There is a newer version of the record available.

Published January 1, 2026 | Version v13
Preprint Open

Semantic Identity Drift in Decision Systems under Formal Certification

Description

This paper provides a purely formal separation between operator-based semantic drift and predicate-based certification in decision systems. It introduces semantic identity drift as a structural phenomenon that is independent of observable behavior, correctness, or certification success.

The contribution is negative and conceptual in character. It establishes that neither behavioral stability nor certification predicates suffice to characterize the preservation of semantic identity over time. The analysis is intentionally abstract: no assumptions are made about observability, metrics, signals, telemetry, detection procedures, or system architectures.

The paper serves as a formal anchor that clarifies which inferences about semantic stability are not justified under standard correctness and certification frameworks. It does not propose operational methods, audit tools, or governance mechanisms. Any empirical or signal-based treatment of drift lies strictly outside the scope of this work.

This version is archived as part of an ongoing research line on semantic stability and meaning-state drift in autonomous and decision-making systems.

#semantic drift #decision identity #formal methods #certification #non-implication results #audit interpretation #semantic stability

 

Files

Semantic_Identity_Drift_in_Decision_Systems_under_Formal_Certification.pdf

Files (155.2 kB)

Additional details

Dates

Created
2025-04-01
Initial formal release of the SnapOS audit model and semantic traceability framework (version 1.0).

References

  • Chaitin, G. J. (1977). Algorithmic Information Theory. IBM Journal of Research and Development, 21(4), 350–359. https://doi.org/10.1147/rd.214.0350
  • Håstad, J. (1987). Computational Limitations of Small-Depth Circuits. Ph.D. thesis, MIT. http://hdl.handle.net/1721.1/14958
  • Gödel, K. (1931). Über formal unentscheidbare Sätze der Principia Mathematica. Monatshefte für Mathematik und Physik, 38, 173–198. https://doi.org/10.1007/BF01700692
  • Floridi, L. (2011). The Philosophy of Information. Oxford University Press. ISBN: 9780199232383
  • Souly, A. et al. (2025). Poisoning Attacks on LLMs Require a Near-Constant Number of Samples. Alan Turing Institute / Anthropic / UK AI Security Institute. arXiv:2508.03114.
  • Bowen, D. et al. (2024). Data Poisoning in LLMs: Jailbreak-Tuning and Scaling Laws. arXiv:2408.02946.