Judging Category
Basic or Experimental Research
Student Rank
Graduate
College
Business
Faculty Sponsor
Nanying Lin, nlin@astate.edu
Description
Textbook finance theories indicate that investors demand risk premia from risky assets as compensation for risk and for hedging against future unfavorable economic states. We design a comprehensive study to examine whether investment advice generated by artificial intelligence (AI) reflects a coherent risk–return pattern in investment decision-making. We find that AI advises investors to increase stock investments when return increases and reduce stock allocations when stock volatility increases. However, AI ignores the hedging role of the stock market with consumption growth. A low risk aversion cannot explain this failure to respond to consumption-related risk. We further find that AI provides slightly more rational investment advice for male investors than for female investors, and that a more advanced model (ChatGPT 5) exhibits more rational behavior than an earlier-generation model (ChatGPT 4.1). Our results suggest that AI provides economically meaningful investment advice but remains less efficient than predictions implied by textbook finance theories.
Disciplines
Finance and Financial Management
License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.
Recommended Citation
Gilbert, Oscar O. and Chu, Tianxiang, "How Rational Is AI Investment Advice? Risk-Return Relevance in Artificial Intelligence (AI) Investments" (2026). Create@State. 21.
https://arch.astate.edu/evn-createstate/2026/posters/21
Included in
How Rational Is AI Investment Advice? Risk-Return Relevance in Artificial Intelligence (AI) Investments
Textbook finance theories indicate that investors demand risk premia from risky assets as compensation for risk and for hedging against future unfavorable economic states. We design a comprehensive study to examine whether investment advice generated by artificial intelligence (AI) reflects a coherent risk–return pattern in investment decision-making. We find that AI advises investors to increase stock investments when return increases and reduce stock allocations when stock volatility increases. However, AI ignores the hedging role of the stock market with consumption growth. A low risk aversion cannot explain this failure to respond to consumption-related risk. We further find that AI provides slightly more rational investment advice for male investors than for female investors, and that a more advanced model (ChatGPT 5) exhibits more rational behavior than an earlier-generation model (ChatGPT 4.1). Our results suggest that AI provides economically meaningful investment advice but remains less efficient than predictions implied by textbook finance theories.
