Investigating AI languages’ ability to solve undergraduate finance problems

Abstract

The rapid advancement of artificial intelligence (AI) has given rise to sophisticated language models that excel in understanding and generating human-like text. With the capacity to process vast amounts of information, these models effectively tackle problems across diverse domains. In this paper, we present a comparative analysis of prominent AI language models—ChatGPT and Google Bard—focusing on their ability to solve undergraduate finance problems. We find that GPT-4 significantly outperforms Bard-1.0, excelling in easy problems but struggling with complex ones. The results suggest that it is crucial to handle AI with care in order to uphold academic integrity.

Keywords:

• Artificial intelligence
• ChatGPT
• financial education
• higher education
• undergraduate finance

Acknowledgments

The authors would like to thank Shishir Paudel, Shiang Liu, and Taggert Brooks for their help.

Disclosure statement

The authors report there are no competing interests to declare.

Additional information

Funding

This work was supported by grants from the University of Wisconsin-La Crosse College of Business Administration and Menard Family Midwest Initiative for Economic Engagement and Research.