人工知能とESG開示の品質：インド上場企業からの証拠

AI and ESG Disclosure in Indian Listed Companies | Submission-oriented draft ARTIFICIAL INTELLIGENCE AND THE QUALITY OF ESG DISCLOSURE: EVIDENCE FROM INDIAN LISTED COMPANIES (原題)

Garv Nijhara

Crossrefプレプリント2026-01-01#AI×ESG経営インパクト: 調達リスク対象セクター: cross_sector

DOI: 10.2139/ssrn.7489218

原典: https://doi.org/10.2139/ssrn.7489218

🤖 gxceed AI 要約

日本語

AI能力がESGデータ分析能力を介してESG開示の品質（完全性・比較可能性・検証可能性・意思決定有用性）を高める統合モデルを提示する。取締役会の技術ガバナンスを調整変数とし、インド上場企業を対象にSmartPLSによる高次PLS-SEMとパネル固定効果で検証する設計。SEBIのBRSR義務化・BRSR Core保証制度を背景に、開示の「量」ではなく「質」に着目する点が貢献。

English

This study proposes and tests an integrated model in which AI capability raises ESG disclosure quality via ESG data-analytics capability, with board technology governance as a moderator. It targets Indian listed companies under SEBI's mandatory BRSR and phased BRSR Core assurance regime, using higher-order PLS-SEM plus panel fixed effects. Its contribution is a transparent, India-specific framework linking AI capability to disclosure quality rather than volume; current results rest on a synthetic pilot dataset.

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

インドBRSRは日本企業が参照する新興国開示制度の代表例であり、SSBJ・有報開示の高度化やAI活用による開示データ整備を検討する日本企業・投資家にとって比較材料となる。AI×ESG開示の実証設計は国内の統合報告書高度化にも示唆を与える。

In the global GX context

As mandatory sustainability reporting spreads (ISSB, CSRD, SEC climate), this paper addresses a gap in global disclosure scholarship: whether AI capability improves disclosure quality, not just volume, under a quantitative regime like India's BRSR. It offers a replicable higher-order SEM design for linking technology governance to disclosure outcomes across jurisdictions.

👥 読者別の含意

🔬研究者:AI能力・取締役会技術ガバナンスとESG開示品質を結ぶ高次PLS-SEM設計と媒介・調整モデルの実証テンプレートとして参考になる。

🏢実務担当者:AI・データ分析投資をESG開示品質向上（BRSR・SSBJ対応）に結びつける社内ガバナンス設計の示唆を得られる。

🏛政策担当者:開示制度の量的拡充だけでなく、企業のAI活用とデータ検証可能性を促す保証・ガバナンス政策の論点を提供する。

📄 Abstract（原文）

Artificial intelligence (AI) is increasingly embedded in corporate decision systems, risk management, analytics, customer processes and sustainability management, while investors and regulators are simultaneously demanding more comparable and decision-useful environmental, social and governance (ESG) information. This study develops and tests an integrated model explaining how AI capability can influence the quality of ESG disclosure by strengthening firms' ESG data-analytics capability. It further examines whether board technology governance conditions the AI-ESG disclosure relationship. The empirical architecture is designed for Indian listed companies because the Securities and Exchange Board of India (SEBI) has progressively institutionalised quantitative sustainability reporting through the Business Responsibility and Sustainability Report (BRSR), with mandatory BRSR reporting for the top 1,000 listed entities from financial year 2022-23 and a phased BRSR Core assurance regime thereafter. The model conceptualises AI capability as a higher-order construct comprising strategic AI orientation, AI applications, AI investment, AI governance and AI-related human capital; ESG disclosure quality is modelled as a higher-order construct capturing environmental, social and governance disclosure quality, with emphasis on completeness, comparability, verifiability and decision usefulness. ESG data-analytics capability is specified as a mediator and board technology governance as a moderator. A SmartPLS-based higher-order PLS-SEM design, complemented by panel fixed-effects and robustness tests, is proposed. A 120 firm-year synthetic pilot dataset is supplied solely to demonstrate the analytical workflow; its illustrative results indicate positive direct and mediated associations and provide benchmark diagnostics for the final archival study. The principal contribution is a transparent, India-specific empirical framework linking AI capability with the quality-not merely the volumeof ESG disclosure under an evolving regulatory regime.

🔗 Provenance — このレコードを発見したソース

openaire https://doi.org/10.2139/ssrn.7489218first seen 2026-11-05 04:40:45

🔔 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。