Deepseek Facts : Evidence from the Selected Markets

Abstract

This research paper focuses on how the introduction of DeepSeek AI in January 2025 affects the stock market in the United States of America, specifically, the companies of the New York Stock Exchange (NYSE) and the Nasdaq. The research utilizes an event study, which is based on a CAPM model, to determine whether the announcement under question results in the abnormal returns that cannot be explained by the normal market risk. The multiple events windows on the day of announcement were used to investigate stock returns of the sampled companies on the day of announcement. CAPM was applied to determine the expected returns and it gave the variance of abnormal returns and cumulative abnormal returns. These results show that the levels of abnormal returns and volatility increases after the introduction of the DeepSeek AI. This means that the market reaction could not have been ideally described using the conventional risk reward models. It was discovered that markets were market specific to AI shocks since technology heavy markets reacted more than diversified markets. Overall, the findings suggest the inefficiency of CAPM under the conditions of technology disruption occurring too rapidly and highlight the part played by event specific and behavioral determinants in the short term market reactions.