Abstract
This paper studies how investors use generative AI in discussions across two major investing social media platforms with distinct governance and user bases: Seeking Alpha and Reddit's r/WallStreetBets. We document large cross-platform differences in adoption patterns and their associated market outcomes. On Seeking Alpha, AI adoption is more prevalent when information is scarce or when contributors cover unfamiliar stocks. AI article sentiment positively predicts returns and is associated with greater information discovery: more informative retail order flow, reduced user disagreement, and narrower bid-ask spread. On WallStreetBets, in contrast, AI adoption tends to rise following surges in retail buying and is linked to sentiment contagion. Adoption is also followed by higher abnormal trading volume, greater volatility, and more lottery-like return distributions. These results indicate that both the adoption of AI and its relationship with market outcomes depend upon the institutional and behavioral context in which the technology is deployed.
| Original language | English |
|---|---|
| Number of pages | 80 |
| Publication status | Published - 5 Dec 2025 |
| Event | NBER Big Data, Artificial Intelligence, and Financial Economics Conference - Charles View Ballroom, Hyatt Regency, 575 Memorial Drive, Cambridge, United States Duration: 5 Dec 2025 → … https://www.nber.org/conferences/big-data-artificial-intelligence-and-financial-economics-fall-2025 |
Conference
| Conference | NBER Big Data, Artificial Intelligence, and Financial Economics Conference |
|---|---|
| Country/Territory | United States |
| City | Cambridge |
| Period | 5/12/25 → … |
| Internet address |
User-Defined Keywords
- Generative AI
- Large Language Models
- Social Media
- Financial Disclosures
- Retail Investors
- Information Frictions
- Market Microstructure