Human-AI Scenario Planning (HASP) is a structured method for exploring plausible future operating environments and testing whether candidate choices remain viable as conditions change. It does not forecast a single future. It makes assumptions, uncertainty, evidence, stakeholder consequences and trade-offs visible so that they can be challenged through dialogue.

HASP sits between the Human-AI Knowledge Base (HAK) and Strategic Thinking. It converts current organisational knowledge and futures intelligence into a mind-sized set of three scenario worlds, then uses those worlds to examine candidate options through the distinct lenses of Prosperity, Planet and People. Learning and early signals return to organisational awareness and the Knowledge Base so the scenarios and choices can be revisited.

Purpose, Ethics & Governance bound the method. They establish the conditions within which choices may be considered and provide the basis for absolute MUST criteria. A choice that breaches these conditions does not remain in comparison merely because it performs well elsewhere.

The dominant flow is:

The Human-AI Knowledge Base provides the foundational evidence: purpose and governance, stakeholders and desired outcomes, capabilities and dependencies, value chains, information, current performance, constraints, risks, and relevant external conditions. HASP is grounded in this knowledge but deliberately tests its boundaries rather than treating current assumptions as fixed.

Futures Intelligence broadens the field of view before writing scenarios. Delphi draws on distributed expert judgement without forcing premature consensus. Seven Questions / Issues surface concerns, threats, opportunities, unpreparedness and assumptions that may prove wrong. Horizon Scanning adds trends, weak signals, emerging issues, discontinuities and potential shocks. AI can synthesise this material, but minority views and contradictions must remain visible.

Critical uncertainties are selected from forces that are both highly important and genuinely uncertain. We use different combinations to construct three plausible, internally coherent futures. These should not be optimistic, expected and pessimistic versions of the same story. Each should challenge different assumptions and create meaningfully different risks, opportunities and operating conditions.

The scenario set uses a consistent structure: name, time horizon, concise overview, critical assumptions, key drivers, economic and market conditions, social and workforce conditions, technology and information, planetary and resource conditions, political and regulatory conditions, and implications for the organisation. The narrative must be rich enough to imagine living and working within it, yet concise enough to remain in working memory during assessment.

Personas and Day in the Life narratives connect each future to the people who must live and work within it. They focus on decision-relevant roles, desired outcomes, responsibilities, knowledge, influence, constraints, opportunities and unacceptable consequences. Their purpose is not fictional decoration; it is to make lived effects on stakeholders, capability and the Common Good visible.

Coherence testing and early signals keep the scenarios disciplined. AI may identify conflicting assumptions, unsupported causal links, dependency conflicts, missing second-order consequences and inconsistencies between narratives and persona experience. Humans decide whether an issue is a genuine inconsistency, an uncertainty or a deliberate assumption. Observable early signals then connect the scenarios back to organisational awareness and indicate when assumptions or choices should be reconsidered.

Evaluation criteria are agreed before scoring any option. Prosperity examines durable value and the economic, operational and organisational capability required to continue fulfilling purpose. Planet examines ecological limits, biodiversity, climate and energy, circular resource use, water, pollution and waste. People examines dignity, wellbeing, equity, accessibility, meaningful work, trust, community and partnership. These dimensions remain separate, so strength in one cannot conceal weakness in another.

The MUST gate comes first. Test each option against every mandatory condition in each scenario. PASS allows comparative scoring to proceed. FAIL means the option is unacceptable in its current form and must be removed or redesigned. FAIL is not a score of 1: a score of 1 indicates achievement is highly difficult but still plausibly possible. Default MUSTs include compliance with purpose, ethics, governance and law; protection of fundamental human rights and dignity; no unacceptable health or safety harm; no loss of biodiversity where affected; and no irreversible or unacceptable environmental harm.

Comparative scoring begins only after the MUSTs pass. Each comparative criterion receives a relative importance weight from 1 to 10. Each option is then scored from 1 to 5 for achievability under the particular scenario, and every score requires a concise rationale. The weighted contribution is Score x Weight. Each dimension score is the sum of weighted contributions divided by the sum of weights for that dimension, retaining the common 1-5 scale. There is deliberately no overall Triple Bottom Line score.

The Scenario Diagram plots Prosperity, Planet and People as a radar profile for each option under each scenario. Options are arranged vertically and scenarios horizontally. This makes robustness, sensitivity, imbalance, dependency and potential collapse visible. The purpose is not to select the largest triangle. The shape prompts human challenge, interpretation, and judgement.

The HASP workbook operationalises the method through seven linked sheets: Scenario Set, Personas, Evaluation Criteria, Option Scoring, Scenario Summary, Scenario Diagram and Decision Notes. It keeps context close to the scorer, applies the MUST gate, calculates weighted dimension scores, updates the radar charts, and records rationale, disagreement, evidence gaps, learning, and actions.

Human-AI dialogue is the decision process, not an optional layer around the spreadsheet. AI may propose scenarios or initial scores, find relevant evidence, test coherence, identify contradictions and stakeholder consequences, and expose evidence gaps or optimism. Humans remain responsible for approving the scenario set, agreeing criteria and weights, assigning final scores, interpreting the profiles and deciding what to do. Relevant learning feeds back into HAK, HASP, Strategic Thinking, Design Thinking, and Organisational Awareness.

The method is governed by several non-negotiable principles:

• Scenarios are plausible alternative futures, not predictions.
• Significant disagreement and minority perspectives are preserved rather than averaged away.
• Criteria and weights are agreed before scoring so the framework is not changed to engineer a preferred result.
• A failed MUST cannot be compensated for by performance elsewhere.
• The rationale is as important as the number because it exposes the assumptions behind the score.
• Prosperity, Planet and People remain separate; no single total conceals imbalance.
• The diagram supports dialogue and learning; it does not make the decision.

Methodological foundations. HASP synthesises established foresight, decision-analysis, human-rights and sustainability principles rather than reproducing any one source. Principal foundations used in the guideline include:

• UK Government Futures Toolkit – critical uncertainties, scenario construction, personas, Day in the Life narratives and early warning signals.
• Shell Scenarios – credible, relevant and challenging alternative stories, supported by modelling and coherence testing rather than prediction.
• Kepner-Tregoe Decision Analysis – separation of mandatory objectives from weighted comparative objectives and relative-importance weighting.
• UN Guiding Principles on Business and Human Rights – the responsibility to avoid causing or contributing to adverse human-rights impacts.
• Future-Fit Business Benchmark – distinction between minimum environmental thresholds and optional positive contributions.
• Adapt, Survive and Flourish – Scenario Planning – the Scenario Diagram tradition of making comparative profiles visible to support learning and dialogue.

HASP should therefore be understood as an adaptive organisational learning and judgement system, not a calculation procedure. Its output is not a winning score; it is a visible account of viability, imbalance, dependency, uncertainty and the evidence on which judgement currently rests.

The spreadsheet calculates. People interpret, judge and decide. AI helps them see what they may have missed.

For a fully example spreadsheet:

NaturFlourish HASP | Scenario Set