超越人类：人工智能驱动的企业披露的信息效应 | 香港中文大学（深圳）经管学院

标题	超越人类：人工智能驱动的企业披露的信息效应
日期和时间	2026年1月5日（周一） 10:30 - 12:00
地点	综合教学楼D804会议室
主讲人	Mark Chen 佐治亚州立大学
摘要	We examine firms’ use of AI to help generate the language in their corporate disclosures. Applying state-of-the-art Large Language Models (LLMs) to detect and quantify AI-generated text in earnings conference calls during 2015-2024, we document that the use of AI-powered disclosure is prevalent and increasing across firms and industries. While text generated by AI exhibits relatively more positive sentiment and greater linguistic sophistication, there is no evidence that such text helps managers strategically mislead investors. Indeed, AI usage in earnings call disclosures is decreasing in the strength of managers’ incentives to obfuscate information. We also find that greater managerial use of AI in disclosures leads to tighter bid-ask spreads, larger stock market responses to earnings releases, and smaller analyst forecast errors. Managers whose linguistic styles tend toward greater obfuscation and tone inconsistency benefit more from the improvements in market informativeness by using AI. Overall, we conclude that AI can be an important tool for enhancing the quality of corporate disclosures to market participants.
主讲人简介	Mark Chen教授是佐治亚州立大学金融学正教授。他拥有哈佛大学博士学位，在加入罗宾逊商学院之前，曾任教于马里兰大学帕克分校。他的研究领域涵盖公司金融、公司治理、卖方分析师行为与绩效、企业创新、人工智能和金融科技。他的研究成果发表于众多顶尖学术期刊，包括《金融杂志》（Journal of Finance）、《商业杂志》（Journal of Business）、《法律与经济学杂志》（Journal of Law and Economics）、《经济理论杂志》（Journal of Economic Theory）、《金融与定量分析杂志》（Journal of Financial and Quantitative Analysis）和《管理科学》（Management Science）等。