The Decalogy on Artificial Intelligence: Ten Papers in Four Movements

This volume collects ten papers written in two phases. The first phase produced the Transition Trilogy (Papers 1–3): a foundation paper on the nature of human cognition, an empirical paper on disciplinary convergence under AI methodology, and a methodological paper on AI-assisted research. The second phase produced Papers 4–10: papers on the definition of AI, workflow, human value, the robot, physical substrates, the world as signal environment, and the Silicon Era. The ten papers argue toward three theses. Thesis One: the foundational definitions have been wrong — intelligence, AI, the robot, and the Silicon Era have all been defined as systems that exist separately from the world and bridge that separation through technology. All four definitions were wrong in the same direction. Correcting them changes the conclusions. Thesis Two: the transition is already underway. What appears to be a future state — AI as the epistemic core of every field, convergence of methodologies, the collapse of the power-access law — is a present condition. Thesis Three: AI does not separate humans from the world. It integrates them more deeply into it, and that integration is expansion. The three theses are not independent claims. They are the same insight at three angles: definitional (Thesis One), historical (Thesis Two), directional (Thesis Three). The Independence Illusion — the common error structure named in Paper 10 — is what all three theses correct.

REVISION NOTES (v7):

Papers 2, 4, and 6 receive deferred inserts: an independent-survival analysis for the convergence claim (Paper 2), engagement with Christiano, Bai, Russell, and Ouyang distinguishing specification risk from execution risk (Paper 4), and an alternative pathway for the model collapse argument that does not require the deficiency-substrate interpretation (Paper 6).

Papers 7–9 receive volume strengthening: three new DAC cases including Knight Capital and Stuxnet (Paper 7), IBD as a recurring technology classification pattern plus neuromorphic and biological interface evidence (Paper 8), and specific benchmarks for GraphCast, ocean/seismic/ecological AI, and an AlphaFold WSS reframing (Paper 9).

Paper 10 was restructured. The Definitional Cascade was compressed from seven subsections to one, preserving the schema-layer argument. The freed weight was redirected to an independent Kardashev critique and an expanded Planetary Synchronization Rate section with sub-civilizational testability and a Fermi paradox reinterpretation. An abstract and a Series Preface update were added. Compilation errors in the prior version have been corrected.