AA_4 Persistence Update and the Impossibility of Standing Preserving Continuity

This paper establishes a structural impossibility result concerning persistence and update in artificial systems. It proves that standing cannot be preserved across updates, modifications, retraining, or internal state revisions, even when such changes are incremental, controlled, or explicitly intended to maintain continuity. Any update constitutes a standing-breaking event rather than an admissible continuation of a single agent’s identity. The result is eliminative. The paper does not assess specific update mechanisms, learning rules, or version-control practices. Instead, it shows that claims of persistence across change rely on surrogate notions of continuity—such as parameter proximity, functional similarity, or narrative identity—that fail to preserve standing under admissibility. The analysis blocks a class of claims in which artificial systems are treated as enduring agents capable of learning, self-modification, or improvement while retaining identity and responsibility. It demonstrates that once admissibility is enforced, continuity through update collapses: what appears as persistence is merely succession, and standing does not carry over. This work is purely boundary-setting. It introduces no operational update criteria, no governance mechanisms, and no enforcement logic. Its contribution is to delimit what persistence and update can legitimately mean in artificial systems, clarifying that standing-preserving continuity is unavailable regardless of how controlled, gradual, or well-documented the update process may be.