Multi-Agent AI Washing Index Panel Data for Chinese A-share Listed Firms, 2015-2024,anonymized

Published: 22 September 2026| Version 1 | DOI: 10.17632/br8kkzhcjt.1
Contributor:
shuyang feng

Description

This dataset provides anonymized firm-year panel measurements of AI washing among Chinese A-share listed companies from 2015 to 2024. It contains structured scores, qualitative classifications, adversarial multi-agent evaluation records, and verification evidence extracted from annual reports and firm-level AI capability indicators. The dataset is designed for academic research on corporate AI disclosure, technology narrative inflation, digital transformation, capital market communication, and AI-related greenwashing-style behavior. The AI Washing Index, also named Hype Index in the data file, measures the degree to which a firm's AI-related disclosure appears overstated relative to its disclosed strategy, technical foundation, and investment support.

Files

Categories

Economics, Digital Transformation

Licence