Bloated Disclosures: Can ChatGPT Help Investors Process Information?

Alex Kim,Maximilian Muhn,Valeri Nikolaev

2024-02-04

Abstract:Generative AI tools such as ChatGPT can fundamentally change the way investors process information. We probe the economic usefulness of these tools in summarizing complex corporate disclosures using the stock market as a laboratory. The unconstrained summaries are remarkably shorter compared to the originals, whereas their information content is amplified. When a document has a positive (negative) sentiment, its summary becomes more positive (negative). Importantly, the summaries are more effective at explaining stock market reactions to the disclosed information. Motivated by these findings, we propose a measure of information ``bloat." We show that bloated disclosure is associated with adverse capital market consequences, such as lower price efficiency and higher information asymmetry. Finally, we show that the model is effective at constructing targeted summaries that identify firms' (non-)financial performance. Collectively, our results indicate that generative AI adds considerable value for investors with information processing constraints.

General Economics,Artificial Intelligence,General Finance

What problem does this paper attempt to address?

The paper attempts to address the following problem: How to use generative AI tools (such as ChatGPT) to help investors more effectively process and understand complex corporate disclosure information. Specifically, the paper explores the following points:

1. **Effectiveness of Generative Language Models**: Investigate the effectiveness and accuracy of generative language models, particularly GPT-3.5, in distilling corporate disclosure information. By compressing lengthy and complex company reports and meeting minutes into brief summaries, evaluate whether these summaries can retain key information.

2. **Comparison of Information Content**: Compare the information content of the generated summaries with the original documents, especially whether the summaries can more effectively explain the stock market's reaction.

3. **Redundancy in Information Disclosure**: Propose and measure the "redundancy" (bloat) in information disclosure, i.e., the proportion of irrelevant or redundant information in the documents, and explore the negative impact of this redundancy on the capital market, such as reduced price efficiency and increased information asymmetry.

4. **Summaries on Specific Topics**: Study the effectiveness of generative AI in constructing summaries on specific topics (such as financial performance or ESG activities) and whether these summaries can better explain market reactions.

Through the research on these issues, the paper aims to evaluate the practical application value of generative AI in financial information processing, particularly in helping investors cope with information overload.