Jon W. Hansen, FCIPS · Procurement Insights | Hansen AI Labs™ · October 2026
Within two days this week, two independent applications of my work came to light: one in a European economics journal, the other in an independent analysis of AI decision tools. Both were written by people with no connection to Hansen Models™, and neither is a mention or a repost. In both cases, the idea was understood, credited and applied.
An application in economic forecasting
In May 2026, Vitalii Opanasiuk and Iurii Mohylevskyi published “Critical Convergence as a Post-Singular Phenomenon: Formalization, Economic Effects, and Methodological Challenges of Forecasting” in the Kyiv Economic Scientific Journal (Issue 13, pages 184–189).
Journal record · Full paper · DOI: 10.32782/2786-765X/2026-13-23
The authors identify Strand Commonality™ by name, credit its origin to this archive, and cite the May 2025 Procurement Insights post Are Black Swan Events Real, or Are They a Product of Hindsight? They describe the theory as holding that economic, political and social strands intersect to create patterns that can be tracked, and they use it to argue that some events called black swans are better understood as grey swans: events of high probability whose precise timing remains uncertain. They then apply that reasoning to COVID-19, the war in Ukraine and the 2026 Iran conflict, arguing that each had observable precursors.
The paper draws on Nassim Nicholas Taleb’s work on black swans, Ray Kurzweil and Nick Bostrom on technological change and risk, and Daron Acemoglu on economics and institutions. Strand Commonality™ appears in that same discussion.
Strand Commonality™ was built to explain procurement outcomes, but nothing in it is confined to procurement. Two researchers have now applied it to systemic economic forecasting, to explain what conventional models tend to miss. A theory becomes useful when others, with no stake in it, can use it without its author in the room.
An application in AI decision governance
On 4 October 2026, Bertrand Maltaverne published Why I care more about an AI’s confidence score than its speed in Procurement Tidbits. [Link to the post]
In it, Bertrand credits me with a distinction I made in a comment exchange: a model can confidently choose among the answers it was given while being unable to recognize that the decision was framed incorrectly, or that the right answer was never among the options. He carries that distinction into his own assessment of an AI decision tool, and concludes that a confidence score cannot correct a defective frame.
A model can be highly confident inside the wrong frame. Measuring how fast or how confident an AI is cannot tell you whether the decision was framed correctly. Thank you to Bertrand for carrying that distinction into his evaluation: validate the frame before trusting the score.
Why it matters
The same body of work came to light in two very different places in the same week: a forecasting paper concerned with wars and pandemics, and an analysis of AI tools for procurement teams. That should not be surprising. Strand Commonality™ holds that outcomes are shaped where economic, political, social and technological strands cross. The principle is not confined to procurement; it applies wherever those crossings occur. Independent applications in two such different fields are what you would expect to see.
What this means for AI use in procurement
Procurement teams are being asked to trust AI tools on the strength of their speed and their confidence. Both citations point to the question that has to come first: what frame is the AI working inside, and does it reflect how the operation actually runs?
A tool can only choose among the options it was given, using the data it was given. If the decision depends on a strand outside that frame, such as a supplier’s ownership change, a policy shift or a workaround the system never recorded, the tool will not see it, and its confidence will not reveal the gap. The answer is not to distrust AI. It is to establish the frame before relying on the answer: trace how the operation actually runs, including the strands that cross it, and then let the tool work inside a frame that is true.
Our thanks to Vitalii Opanasiuk, Iurii Mohylevskyi and Bertrand Maltaverne for the care with which they read the work, and for the accuracy with which they applied it.
The original post cited by the Kyiv paper:
Truth Is Believing. Accuracy Is Knowing. Outcome Is Proof.™
-30-
piblogger
October 26, 2026
Correction, 26 October 2026: The Acemoglu work cited in the Kyiv paper is The Race Between Man and Machine (with Pascual Restrepo, American Economic Review, 2018), on technology, automation and economic growth, not institutions, as this post states. Note also that the paper is published in Ukrainian; the journal’s website can be switched to English using the language option on its page.