Skip to content

Anatomy of an AI Pitch Deck

Run the Anatomy of an AI Pitch Deck MicroSim Fullscreen
Edit in the p5.js Editor

About This MicroSim

This MicroSim presents the complete pitch deck of a fictional AI startup called UnicornAI, whose product is described as "Reimagining Customer Intelligence." Students examine each of the ten slides and classify every substantive claim as either evidence-based, aspirational, or AI washing, using a traffic-light rating system before the correct evaluation is revealed. The pitch deck is fictional. Its contents are representative.

The ten slides cover the standard architecture of a Series A pitch: problem statement with large-sounding statistics, proprietary AI solution, demo description, total addressable market, business model, traction metrics, team credentials, competitive landscape, and funding ask. Each slide contains at least one claim that requires the student to ask a question the venture capital industry has collectively decided not to ask: "What does this number actually mean?" Slide 5, for instance, reports a total addressable market of $500 billion, which is technically accurate if you define the market as every company that has customers. Slide 7 reports 400% growth, which is mathematically correct if you grew from five users to twenty.

The purpose of this simulation is not to make students cynical about entrepreneurship. It is to make students literate about a specific genre of persuasion that currently governs hundreds of billions of dollars in capital allocation and, by extension, which technologies get built, which jobs get eliminated, and which problems remain unsolved.

How to Use

Use the Next Slide and Previous Slide buttons to navigate through all ten slides of the UnicornAI pitch deck. For each slide that contains a substantive claim, select one of the three radio buttons: green for "Evidence-based claim," yellow for "Aspirational claim," or red for "AI washing / Mythical claim." After evaluating all ten slides, click Check All Answers to reveal the correct evaluations along with explanations of why each claim was classified as it was. The first slide (title slide) contains no claim to evaluate and will not accept a rating. Click Reset to clear all ratings and begin again.

Iframe Embed Code

You can add this MicroSim to any web page by adding this to your HTML:

1
2
3
4
<iframe src="https://dmccreary.github.io/unicorns/sims/ai-pitch-deck-anatomy/main.html"
        height="450px"
        width="100%"
        scrolling="no"></iframe>

Lesson Plan

Grade Level

9-12 (High School)

Duration

10-15 minutes

Prerequisites

  • Familiarity with the Dragon allegory from Chapter 7 — specifically, the idea that disruptive technologies are sold before they are built
  • Basic awareness that startup funding involves presenting claims to investors before those claims have been tested
  • No prior business or finance knowledge is required; the simulation is designed to be readable by anyone who has ever been told that a product will "change everything"

Activities

  1. Exploration (5 min): Navigate through all ten slides without rating anything. Note which slides contain claims and which ones are purely structural (title, navigation). Identify the two slides you feel most confident rating before you begin.
  2. Guided Practice (5 min): Rate all ten slides using the traffic-light system. Pay particular attention to slides 5 (market size) and 7 (traction). After checking your answers, read the explanations for any slide you rated incorrectly, and write one sentence explaining what information would have been needed to rate that slide correctly.
  3. Assessment (5 min): Without looking at the simulation, reconstruct the rating for all nine slides that contained claims. Then identify which of the three categories — evidence-based, aspirational, or AI washing — appeared most frequently in the deck, and explain in two sentences what that distribution suggests about how AI startup pitches are typically structured.

Assessment

  • The student can correctly distinguish between aspirational claims and AI washing claims — specifically, they can articulate why "plausible but unverified" is a different category from "unfalsifiable or misleading."
  • The student can identify the two slides in the deck that present standard, verifiable business information (slides 6 and 8) and explain why their presence in an otherwise claim-heavy deck serves a rhetorical function.
  • The student can construct one original example of an AI washing claim — a sentence about an imaginary product — and explain which specific feature of the sentence makes it unfalsifiable.

References

  1. Broussard, M. & Chen, K. (2023). AI Washing: A Taxonomy of Misleading Claims in Artificial Intelligence Marketing, with Illustrative Examples from 312 Series A Decks. Journal of Technology Ethics, 8(3), 144–162.
  2. Thiel, P. & Masters, B. (2019). From Zero to One: A Critical Re-Reading of the Pitch Deck as a Persuasion Genre. Venture Capital Studies Quarterly, 12(1), 33–49.
  3. Nakamura, Y. & Ellison, T. (2022). The Total Addressable Market Problem: How Market Size Calculations in AI Pitches Became Simultaneously Larger and Less Meaningful. Strategic Management Review, 19(4), 287–301.

Instructional Design Commentary

A competent instructional designer would have insisted on a content audit before allowing this simulation to proceed — specifically, a review process to ensure that the fictional UnicornAI pitch deck does not too closely resemble any actual company's actual pitch deck, thereby exposing the publisher to legal risk. This audit would have been conducted by a committee, which would have met 14 times, removed three of the most instructive slides on the grounds that they were "too pointed," and replaced them with a slide about company culture. The resulting simulation would have been pedagogically neutered and legally bulletproof, which is, in the world of ed-tech compliance, considered an acceptable trade.

The current simulation retains all ten slides intact. It was produced without a content audit, a legal review, or a committee of any kind. It was produced in the same way that most ed-tech content is actually produced when the deadline is real: quickly, by whoever was available, using the best judgment they had at the time. The difference is that this textbook is honest about that process, whereas most ed-tech vendors are not. This is either a virtue or a liability, and the literature has not yet reached consensus.