{"graphSha256":"f29d7b1eedeb88387238d2abda347f810d2179c5fdbead47c3ef48507ab20ca5","node":{"id":"T385","type":"ficha","title":"AI Integrity Architecture: Toward Expert-System Envelopes Around Statistical AI","url":"https://dakleyer.github.io/perspective-beyond/sources/ai-integrity-architecture-toward-expert-system-envelopes-around-statistical-ai-t385/","original":"https://tegrity.ai/ai-integrity-architecture-toward-expert-system-envelopes-around-statistical-ai/","aliases":[],"clusters":["tegrity"],"tags":["evidencia","integridad-ia"],"summary":"","commit":null,"sourcePath":null,"commitUrl":null,"documentId":"doc-fa687f604847765f99","tagWeights":{"evidencia":0.3636,"integridad-ia":0.6364},"reading":"The integration of artificial intelligence based on statistical and foundational models into mission-critical systems is giving rise to the emergence of a new architectural “integrity” layer, whose purpose is to impose constraints, guarantee explainability, and maintain traceability under increasingly strict regulatory and safety requirements. This layer adopts patterns very close to those of classical expert systems: explicit knowledge, deterministic inference, auditable rules, and structured explanation capability. This article analyzes the regulatory and technical trends that push toward constrained and explainable architectures, reviews the notion of expert systems, and argues that modern AI guardrails in mission-critical environments are, in essence, expert systems that wrap and condition probabilistic models.","readingSource":{"url":"https://tegrity.ai/ai-integrity-architecture-toward-expert-system-envelopes-around-statistical-ai/","section":"Opening paragraphs","observed":"2026-11-04","sourceSha256":"6e4315aa75fe68e5291631c3da13eeae293e01d90ec77c3adb890f9184fc7f49"},"readingSummary":"The integration of artificial intelligence based on statistical and foundational models into mission-critical systems is giving rise to the emergence of a new architectural “integrity” layer, whose purpose is to impose constraints, guarantee explainability, and maintain traceability under increasingly strict regulatory and safety requirements.","contextTokens":235},"relations":[["apoyo:T385:C59","T385","C59","documenta el concepto","",null,"apoyo",""],["apoyo:T385:C86","T385","C86","documenta el concepto","",null,"apoyo",""],["documento:F017","T385","T1391","documented use","A perfectly traceable rule chain cannot recover a decision-relevant distinction that input formation has already removed.",null,"documento","La limitación es condicional a recibir la misma representación; no afirma que ninguna tecnología pueda adquirir nueva evidencia o rediseñar la captura."],["documento:R016","E1704","T385","editorial comparison","For a complementary reading, see [Architecture for operational AI integrity](https://tegrity.ai/ai-integrity-architecture-toward-expert-system-envelopes-around-statistical-ai/).",null,"documento",""],["documento:R050","E1570","T385","editorial comparison","For a complementary reading, see [Expert-system envelopes around statistical AI](https://tegrity.ai/ai-integrity-architecture-toward-expert-system-envelopes-around-statistical-ai/).",null,"documento",""],["documento:R106","N1826","T385","editorial comparison","For a complementary reading, see [Expert-system envelopes and authorization boundaries](https://tegrity.ai/ai-integrity-architecture-toward-expert-system-envelopes-around-statistical-ai/).",null,"documento",""],["documento:R163","T888","T385","editorial comparison","For a complementary reading, see [The architecture beneath AI Integrity Management](https://tegrity.ai/ai-integrity-architecture-toward-expert-system-envelopes-around-statistical-ai/).",null,"documento",""],["documento:R164","T385","T387","editorial comparison","For a complementary reading, see [The structural limits of integrity under regime change](https://tegrity.ai/structural-limits-of-current-ai-integrity-under-regime-change/).",null,"documento",""]]}