The\u00a0previous post\u00a0<\/a>in this series introduced a general framework for AI-assisted scenario building: Force Blank, Penalize Guessing, Show the Source. The framework produces output where every claim is tagged as VERIFIED, ASSUMED, or PROJECTED, and where gaps are explicitly labeled instead of silently filled.<\/p>\n That’s the\u00a0what<\/em>. This post is about the\u00a0how<\/em>\u00a0\u2014 three practical challenges that anyone implementing the framework will encounter:<\/p>\n Tagging an assumption is necessary but not sufficient.\u00a0 Four calibration methods work well with the tagged output:<\/p>\n Daniel Kahneman and Amos Tversky’s distinction between the “inside view” (planning based on the specifics of this project) and the “outside view” (what happened in similar projects historically) is the single most useful concept for calibrating assumptions. The planning fallacy \u2014 systematically underestimating costs and timelines \u2014 is so well-documented that the American Planning Association\u00a0officially endorsed reference class forecasting<\/a>\u00a0in 2005 as a corrective.<\/p>\n In practice, this means: for every ASSUMED tag, ask the model (or yourself) to identify 3\u20135 comparable situations and their actual outcomes. If you assume 15% growth, what growth did similar products in similar markets actually achieve? If you assume a 6-month regulatory timeline, how long did comparable approvals actually take? The tagged format makes this step natural \u2014 you have a list of assumptions; now walk down it with an outside view on each one.<\/p>\n You can even build this into the prompt:<\/p>\n For every ASSUMED tag, add a “Calibration” note: identify 2\u20133 comparable historical cases and their actual outcomes. If no comparable data exists, note [NO REFERENCE CLASS].<\/em><\/p><\/blockquote>\n Not all assumptions are equally important. RAND’s Assumption-Based Planning calls this “criticality” \u2014 an assumption is critical if its failure would require fundamental changes to the plan. In practice, this means testing: what happens to the conclusion if this assumption is 50% wrong? If the answer is “not much,” the assumption is low-priority. If the answer is “the entire business case collapses,” that’s your highest-priority validation target.<\/p>\n The tagged format enables this directly. You can ask the model:<\/p>\n Take the three ASSUMED items with the highest downstream impact on the final projection. For each, recalculate the projection with the assumption at 50% of stated value and at 150%. Show me which assumptions the conclusion is most sensitive to.<\/em><\/p><\/blockquote>\n Gary Klein’s pre-mortem technique inverts the question: instead of asking “will this work?”, you start from “it failed \u2014 why?” This is particularly effective for ASSUMED tags, because it surfaces failure modes that optimism hides. Ask the model:<\/p>\n Assume this scenario failed after 12 months. Which of the ASSUMED items were most likely the point of failure? For each, describe a plausible narrative of how that assumption broke down.<\/em><\/p><\/blockquote>\n Assumptions have shelf lives. A market size estimate from a 2025 Gartner report is still reasonable in 2026. A competitive landscape assumption from 2024 may already be wrong. Adding a temporal dimension to ASSUMED tags helps:<\/p>\n For each ASSUMED tag, add an expiry estimate: how long is this assumption likely to remain valid? Mark anything older than 12 months or based on pre-2025 data as [STALE ASSUMPTION].<\/em><\/p><\/blockquote>\n The framework works well when one person uses it in one chat session. The governance question is: how does it survive contact with an organization \u2014 multiple people, multiple AI tools, multiple documents, over months?<\/p>\n What typically happens: an analyst generates a beautifully tagged scenario. They copy it into a slide deck. The tags disappear. A manager reads the deck, sees “Year 1 revenue: \u20ac310K” with no indication that the number is PROJECTED from two unvalidated ASSUMED inputs. The ghost scenario lives again.<\/p>\n This is a knowledge management problem, not an AI problem. And it has knowledge management solutions.<\/p>\n The simplest governance mechanism is a template. If your organization uses AI for scenario planning, the output template should have provenance columns built in. Not optional, not “add if useful” \u2014 structurally required. A scenario document without source tags should be treated the same way as a financial report without citations: incomplete.<\/p>\n Concretely: create a standard table format for all AI-assisted scenario outputs:<\/p>\n\n
\n1. Calibrating Assumptions: From “Tagged” to “Tested”<\/h2>\n
(ASSUMED: market grows 15% annually)<\/code>\u00a0is better than an unlabeled 15% baked into the projection \u2014 but it still doesn’t tell you whether 15% is defensible. The framework surfaces assumptions; calibration tests them.<\/p>\nReference Class Forecasting: The Outside View<\/h3>\n
Sensitivity Testing: What Breaks If This Is Wrong?<\/h3>\n
Pre-Mortem: Imagine It Failed<\/h3>\n
Temporal Decay: When Does the Assumption Expire?<\/h3>\n
\n2. Governance: Making the Framework Stick Beyond One Person’s Prompt<\/h2>\n
The Problem: Tags Die in Translation<\/h3>\n
Level 1: Template Enforcement<\/h3>\n