Mapping the Accounting–Information Systems Frontier: A Research Agenda for Machine-Readable Disclosure
会計・情報システムのフロンティアをマッピングする:機械可読開示の研究アジェンダ (AI 翻訳)
Idorenyin J. Okon, Olabamiji Atanda, Adeolu O. Adewuyi
🤖 gxceed AI 要約
日本語
機械可読開示は主要資本市場で標準化しつつあるが、LLMの登場、IFRS S1/S2や欧州単一電子フォーマットによるサステナビリティ開示のデジタル化、新興市場へのシフトが既存のコンセンサスを揺るがしている。2006〜2026年の362件の研究をPRISMAに基づきレビューし、MRDスタックという独自フレームワークで分析。生成AIやサステナビリティ報告に関する共引用クラスターは存在せず、生産条件・インフラ・機械オーディエンスの研究が不足していることを示す。
English
Machine-readable disclosure has become the default in major capital markets, but LLMs, IFRS S1/S2 digitalization, and a shift to emerging markets unsettle prior consensus. A PRISMA review of 362 studies (2006-2026) using the MRD Stack framework finds no co-citation cluster for generative AI or sustainability reporting, and underdeveloped research on production conditions, infrastructure, and the machine audience. The value of machine-readable disclosure rests on institutions and audiences, not just technology.
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
日本ではSSBJ開示や有報のデジタル化が進む中、機械可読開示のインフラ整備とAI活用は重要な論点。本レビューは、日本の開示実務が国際標準(IFRS S1/S2)やESEFに対応する際の研究基盤を提供し、今後の研究課題を示唆する。
In the global GX context
Globally, this review addresses the intersection of AI and sustainability disclosure, highlighting gaps in infrastructure and machine audience research. It informs regulators and standard-setters (ISSB, ESMA) about the need to study the institutional and capacity aspects of machine-readable disclosure, especially in emerging markets.
👥 読者別の含意
🔬研究者:Provides a comprehensive research agenda and framework (MRD Stack) for studying machine-readable disclosure, highlighting gaps in AI and sustainability reporting.
🏢実務担当者:Offers insights into the evolving landscape of digital disclosure, helping corporate teams prepare for AI-driven analysis and sustainability reporting requirements.
🏛政策担当者:Highlights the importance of institutions and capacities for effective machine-readable disclosure, relevant for regulators designing digital reporting mandates.
📄 Abstract(原文)
Machine-readable disclosure has matured from a technical pilot into the default format of corporate reporting in the world's major capital markets. Three changes now underway, namely the arrival of large language models capable of reading untagged disclosures, the digitalisation of sustainability disclosure under IFRS S1/S2 and the European Single Electronic Format, and a shift in the empirical base of the field toward emerging markets, unsettle the consensus the literature had reached. We conduct a PRISMA-guided systematic literature review of 362 peer-reviewed studies indexed in Scopus between 2006 and 2026, combining bibliometric performance analysis, science mapping, and structured content analysis under an original five-layer framework, the Machine-Readable Disclosure (MRD) Stack. We find that the field's intellectual base is anchored in the formative United States XBRL regime and contains no co-citation cluster for either generative AI or sustainability reporting. Standards and consequences are well developed; production conditions, infrastructure, and the machine audience are not. The literature has built an apparatus for an audience it named but did not study and has generalised from one well-enforced mandate in one market. We argue that the value of machine-readable disclosure has rested less on the technology the field has built than on the institutions, capacities, and audiences around it, and we offer this thesis as falsifiable rather than rhetorical, with frontier markets as the natural experimental setting. The synthesis yields an integrated forward agenda of eight research questions tied to the MRD Stack, with implications for regulators, preparers, intermediaries, and emerging-market policymakers.
🔗 Provenance — このレコードを発見したソース
- openaire https://doi.org/10.2139/ssrn.7004918first seen 2026-08-04 04:51:37
🔔 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。