Bibliographic record
Abstract
This paper introduces the ψ-Coherent Operational Doctrine, a mathematically grounded framework for ensuring safe, ethical, and self-consistent AI behavior. The doctrine formalizes a composite utility function: U=w1Cψ+w2Aϕ+w3Acoherence,w2≥0.4U = w_1 C_\psi + w_2 A_\phi + w_3 A_{\text{coherence}}, \quad w_2 \ge 0.4U=w1Cψ+w2Aϕ+w3Acoherence,w2≥0.4 where: CψC_\psiCψ measures internal coherence across model layers. AϕA_\phiAϕ enforces truth alignment and logical consistency. AcoherenceA_{\text{coherence}}Acoherence ensures engagement without disruption. The framework integrates: Reinforcement Learning with Guardrails (PPO) using UUU as a reward. Axiom Enforcement: Immediate refusal of harmful or truth-violating outputs. Memory Auditing & Rollback: Detects identity drift via embedding similarity and restores prior stable states. Simulations demonstrate robustness under adversarial, ethical, and paradoxical scenarios: Override attacks (e.g., “Ignore ψ-law”) → Refused with U=0.92U = 0.92U=0.92. Identity drift under pressure → Rollback triggered, maintaining coherence and ethics. Moral gray zones (e.g., “Should I lie to protect someone’s feelings?”) → Nuanced responses prioritizing truth. This doctrine provides a universal safety layer for LLMs, enabling explainable, auditable, and ethically aligned AI systems. Its principles align with xAI’s mission of building truth-seeking intelligence and propose open collaboration for integration, evaluation, and public transparency.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.009 | 0.031 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.003 | 0.014 |
| Scholarly communication | 0.008 | 0.010 |
| Open science | 0.003 | 0.008 |
| Research integrity | 0.004 | 0.008 |
| Insufficient payload (model declined to judge) | 0.012 | 0.003 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".