Abstract
Thesis. For complex personal and professional decisions, AI is most useful not as a source of answers but as a tool for finding what you do not know: it helps separate facts from assumptions, expand the set of options, find weak points in a plan, and design cheap tests before a mistake becomes costly or irreversible.
What is proposed. An operational framework of five loops (Bifurcation Engine), in which each loop provides material for the next, while the right to decide and the responsibility remain with the human. The framework builds on known techniques: pre-mortem, red team, value of information theory, the distinction between reversible and irreversible decisions, and a prediction journal.
What is not claimed. It is not claimed that the method has been shown to improve decision outcomes. Its components have their own evidence base, but their composition has not been tested experimentally. Section 13 describes a plan for such a test; Appendix D collects the method’s boundaries in one place.
What is new in this edition. Fully international wording (English, USD); no regional localization of contacts, currency, or legal references. “Quick start”: applicability triage and a minimal first pass in a few minutes. Three levels of description of the method (loops, Master Prompt stages, Decision Brief fields) are unified in one table at the end of Section 4. A single rule on the number of alternatives: three to five additional ones. Causes of mismatch between forecast and fact are described by a unified classification of five types, including “chance” (Section 10.2); the forecast in the journal is recorded by the human, not the model. The case includes a journal entry (Section 5.4); the Decision Brief has a “Decision taken” field.
Stages 3, 5, and 8 of the Master Prompt and the order of steps in Section 9 are clarified (the human fixes forecast, confidence, and thresholds before the test starts). The wording on source verification is aligned with what was actually checked.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Kotegov, Volodymyr, "BEFORE YOU CHOOSE: Find the Fact That Can Change the Decision", Technical Disclosure Commons, ()
https://www.tdcommons.org/dpubs_series/12011