Redundancy-as-Masking: Formalizing the Artificial Age Score (AAS) to Model Memory Aging in Generative AI

Seyma Yaman Kayadibi^*

• Victoria University, Australia, Melbourne, Australia

Artificial intelligence can exhibit aging-like patterns not as a function of chronological time, but through systematic asymmetries in output-level observable memory performance under different context-persistence conditions. In large language models, semantic cues, such as the name of the day, often remain stable across sessions, while episodic details, like the sequential progression of experiment numbers, tend to collapse when conversational context is reset in the 25-day bilingual recall protocol. To capture this phenomenon, the Artificial Age Score (AAS) is introduced as a log-scaled, entropy-informed metric of memory age derived from observable recall behavior and defined purely at the output level, without access to internal latent-state representations. The score is formally proven to be well-defined, bounded, and monotonic under mild and model-agnostic assumptions, supporting mathematical reuse across evaluation settings while not implying empirical generalization across models or domains. In its Redundancy-as-Masking formulation, the score interprets redundancy as overlapping information that reduces the penalized mass. However, in the present study, redundancy is not explicitly estimated; all reported values assume a redundancy-neutral setting (R = 0), yielding conservative upper bounds. The AAS framework was tested over a 25-day bilingual study involving ChatGPT-5.0, structured into stateless and persistent interaction phases. During persistent sessions, the model consistently recalled both semantic and episodic details, driving the AAS toward its theoretical minimum, indicative of behavioural youth in recall. In contrast, when sessions were reset, the model preserved semantic consistency but failed to maintain episodic continuity, causing a sharp increase in the AAS and signaling an aging-like behavioural signature of continuity loss in recall behavior. These results are interpreted behaviorally and do not constitute evidence about internal memory mechanisms or latent-state dynamics. These findings support the utility of AAS as a theoretically grounded, task-independent diagnostic tool for evaluating memory degradation in artificial systems. The empirical validation reported here is limited to this specific model version and protocol; applications to other architectures or training regimes require revalidation. The study builds on foundational concepts from von Neumann’s work on automata, Shannon’s theories of information and redundancy, and Turing’s behavioral approach to intelligence.