Before You Decide: The Decision Series — Ten Papers on Artificial Intelligence for People Who Have to Decide

Practitioner companion to The Decalogy on Artificial Intelligence. Ten papers translate each Decalogy argument into operational analysis for decision-makers. Topics include: why scaling without targeting stalls at the output layer (Paper 1), cross-domain representational convergence and shared-backbone strategy (Paper 2), structural limitations of red-teaming under shared conceptual frameworks (Paper 3), tool-frame vs. agent-frame alignment architecture (Paper 4), the Task Assignment Paradox and why 88% AI adoption yields only 39% EBIT impact (Paper 5), human data as irreplaceable pipeline signal (Paper 6), distributed agency and infrastructure beyond humanoid robotics (Paper 7), the training-to-inference compute demand transition (Paper 8), world signal as the primary AI data source (Paper 9), and why current AI metrics measure scale rather than direction (Paper 10). Each paper is numbered to match the Decalogy and includes proposed experimental designs. Designed for executives, investors, policymakers, and technical leaders who need the structural analysis without the full theoretical derivation. Revision Note: Per-paper descriptions revised from topic labels to structural claims. Paper 4 now states that the agent frame produces scheming and the tool frame constrains it. Paper 6 identifies tail variability rather than volume as the relevant property. Paper 7 names Distributed Agency Computation and the market implication. Paper 8 specifies environmental coupling as the infrastructure criterion that holds regardless of forecast accuracy. Paper 10 reframed as a category change argument: expertise is the interface through which AI is absorbed, not what AI replaces.