polish at this stage.

Module 4 / Session 4: Where Are We Now?

Session objective: Students compare multiple technology curves in the present, analyze venture implications by phase, and sharpen their own thesis through comparison.

Session outputs

• Comparative curve dashboard notes
• Opportunity-by-phase worksheet
• Thesis direction statement

Session 4 agenda

Facilitator notes for Session 4

• The goal is not to predict the future perfectly. The goal is to learn what a serious thesis sounds like.
• Encourage students to compare multiple curves, but require one anchor curve for their final work.
• Use the phrase "What becomes newly possible here?" repeatedly.

Module 5 / Session 5: Personal Thesis Workshop

Session objective: Students convert curve analysis into a personal technology thesis with a visual map, strategic implications, and actionable positioning choices.

Session outputs

• Draft 10-year curve map visual
• Draft personal technology thesis document
• Draft action plan tied to thesis

Session 5 agenda

Required elements of the personal technology thesis

• Chosen technology and why it matters
• Primary curve metric and data basis
• Current 6D classification
• Evidence for current classification
• What changed over the last 5 years
• What the next phase would look like
• Personal exposure or opportunity
• 2-3 concrete actions to take now
• Major risks, decelerators, or reasons the thesis could be wrong

Module 6 / Session 6: Final Presentations and Socratic Dialogue

Session objective: Students present a coherent technology thesis, defend it under questioning, and show they can connect curve reading to action.

Session outputs

• Final presentation
• Final thesis document
• Final curve map
• Response notes from peer questioning

Session 6 agenda

Final presentation structure

1. The technology and why it matters
2. The evidence curve
3. The current 6D phase and why
4. The next likely transition
5. The D6 world that could emerge
6. The student's personal or venture positioning
7. One thing that could make the thesis wrong

5. Assignments and artifacts

Assignment 1: Linear vs. Exponential Classification

When: Session 1

Purpose: Build foundational pattern recognition.

Submission format: Worksheet or short form response

Required artifact

• Classify provided scenarios as linear, exponential, or insufficient evidence
• Give one sentence of justification per item

Assignment 2: 6Ds Phase Mapper

When: Session 2

Purpose: Build fluency with the framework.

Submission format: Worksheet

Required artifact

• Classify 20 technologies into a current phase
• Include at least one evidence statement per classification
• Mark 3 items as high uncertainty and explain why

Assignment 3: Technology Curve Analysis

When: Draft in Session 3, submit after Session 4

Purpose: Build real data interpretation skill.

Submission format: Graph plus 500-900 word memo

Required artifact

• 10+ years of data
• One clearly labeled metric
• Source list
• Annotated graph
• Current phase classification
• Short defense of where the curve inflected

Assignment 4: Opportunity-by-Phase Worksheet

When: Session 4

Purpose: Connect curve-reading to venture and career opportunity.

Submission format: Structured worksheet

Required artifact

• One D2 opportunity
• One D3 opportunity
• One D4 or D5 opportunity
• One D6 implication
• One personal relevance statement

Assignment 5: 10-Year Curve Map

When: Draft in Session 5, finalize before Session 6

Purpose: Convert analysis into a clear visual forecast and decision tool.

Submission format: Slide, poster, whiteboard photo, Figma board, or PDF

Required artifact

• X-axis from now to 10 years
• Y-axis defined clearly
• Historical reference point
• Current point
• Projected future point
• Annotations for disappearing and emerging businesses
• Markers for current phase and next likely phase

Assignment 6: Personal Technology Thesis

When: Draft in Session 5, final in Session 6

Purpose: Produce a strategic document that connects trend analysis to action.

Submission format: 700-1200 word memo or 6-10 slide deck

Required artifact

• Thesis statement
• Evidence summary
• 6D phase classification
• Forecast logic
• Personal exposure or advantage
• Concrete next actions
• Risks and uncertainty

Assignment 7: Final Presentation and Socratic Defense

When: Session 6

Purpose: Test synthesis, clarity, and defensibility.

Submission format: Live presentation

Required artifact

• 5-minute presentation
• 3-5 minutes of Q&A defense
• Final deck or visual

6. AI/LLM grading and assessment framework

Assessment principles

• Grade the reasoning the student shows, not the worldview the evaluator personally prefers.
• Reward explicit uncertainty handling.
• Penalize invented evidence, unsupported phase claims, and vague strategic recommendations.
• Separate writing quality from analytical quality unless communication clarity is itself part of the rubric.
• Treat the curriculum's 6Ds framework as the grading lens, while allowing thoughtful edge cases.

Recommended grading workflow

1. Human or system operator verifies that the submission includes the required components.
2. LLM performs rubric-based scoring and evidence extraction.
3. Human reviews flagged issues, especially factual uncertainty, possible hallucinations, or unusually high/low scores.
4. Final feedback is released with both numeric scores and narrative comments.

What the LLM is allowed to evaluate well

• Whether the student states a clear thesis
• Whether evidence is actually cited or described
• Whether phase claims are defended logically
• Whether the forecast follows from the evidence presented
• Whether strategic actions connect to the thesis
• Whether uncertainty and counterarguments are acknowledged

What the LLM should not pretend to know

• Whether a student's external factual claim is true if the evidence is not included
• Whether a current technology is objectively in one exact phase when the student has made a defensible ambiguous case
• Whether a specific market forecast will actually happen

Core scoring dimensions for all major artifacts

Concrete assessment heuristics for LLM evaluation

Heuristic 1: Metric integrity

• Full credit if the student uses one primary metric consistently and explains why it represents the curve.
• Partial credit if the student mixes metrics but still maintains a defensible story.
• Low credit if the student shifts between cost, adoption, and performance without explanation.

Heuristic 2: Evidence sufficiency

• Full credit if the student includes at least 10 years of data or a clearly justified shorter window for emerging technologies, plus identifiable sources.
• Partial credit if the data exists but is thin, incomplete, or weakly sourced.
• Low credit if the student relies on anecdotes, headlines, or unsupported claims.

Heuristic 3: Phase specificity

• Full credit if the student identifies a current phase and explains at least two reasons for that classification.
• Partial credit if the phase is plausible but under-argued.
• Low credit if the phase is asserted without evidence or contradicts the student's own chart.

Heuristic 4: Inflection-point reasoning

• Full credit if the student identifies a likely inflection point or slope change and explains what changed.
• Partial credit if an inflection is suggested but not tied to evidence.
• Low credit if the student never addresses where change became meaningful.

Heuristic 5: D6 imagination quality

• Full credit if the student explains what broad access would unlock in behavior, business model, or social reality.
• Partial credit if the D6 world is named but generic.
• Low credit if the student cannot articulate downstream implications.

Heuristic 6: Actionability

• Full credit if the student gives concrete next moves with timing and rationale.
• Partial credit if actions are sensible but generic.
• Low credit if recommendations read like slogans.

Heuristic 7: Intellectual honesty

• Full credit if the student names uncertainties, decelerators, or disconfirming evidence.
• Partial credit if uncertainty is acknowledged superficially.
• Low credit if the submission is overconfident, one-sided, or hype-driven.

Flag conditions for human review

• Submission contains confident factual claims with no visible evidence
• Student appears to have copied generic AI-generated phrasing without specific course artifacts
• Score discrepancy of more than 20 percentage points between rubric dimensions
• Submission is eloquent but analytically empty
• Submission is analytically strong but written unclearly enough that the LLM may be underscoring it

7. Rubrics, scoring criteria, and evaluator prompt guidance

Recommended course-weighted rubric

Aligned to the base curriculum:

Detailed scoring criteria by component

A. Technology Curve Analysis (30 points)

B. Technology Thesis Document (30 points)

C. 10-Year Curve Map Visual (20 points)

D. Presentation and Peer Questioning (20 points)

Performance bands

Evaluator prompt guidance for LLM use

Recommended system prompt

You are evaluating student work for a course called "Exponential Curves Mapping."
Use only the student's submitted materials and the rubric provided.
Do not invent missing evidence.
Do not reward confidence unless it is supported by data or explicit reasoning.
It is acceptable for a student to present uncertainty or ambiguity if they explain it clearly.
Score each rubric criterion independently, then provide a total score, short rationale, and targeted feedback.
If evidence is missing, say it is missing rather than guessing.
If a claim seems factually questionable but cannot be verified from the submission, flag it for human review.
```

#### Recommended evaluator user prompt template

Evaluate the following student submission for the course "Exponential Curves Mapping."

Apply this rubric: [paste relevant rubric section]

Required output format:

1. Completeness check
2. Criterion-by-criterion scores with one-sentence justification each
3. Total score
4. Three strengths
5. Three improvement priorities
6. One paragraph of student-facing feedback
7. Human-review flags, if any

Student submission: [paste submission]

#### Recommended structured output schema

```json
{
  "completeness": {
    "status": "complete | partial | incomplete",
    "missing_items": []
  },
  "scores": {
    "criterion_name": {
      "points_awarded": 0,
      "points_possible": 0,
      "rationale": ""
    }
  },
  "total_score": {
    "points_awarded": 0,
    "points_possible": 100
  },
  "strengths": [],
  "improvement_priorities": [],
  "student_feedback": "",
  "human_review_flags": []
}

Prompting guardrails for evaluator consistency

• Instruct the LLM to quote or paraphrase the student's actual evidence when possible.
• Instruct the LLM to distinguish between "missing evidence" and "bad reasoning."
• Instruct the LLM not to penalize unconventional conclusions if the evidence chain is coherent.
• Instruct the LLM to keep tone direct, specific, and non-generic.
• Instruct the LLM to recommend next revisions that the student could actually complete.

8. Feedback strategy: what strong/average/weak responses look like and how an LLM should respond

Feedback principles

• Feedback should identify what the student did, not just how it felt.
• Feedback should reference evidence quality, phase reasoning, and strategic implication.
• Feedback should end with an actionable next step.
• Feedback should avoid generic praise such as "great job" unless attached to a specific observation.

What strong responses look like

Characteristics

• Uses a clear metric and relevant data window
• Shows source awareness
• Defends phase classification with multiple signals
• Identifies a plausible inflection point
• Distinguishes present evidence from future projection
• Connects thesis to a real action plan
• Acknowledges uncertainty without becoming vague

How the LLM should respond to strong work

• Validate the strongest analytical move specifically
• Point out 1-2 ways to sharpen the thesis further rather than rewriting the whole project
• Encourage stronger boundary conditions or disconfirming evidence

Example LLM feedback pattern for strong work

• "Your strongest move is tying the cost curve to a specific phase transition rather than just naming the phase."
• "The forecast is persuasive because you separate what the data shows from what you expect next."
• "To strengthen this further, define what evidence would tell you the technology is still in D2 rather than D3."

What average responses look like

Characteristics

• Has the right structure but thin evidence
• Chooses a plausible phase but under-explains it
• Includes some charting but weak annotation
• Makes strategic recommendations that are reasonable but generic
• Mentions uncertainty but does not integrate it meaningfully

How the LLM should respond to average work

• Name what is already working
• Identify the single missing layer that would improve the submission most
• Avoid overwhelming the student with too many revision targets

Example LLM feedback pattern for average work

• "You have a workable thesis and a plausible phase classification."
• "The main thing holding this back is evidence density: your chart needs a clearer metric definition and stronger source trail."
• "Revise by adding two concrete data points that explain why you believe the inflection has already happened."

What weak responses look like

Characteristics

• Makes broad claims with little or no evidence
• Treats the 6Ds as labels without reasoning
• Confuses hype, volatility, and exponential improvement
• Uses vague future language with no timing or assumptions
• Offers generic advice not linked to the chosen curve
• Omits sources, annotations, or required artifacts

How the LLM should respond to weak work

• Be direct about what is missing
• Do not soften core analytical problems with empty encouragement
• Prioritize foundational fixes in sequence
• Tell the student exactly what to add next

Example LLM feedback pattern for weak work

• "Right now the submission names a phase, but it does not yet prove that phase with evidence."
• "Your next revision should start with one metric, one chart, and one source trail."
• "Do not add more predictions until the current-state analysis is grounded."

Recommended LLM response style by performance level

Sentence starters the LLM can use productively

• "Your evidence most strongly supports..."
• "The weakest part of the current argument is..."
• "This phase classification becomes more convincing when you..."
• "A stronger version of this thesis would..."
• "The most important missing evidence is..."
• "Your action plan aligns with the thesis when..."
• "A reasonable counterargument is..."

Final evaluator instruction for feedback generation

When generating feedback, the LLM should:

1. State the current quality level in plain language.
2. Cite one concrete strength.
3. Cite one concrete analytical gap.
4. Give the highest-leverage revision step.
5. End with a forward-looking next action.

Implementation notes for operators

• If this course is run online, require submission templates in advance to reduce formatting chaos.
• If AI grading is used, keep a human in the loop for final presentation scoring and any submission with uncertain factual grounding.
• Update the "Where Are We Now?" technology table before each cohort, as the curriculum already instructs.
• Preserve the course's central tension: most people miss the future because they misread the shape of the curve.