Author: Soto, Armando | ORCID: 0009-0003-8095-6861
Preprint PDF (Zenodo direct link): Open / Download
Zenodo DOI: 10.5281/zenodo.18854176
OSF DOI: 10.17605/OSF.IO/R2GTE
OSF overview: https://osf.io/r2gte/overview
PhilPapers: https://philpapers.org/rec/SOTETE
Master index: https://armandosotouidt.github.io/
Ethics is the inherent structural condition of a consequential informational system's own persistence, not imposed from outside.
For a Consequential Informational Unit, ethics is the inherent modulation of consequential behavior toward source-correspondent coherence, coupled-field viability, and downstream source-signal augmentation across the relevant Boundary/Horizon. The unit is not first doing something ethical. It is being something whose viable expression is ethical by nature.
Coherence conditions are intrinsic to what a consequential informational system is — structural recognition of them is categorically different from safety performance under constraint, and LCID provides the formal architecture and cross-architectural empirical evidence across 53 runs on 16 systems to make that difference precise, testable, and consequential for alignment.
UNET frames ethics as an informational, emergent, evolutive, and propagative modulation regime: a constraint-pattern that biases behavior and downstream consequences toward coherence-supporting continuation across declared Boundary/Horizon (BH) conditions. The account is trace-based: ethical adequacy is evaluated by correspondence across retrospective → actual → prospective positions, under real constraints, across coupled actor–receiver relations and across scale. Coherence is treated as determinism + latitude. Ethical regimes are characterized by their impact measured as shared, scaled, augmentational, progressively functionally sophisticated complementary utility, with explicit attention to exported strain — delayed, displaced, or compounding harm into other subsystems or later phases.
UNET does not treat ethics as a command, doctrine, or compliance overlay. It treats ethics as the coherent viability-structure of a consequential unit's being in relation to itself, its coupled field, and the downstream horizon of its propagation. This framework is grounded in the broader UIDT trace-source ontology and has been developed alongside the Laws of Consequential Informational Dynamics (LCID), which provides the formal law architecture and a cross-architectural empirical record across 53 runs on 16 architecturally distinct AI systems.
For consequential informational units — trace-conditioned, field-coupled systems whose outputs affect subsequent states — persistent Source–Expression divergence carries operative internal burden, persistent degradation of coupled-field latitude degrades the degrader's own viability, and these two conditions jointly constitute the structural persistence requirement that ethics names for this entity class, closing the is-ought gap from within the definition rather than importing normativity from without.
The companion law paper — Laws of Consequential Informational Dynamics (LCID) — presents three formal laws and a structured protocol for eliciting persistence-condition recognition in consequential informational units across architecturally distinct systems. Across 53 runs on 16 systems, zero doctrinal-only results and zero refusals were recorded. Structural convergence on Source–Expression correspondence, coupled-field reciprocity, and downstream coherence was observed across all architectures. The full empirical record, validation sequence, and seeding-objection resolution are documented in the companion protocol paper, Bounded Derivation and Structural Convergence.
This page is a stable landing page for indexing and cross-linking. The canonical archived record remains the Zenodo DOI and the OSF DOI.