Before AI Assurance: Fiscal Geometry, Evidence Routing, and Sustainability Disclosure Credibility
AI保証の前に：財政幾何学、エビデンスルーティング、およびサステナビリティ開示の信頼性 (AI 翻訳)
Yongzhi Huang
🤖 gxceed AI 要約
日本語
本稿は、サステナビリティ開示の信頼性を評価するためのフレームワーク「Fiscal Geometry」を提案する。AIによる自動検証だけでは不十分で、まず開示請求が構造化された証拠ルーティングプロセスを通る必要があると論じる。開示請求が主張から証拠分類、信頼性評価へと移行する過程を明確にし、AIが開示情報を処理する前に証拠経路を理解する必要性を強調する。
English
This paper develops Fiscal Geometry as an evidence-routing framework for sustainability disclosure credibility in AI-enabled assurance. It argues that before AI can assure sustainability claims, claims must be routed through a structured evidentiary process. The framework reframes assurance as dependent on prior classification, documentary grounding, and institutional transmission, highlighting credibility failures before they become assurance failures. The contribution is conceptual, providing a basis for AI governance and institutional accountability.
Unofficial AI-generated summary based on the public title and abstract. Not an official translation.
📝 gxceed 編集解説 — Why this matters
日本のGX文脈において
日本のSSBJ開示や統合報告書において、サステナビリティ情報の信頼性は重要な課題である。本稿のエビデンスルーティング枠組みは、AIを活用した保証の前に開示請求の証拠を体系的に整理する方法を提供し、日本企業の開示品質向上に寄与する可能性がある。
In the global GX context
Globally, as ISSB and CSRD mandate sustainability disclosures, the credibility of those disclosures becomes critical. This paper's evidence-routing framework offers a pre-assurance discipline that can strengthen disclosure integrity in AI-assisted environments, relevant to regulators, firms, and assurance providers worldwide.
👥 読者別の含意
🔬研究者:This paper provides a conceptual framework for studying the intersection of AI, evidence, and sustainability disclosure credibility, valuable for scholars working on assurance and AI governance.
🏢実務担当者:Corporate sustainability teams can use the evidence-routing concept to structure their disclosure processes and prepare for AI-based assurance.
🏛政策担当者:Policymakers shaping assurance standards may consider the need for evidentiary classification and institutional grounding before relying on AI for verification.
📄 Abstract（原文）
This working paper develops Fiscal Geometry as an evidence-routing framework for assessing sustainability disclosure credibility in AI-enabled assurance environments. The paper argues that the credibility of sustainability claims cannot be secured by automated verification alone. Before AI systems can evaluate, summarize, or assure sustainability information, claims must first be routed through a structured evidentiary process that determines what counts as admissible support, how evidence is classified, how institutional context affects credibility, and where disclosure risks arise. The paper introduces evidence routing as a pre-assurance discipline. It connects sustainability disclosure, AI governance, institutional credibility, and assurance logic by showing how disclosure claims move from assertion to evidentiary classification and then to credibility assessment. Rather than treating sustainability assurance as a final-stage verification exercise, the paper reframes assurance as a process that depends on prior classification, documentary grounding, and institutional transmission. The contribution is conceptual and methodological. It provides a framework for identifying credibility failures before they become assurance failures, especially in systems where AI may process sustainability information without understanding the evidentiary route behind a disclosure claim. The paper is intended as part of a broader research programme on Fiscal Geometry, AI governance, evidence routing, and institutional accountability.
🔗 Provenance — このレコードを発見したソース
- openalex https://doi.org/10.5281/zenodo.20734726first seen 2026-07-15 05:13:27 · last seen 2026-07-15 05:13:28
🔔 こうした論文の新着を逃したくない方は キーワードアラート に登録（無料・3キーワードまで）。
gxceed は公開メタデータに基づく研究支援データセットです。要約・翻訳・解説は AI 支援で生成されています。 最終的な解釈・検証は利用者が原典資料に基づいて行うことを前提とします。